Back to previous post: A savory pie for the first day of winter

Go to Making Light's front page.

Forward to next post: Egregious Self-Promotion

Subscribe (via RSS) to this post's comment thread. (What does this mean? Here's a quick introduction.)

December 2, 2007

The sinople planet
Posted by Avram Grumer at 07:19 PM * 152 comments

Has anyone else here been wondering about the color-blind synesthete?

You know about synesthesia, of course, the neurological condition in which one sensory phenomenon invokes other, separate phenomena. In one common form, letters or numbers invoke the perception of colors.

So there’s this guy, he’s got synesthesia. But he’s also got color-blindness. So some of the colors that get conjured up in his head when he sees certain numbers are colors that he’s never seen in the external world. He calls them “Martian colors”.

I’ve been wondering about the implications of this for the philosophy of consciousness. If you read people arguing about whether consciousness can be explained by physical processes, you run into a lot of thought experiments about color perception, like the Inverted Spectrum, Mary’s Room, and the whole issue of qualia. Because I’m a physicalist, I don’t believe in qualia unentangled from neurology, so I’m interested in this quote from VS Ramachandran, the neuroscientist who wrote about the color-bland synesthete:

The effect is most obvious and pronounced in the colorblind synesthetes, but occurs in “regular” synesthetes as well. The colors evoked by cross activation in the fusiform gyrus “bypass” earlier stages of color processing in the brain, which may confer an unusual tint to the colors evoked. This is important for understanding the phenomenon of synesthesia, because it suggests that the qualia label — that is, the subjective experience of the color sensation — depends not merely on the final stages of processing but on the total pattern of neural activity, including earlier stages.

That passage I’ve emphasized reinforces my belief that there is no perception of color that doesn’t rest on neural activity. As I see it, it refutes the premise of the Mary’s Room argument, by establishing the only way for Mary to have full knowledge of how the color red is subjectively experienced would be for her to also have full knowledge of how that experience is physically generated by neural activity, and that knowledge would itself refute the non-physicalist conclusion that the argument tries to advance.

Comments on The sinople planet:
#1 ::: Rose Fox ::: (view all by) ::: December 02, 2007, 08:09 PM:

I'm immediately reminded of Terry Pratchett's "octarine" (referenced frequently in the Discworld books) and David Lindsay's "jale" and "ulfire" (in A Voyage to Arcturus). I'm sure other spec fic authors have made reference to colors that human beings can't usually perceive, either because they don't exist on earth (dubious) or because we're not equipped to see them (more plausible). Makes me wonder whether some synesthetes see ultraviolet, or infrared, or sound waves, when they look at numbers or letters or shapes.

#2 ::: Greg London ::: (view all by) ::: December 02, 2007, 08:33 PM:

Because I’m a physicalist ... it refutes the premise of the Mary’s Room argument

I'm completely unfamiliar with all of this as a formal study. Reading Mary's Room, though, the first thing that came up was something my akido instructor said.

"You can 'know about' something. And then you can 'know' something."

He was using it in reference to the Aikido textbook that went with the class. You can read it all you want, but if you can't actually do it, can't practice it, can't actually experience it, then there's something that you don't know.

I read the "Mary's Room" thought experiment as saying you can learn everythign you want to learn about Aikido, but you can't actually practice it on another human being. Once Aikido-Mary meets an Aikido instructor in a dojo, can she learn anythign?

And I'd say absolutely.

There is nothing you can do that will prepare you to hold a helicoper in a hover but to practice it until you can do it, until you experience it. And now that I can do it, there is nothing I can do that will have you hold a hover on your first attempt. I can guide you and offer tips and whatnot, but you've got to do it. Statistically, you'll have to spend at least a few hours flying helicopters before you can do it reliably and consistently.

If "Mary's Room" is basically asking whether we can know everything there is to know about something without experiencing the thing itself, my experience says 'no'.

#3 ::: Kayjayoh ::: (view all by) ::: December 02, 2007, 08:33 PM:

A discussion i used to have with friends back in high school when we were feeling philosophical: "What color is your orange?"

By this we mean that, two people looking at an orange both agree that it is orange. Those two people also look at a traffic cone and a safety vest and agree that they are orange. The color wave length that reaches both sets of eyes is named orange. But does orange look the same to both people? Maybe what you see as orange looks like my purple. And what you see as purple looks like my turquoise. The wavelengths of light are given agreed upon names, but maybe the subjective experience of what that looks like are different.

We loved this conversation, because there was no way to prove or disprove it. It went well after ingesting large qualities of sugar and caffiene after a late pep band rehearsal.

#4 ::: B. Durbin ::: (view all by) ::: December 02, 2007, 08:50 PM:

R.A. Lafferty wrote a short story on the subject. The protagonist creates a device that lets his see through the eyes of another. A very telling bit is that he has a crank experience the world as something imperfect and disgusting; the protagonist realizes that living in such a world would make anyone cranky.

Of course, the center of the story is the disgust the protagonist experiences when he sees the world through the eyes of a woman he's attracted to.

I love Lafferty's writing. He doesn't bother with reasonable premises— he starts from sheer absurdity and goes from there.

#5 ::: Vicki ::: (view all by) ::: December 02, 2007, 09:04 PM:

Kayjayoh @ 3: It gets interesting when you find someone who demonstrably has slightly different color perceptions than you do. My partner Andy and I can agree pretty well on the colors of clothing, books, painted walls, and such--things dyed or colored by humans. But we disagree in consistent ways about certain things in the natural world, especially flowers. There are flowers I see as blue that he sees as purple.

More interestingly, there are flowers I see as entirely white that he sees as white with purple patterns. We're moderately confident, because of this, of a hypothesis that he sees slightly further into the ultraviolet than most humans.

(There are also a few things that I see as blue that are gray to him; there's a narrow range of blue that he doesn't see as colored at all.)

So it's not just that I can't ever know that the sensation I describe as "blue" is the same one that Andy or Jo describes as "blue"--it's that I know that even if we had the same subjective sensation, it wouldn't be stimulated by all the same selections of wavelengths.

#6 ::: Doctor Science ::: (view all by) ::: December 02, 2007, 09:09 PM:

I have full (normal) color vision and my husband has a mild form of color blindness. He can see red in bright light, but there is no question in my mind that what he experiences when he says or sees red is *not* what I experience. For me, part of the "red" experience is that the color leaps out; for him, it does not. Is that the sort of thing you mean by "qualia"?

#7 ::: Diatryma ::: (view all by) ::: December 02, 2007, 09:10 PM:

Some time ago, I read that one reason so many kids say blue is their favorite color is that children see more shades of blue than adults-- you see it more strongly, or something along those lines. I would much like to expand my visual spectrum somewhat. People are used to differences in hearing, but not so much this part of sight.

#8 ::: Julie L. ::: (view all by) ::: December 02, 2007, 09:14 PM:

Rose @1: I'm sure other spec fic authors have made reference to colors that human beings can't usually perceive, either because they don't exist on earth (dubious) or because we're not equipped to see them (more plausible).

IIRC there were a bunch of those in the early 20th century, though not always given discrete names-- Lovecraft, of course; possibly also Edgar Rice Burroughs and Robert Chambers-- I suppose because the infrared and ultraviolet parts of the spectrum must've been brought to general public awareness.

Good Omens has "ultra-black" in a marginal note somewhere. WRT "octarine", I regret to report that there are people who have adopted the term in all apparent seriousness to discuss new evolving chakras or auras or something.

There was some vague stuff in the news a year or three ago about people with visual pigments that *can* see a slight way into the ultraviolet spectrum; I can't recall whether the explanation had to do with tetrachromats-- people with fourn visual pigments instead of three, mostly women who are heterozygous for red/green color-blindness.

#9 ::: Shannon ::: (view all by) ::: December 02, 2007, 09:31 PM:

As we've been discussing, how we see the world in color differently from others isn't an issue most of the time. In high school, I had one friend who I knew was color-blind, but it was a fact that didn't seem to affect anything. Except for one day when we sat down to play Uno. I think he agreed to play because normal Uno was fine for him - the bright colors distinguished themselves enough that he could make out differences between them. But we were playing Extreme Uno, which has different cards. The colors are set against a black checked background, making it extremely difficult to him to see which color was which. Unfortunately, he didn't inform us of this until more than halfway through the game. I think our universal thought was, "No wonder he was losing!"

It was a moment I still remember because it reminded me at how differently we perceive the world and how easy it is to forget that.

#10 ::: Lee ::: (view all by) ::: December 02, 2007, 09:38 PM:

Kayjayoh, #3: Interestingly, the first thing I thought on reading that article was that it looked like a definitive answer to the "Is your red the same as my red?" question.

B. Durbin, #4: Do you happen to recall the title of that short story?

#11 ::: Bruce E. Durocher II ::: (view all by) ::: December 02, 2007, 09:57 PM:

Remember, for those of us with color blindness we can see more colors than we normally do by wearing one red contact and keeping both eyes open. (The approximately two working color pigments we use + red-shifted images from the two working pigments--2+2=4. I'm told it makes working out resistor bands a snap.) For those that aren't color-blind who do the same they should see 3 + 3=6. What would the difference be? Beats me--I've never heard of someone with normal color vision desperate enough to have a red contact lens ground for one eye just to check.

#12 ::: Avram ::: (view all by) ::: December 02, 2007, 10:09 PM:

Greg (in #2), I think you're misunderstanding the intent of Mary's Room. It's not a parable about book learning versus practical experience, it's a thought experiment asserting that there's an irreducibly non-physical aspect to mental experiences.

Of course, the real problem with Mary's Room is that its premise assumes something that's obviously impossible, and it then goes on to assert that the "obvious" conclusion is a description of the real world.

#13 ::: Avram ::: (view all by) ::: December 02, 2007, 10:10 PM:

Greg (in #2), I think you're misunderstanding the intent of Mary's Room. It's not a parable about book learning versus practical experience, it's a thought experiment asserting that there's an irreducibly non-physical aspect to mental experiences.

Of course, the real problem with Mary's Room is that its premise assumes something that's obviously impossible, and it then goes on to assert that the "obvious" conclusion is a description of the real world.

#14 ::: C. Wingate ::: (view all by) ::: December 02, 2007, 10:36 PM:

People who have had lenses removed for cataract surgery can see further into the ultraviolet, though I don't know whether the replacement lenses undo that.

That color perception isn't objective has been known for some time. I remember reading Edwin Land's research on the subject in Sci. Am. some decades back.

#15 ::: R. Emrys ::: (view all by) ::: December 02, 2007, 10:48 PM:

B. Durbin @ #4: Thank you. I read that story in a collection in middle school, and have been trying to think of the title and author ever since.

Mary's Room seems (to me) to prove that there are aspects of physical existence that can't be understood without direction perception. But I suppose that's a less exciting conclusion.

#16 ::: Bruce E. Durocher II ::: (view all by) ::: December 02, 2007, 10:50 PM:

Remember, for those of us with color blindness we can see more colors than we normally do by wearing one red contact and keeping both eyes open. (The approximately two working color pigments we use + red-shifted images from the two working pigments--2+2=4. I'm told it makes working out resistor bands a snap.) For those that aren't color-blind who do the same they should see 3 + 3=6. What would the difference be? Beats me--I've never heard of someone with normal color vision desperate enough to have a red contact lens ground for one eye just to check.

#17 ::: Peter Hollo ::: (view all by) ::: December 02, 2007, 10:53 PM:

Avram,
I was going to say something similar in response to Greg @2 - but you beat me to it :) Yes, Frank Jackson in the Mary thought experiment was attempting to prove that physicalism isn't true, but positing that no amount of third-person data will lead to first-person knowledge of something.

One of the problems, as someone else has pointed out, is the usual thought experiment problem - you're asked to take on an absurd premise (in this case, that someone can know all there is to know about colour vision) and then try to draw some sort of common-sense conclusion from a story using that premise.

Searle's Chinese Room comes across the same problem, and indeed the sensible solution is the same one given for the Chinese Room - it's effectively the "Systems Response" to Searle. In terms of the Chinese Room, Searle is tricking us into a category error by pointing to the person in the room as the Chinese-unknower ("But look! He can respond to Chinese questions, but he still doesn't understand Chinese! QED"). The analogous entity to the human brain is the room itself (the whole system); it's the book plus communications slot, with the neural activity carried out at a snail's pace by the bloke in the room.

Mary, with her comprehensive knowledge of human colour perception, is like the bloke in the Chinese Room after he's memorised the entire book. Maybe she's a bit further ahead than him - all he has is a whole lot of incomprehensible synapse gates or something. But even so, what Mary needs to do in order to model colour perception at an experiential level is to instantiate that knowledge as a kind of mental subroutine. The knowledge has to become part of what Mary is.
Whether that's "possible" or not is a practical question, and no more absurd as a premise than asking that Mary memorise all that stuff in the first place. If she could do it, then she could indeed translate a whole lot of data into "knowing how it feels".

Searle's man in the Chinese room is actually slowly ticking over the mental activities of a completely different person - that's another trick that Searle performs, in distracting us from the fact that different people will respond differently to questions in Chinese. So what's being modeled is a complete person, not just some kind of idealised "comprehension-of-Chinese".
Mary is similarly supposed to be able to ingest "apprehension-of-colour" to the extent that when she sees a red rose she just goes "Yup, a red rose, that's what I thought it'd feel like to see one".

There's certainly something absurd going on in these thought experiments, but it's not due to the assumption of physicalism!

And by the way, thanks for the actual post, Avram! It's absolutely fascinating, and does indeed lead to some really interesting kinks in Jackson's Mary-style thought experiments, but it's anything but a win for the dualists/qualiaphiles.

#18 ::: Peter Hollo ::: (view all by) ::: December 02, 2007, 10:55 PM:

("someone else" = Avram. Sorry!)

#19 ::: Julie L. ::: (view all by) ::: December 02, 2007, 11:06 PM:

Vicki @5: There are flowers I see as blue that he sees as purple.

Gosh. Nearly the same thing happens with me and my husband, except with more things than just flowers-- but there's a consistent color range which I see as (or at least call) "blue" while he insists they're "purple". OTOH, this could easily be a difference in mental categorization rather than biological function-- there was that interesting survey/discussion here in ML a while back about the difference between "gray" and "grey".

And then there's the Berlin/Kay theory of the development of basic color terminology in various languages, which is nifty but not without its detractors.

#20 ::: Patrick Nielsen Hayden ::: (view all by) ::: December 02, 2007, 11:24 PM:

Rose Fox, #1: "I'm sure other spec fic authors have made reference to colors that human beings can't usually perceive, either because they don't exist on earth (dubious) or because we're not equipped to see them (more plausible)."

Yes. Hugh Lofting, Dr. Dolittle in the Moon (1928), in which Dr. Dolittle voyages to the moon and discovers new colors entirely outside the red-blue-yellow spectrum familiar to our eyes. (Or red-blue-green, if you want to get all conical about it.) I must have been all of five or six years old when I read this, but the pure adrenal charge of sense-of-wonder it delivered to my brain remains with me in memory to this very day.

(I'm absurdly pleased to discover via a Google search that Richard Dawkins, of all people, remembers this bit of Dr. Dolittle, and discusses it in the Barcodes in the Stars section of his book Unweaving the Rainbow: Science, Delusion and the Appetite for Wonder.)

#21 ::: Adrian Bedford ::: (view all by) ::: December 03, 2007, 12:36 AM:

I found out I was red-green colour-blind at age 10, and it was at once a huge blow (because it meant I couldn't be a pilot or an astronaut), and a revelation, because it explained why I saw things differently from how other people saw things. Like the vividly red grass over there, for example. Or people with green hair. See a lot of green, actually, often in places where people swear there's actually a kind of light brown. And don't get me started about blue and purple. My wife and I run into this a lot. "Ooh, check out the fabulous purple on that car!" "Sweetie, that's blue." "I think you'll find it's purple." "Blue." etc.

Once I tried a German language class. Thought it might be interesting. It was the kind of language class where the teacher only speaks in the language concerned. And one of the early lessons was identifying coloured pencils, quickly, within a time limit, in German. Guess whose brain exploded?

The other problem I've had with colour perception all my life is when people find out I see things differently (usually after I ask someone to pass me that green thing, and they go, "What green thing? Oh, you mean this brown thing?") there immediately follows all the questions, "What colour is X?" where X is every mundane object under the sun, including the sky!

The "how do you know this is red?" question is a curly one I've often had to deal with. I see a lot of red. Sometimes correctly. When asked the question, I have to refer back to situations where someone has indicated that a given object is red (regardless of what it might look like), and this new phenomenon matches the way that previous example looked.

My wife and I play a lot of videogames, too. Frequently these feature a lot of tricky details using red or green as toggled options of some kind. At such moments I have to hand it all over to Michelle to sort out because I have no idea which is which. Doesn't help when the "green" thing often looks like a shade of grey (the green of traffic lights, for example, looks grey).

I also need various assistance playing games like snooker, as well, to distinguish all the differently coloured balls, many of which look either identical, or at least so similar as to be a problem.

#22 ::: Bruce Cohen (SpeakerToManagers) ::: (view all by) ::: December 03, 2007, 02:51 AM:

Avram,

I, too, am a physicalist, mostly because none of the arguments I've seen for dualism are either testable or really intelligible in terms of the world we all live in*. Most of them involve some bizarre thought experiment that actually turns out to be a shellgame or some other sort of intellectual sleight of hand. The various species of zombie and zimbo are such cons**.

Mary's room fails as a useful test case because it's a rigid categorization of the possible outcomes of Mary's entry into a technicolor world into either "getting new knowledge" or "not getting new knowledge". But suppose she learns nothing new, but has a new experience, one she never had before, which gives her new sensations, but not new knowledge? Sensations are not qualia, so they're allowed to be based entirely on physical processes; they can be neural states. But what room does that leave for qualia? What additional phenomena do they explain? None I'm aware of.

Peter Hollo

Thanks for bringing up Searle; I was going to if no one else did. In Searle's case I think it's pretty obvious that he doesn't want to believe in physicalism, and is grasping at whatever straw he can find to deny it. I take from his arguments that he finds the notion of a non-physical consciousness somehow ennobling, and the idea that the human mind and consciousness†† are physical is just degrading to human dignity. Do I need to point out that this is a religious argument, and not in any sense either a scientific or philosophical one?


* Unless you live in a different one. I've been leaning towards the belief that a lot of philosophers come from somewhere else.
** "Imagine a world just like ours, except that no one has conciousness.† Now doesn't that prove that consciousness must be non-physical?" Nope, just proves I can believe six impossible things before breakfast. Waffles, anyone?
† I've never heard anyone remark on that particular puzzle that only people who already believe in souls would even entertain such a question.
†† There are philosophers, and I can't remember if Searle is one of them, who hold that the mind and consciousness are disjoint, so that even if the mind is based on physical processes, the consciousness, which is some sort of ineffable "other", is not. Sorry, guys, I can eff that easily. There's nothing that the concept of a non-physical consciousness adds to the concept of "mind" that would explain anything else, or predict phenomena we can't otherwise explain with fewer mental contortions.

#23 ::: Bruce Cohen (SpeakerToManagers) ::: (view all by) ::: December 03, 2007, 03:01 AM:

C. E. Wingate,

Oddly enough, one of my eyes has always been sensitive just a little deeper into the ultraviolet than the other, so I've always had evidence that physical color perception was not limited to a single mapping of colors. I first noticed this as a child, so it has nothing to do with the fact that I have had one of my corneas replaced because of cataracts. In theory, the replacement could allow me to see a little farther into the UV because it's plastic, but in fact the difference is very small, if any.

And I don't believe that Land's experiments prove that color vision "isn't objective"*, but just that it isn't the straightforward implementation of trichromatic processing that everyone had thought based on 19th century experiments**.

* Unless you mean something very different from what I think you mean by that.
** There was an article in Sci Am in the 1960s that described those experiments (by Maxwell, IIRC), and pointed out that they didn't prove what was thought because none of the filters and photographic emulsions had the spectral sensitivities that the experimenter thought they had.

#24 ::: A.R.Yngve ::: (view all by) ::: December 03, 2007, 04:08 AM:

Fascinating issue.

I'm going to go out on a limb here and ask: Is sex a "qualia" or not?

I think now that before puberty, I didn't understand what sex "was". It was more like a science-fiction concept.

Imagine an alternate version of the "Mary's Room" experiment... where Mary is raised up to adult age in some ultra-Victorian environment where sex is explained in the most clinical way possible, but sensuality is never so much as suggested. Will the actual experience of sex teach Mary something new?

(This probably sounds familiar to anyone who's seen George Lucas' film THX 1138, but a film can't be used as a scientific argument.)
:-S

#25 ::: Peter Erwin ::: (view all by) ::: December 03, 2007, 06:42 AM:

I'll admit I'm not sure what, precisely, the Mary's Room scenario is supposed to be testing. The Wikipedia article Avram links to does say this:

It is important to note that in Jackson's article, physicalism refers to the epistemological doctrine that all knowledge is knowledge of physical facts, and not the metaphysical doctrine that all things are physical things.

What that seems to suggest is that "physicalism" posits an extremely limited definition of "knowledge" as something akin to "book learning." And, indeed, the quotation from Jackson's original article -- Mary as "a brilliant scientist forced ... to investigate the world from a black and white room via a black and white television monitor" -- seems to support this. (But a quick look at the Wikipedia article for physicalism suggests that "physicalism" may mean something different and more complicated, so I'm reluctant to go too far with this.)

Knowledge exists on various different, overlapping levels. There's the abstract, "knowledge of facts" level ("The capital of Nevada is Carson City"; "Color processing in the brain leads to this particular set of neurons being active"); the pre-dictionary[*] knowledge of the meaning of words in your native language; the memory of a particular pain; the sound of a familiar voice; the knowledge of how to run a maze; the habitual "muscular" knowledge of how to throw a ball; etc.

It's perfectly possible to suppose that all these different sorts of knowledge are physically instantiated/encoded in the brain, while being accessed and processed on different levels. Thus, as Greg points out, knowing via book learning all the details of how to fly a helicopter is not the same as learned physical training in flying one; the two sorts of knowledge are probably encoded, stored, and accessed in different forms (and in different parts of the brain).

(I suspect I'm largely agreeing with Bruce Cohen @22, except that I'm happy to include "sensation" in the category of "knowledge.")


[*] In the sense that people knew how to use words in their native language long before lexicographers came on the scene.

#26 ::: C. Wingate ::: (view all by) ::: December 03, 2007, 06:48 AM:

re 21: Back when our offices were in Silver Spring, my window looked out onto the railroad tracks. (I once helped the police catch a bad guy on the tracks from my lofty aerie.) One day another fellow was in the room with me and I pointed out to him the private car on the end of the Capitol Limited.

Him: "You mean the brown one?"

Me: "You're red-green colorblind, aren't you?"

It was Pullman Green, a very deep olive color.

#27 ::: John Stanning ::: (view all by) ::: December 03, 2007, 06:56 AM:

Surely the only way for Mary to have full knowledge of the color red is to experience it, because her mind (that knows all about the physics of color and how the eye perceives it and how the brain processes it, and so on) is not a mind existing somewhere all by itself, but a mind that inhabits a brain that has intimate neurological connections with sensory organs. When Mary first sees red, things happen in her body: receptors in her eyes see red, and transmit new red signals to her brain, and her brain gets a new experience, that of seeing red. Without Mary having that bodily sensation which connects to her brain and her mind, she can't fully know how red is subjectively experienced.

The same goes for hovering a helicopter, or riding a bicycle, or sex. They are all physical activities which cannot be fully understood until your body has done it and your brain and mind have taken in the sensations involved in using and controlling parts of your body which it had never used or controlled in exactly that way before. Those sensations cannot be communicated by book learning because they are different inputs into the brain. And they are not simply new sensations, but new knowledge because the sensations that have been transmitted to the brain have given learning and understanding to the mind.

#28 ::: Heresiarch ::: (view all by) ::: December 03, 2007, 08:06 AM:

Rose Fox @ 1: "I'm sure other spec fic authors have made reference to colors that human beings can't usually perceive, either because they don't exist on earth (dubious) or because we're not equipped to see them (more plausible)."

Gene Wolfe's fugilin, the shade that is blacker than black. There's another that's whiter than white, but I forget what it's called.

Mary's Room:

From the description I just read on Wikipedia, I'm certainly a physicalist. Yet I have no problem at all granting that Mary learns something new when she walks out the door. That the experience of seeing red is qualitatively different than the experience of studying it doesn't seem to me to prove anything at all about qualia.

A child, opening its eyes and seeing a red object for the first time knows what the color red is, even before they know the word for it. This does not teach them anything about how their eyes accomplish this feat of perception. If the experience of red has no relation with knowledge of the neurophysiology of color perception, then why assume the inverse?

#29 ::: Nix ::: (view all by) ::: December 03, 2007, 08:42 AM:

Heresiarch@#28, far be it from me to dispute Wolfe with someone with your nym, but I think it's `fuligin'. (It's also a word whose source is quite obvious, for Wolfe: a backformation from `fuliginous', which is a perfectly good Latinate English word attested from at least 1621 (according to a quick web search, others will have better resources for this sort of thing than I).

#30 ::: Eleanor ::: (view all by) ::: December 03, 2007, 08:47 AM:

Rose Fox @ 1: I think I remember that in one of Enid Blyton's books (The Mountain Of Adventure, I think) the children reach the secret base and discover that the villainous scientists are making something of a colour that they've never seen before. I can't remember what the thing was, but I think it was liquid, a big pool of it, and in my mind's eye I see it as turquoise.

It's practically the only thing I remember about that book, because I found the idea so startling. But I'm not sure if she even thought she was writing SF. Perhaps she thought scientists could actually create brand new colours. As an 8-year-old reader, I believed that.

#31 ::: Sian Hogan ::: (view all by) ::: December 03, 2007, 09:24 AM:

And now I really REALLY wish there was a way to cure colour-blindness, even if only temporarily. Because I want to know whether the above-mentioned colour-blind synesthete is really "seeing" the colours he normally misses from the spectrum that other people perceive. In which case, is the range of possible colours hard-wired into the brain, even without being experienced? Which would seem odd, given that it doesn't seem as though blind-from-birth people are able to adequately envision colour. Or is it that, once certain colours HAVE been experienced (those in this guy's visual spectrum), the fuller range of colours can be deduced?

Or is this something else all together- is he "seeing" colours which, for most people, don't even exist? Is it only this staid practice of using our eyes for seeing that is preventing us from accessing the much wider spectrum with which our brain is able to cope?

All very interesting, leaving aside Mary's Room. (Which I agree seems flawed as a thought-experiment.)

#32 ::: Jamie ::: (view all by) ::: December 03, 2007, 09:27 AM:

Lemon Lime Gatorade:

Green or Yellow?

You'll find that females generally identify it as yellow, and males as green.

I'm told it has to do with the ratio of rods and cones being different for men and women.

I'm not sure how it fits in, but hey, it feels like it should be relevant. Aproposness? Relevantiness?

#33 ::: Greg London ::: (view all by) ::: December 03, 2007, 09:28 AM:

hm. Well, perhaps reading about Mary's Room in an attempt to understand physicalism isn't the best way. So I went and looked up physicalism in wikipedia. It didn't help.

It seems to be saying that there is no difference between "mind" and "brain", or that anything "mind" can be explained solely as "brain". or something.

Either I'm not understanding the point, or it's parsing as absurdity. I'm certain my experience of emotions could be mapped into a collection of synaptic firings (given another millenium of technical advances or so), but that doesn't explain anything about my experience of whatever emotion I'm feeling.

It seems to boil down to the "mind/body" problem, and personally, I never understood what the big fuss was about whether it's mind/body, or whether it's all mind, or whether it's all body. To borrow from Luther, each individual has their own personal relationship with mind. And I don't need some pope, or some physicist, or some philosopher, telling me how I should relate to God or how I should relate to my mind.

Am I missing something?

Can someone explain physicalism in such a way that it doesn't sound like a turf war with the religious types over the ownership of my mind?

#34 ::: Remus Shepherd ::: (view all by) ::: December 03, 2007, 09:34 AM:

I'm red-green colorblind, so every time I've used red or green in a story I've made references to colors I've never seen.

I don't get the argument against qualia at all. Psychologists have proven that some colors evoke certain emotions, and these emotions are subjective. Even assuming you could predict these responses down to the firing of every synapse, that's a long way from actually experiencing it yourself.

#35 ::: Carrie S. ::: (view all by) ::: December 03, 2007, 09:34 AM:

My dad's red-green colorblind, and he says that he can distinguish the colors, but that the difference doesn't leap out at him. I am forced to accept this as truthful, since he has absolutely no reason to lie to me about it, but it makes no sense to me--how can it be hard to tell the difference between red and green if you can tell the difference at all? I can picture just not being able to distinguish; in my imagination red and green both come out as a sort of muddy color...

When I was little Dad used to take me clothes shopping to ensure he got the color he wanted. Once we walked out of a shop on the verge of paying for something when the salesman who'd been helping us went into the back room for something and I said, "Dad, I thought you wanted a brown suit." The salesman had been told brown, but had gone through all the fitting with a grey suit which he was about to sell. The most charitable explanation is that he was colorblind too, but if so he shouldn't have been working in clothing sales.

Meanwhile, Liam has the same shifted-into-UV thing that Vicki's partner has. Somewhere around blue-green our perceptions start diverging. However he seems to be able to see a little further into IR as well, which suggests he just has a broader range of visible light than I do. And we have a nomenclature disagreement about a particular shade, which is a very red orange to me and which he insists on referring to as "process red" even while admitting that it's orange. He says process red is the correct technical name; I say that the technical name and the name to be used when talking to one's partner who's never done any printing are two different things. :)

#36 ::: steve ::: (view all by) ::: December 03, 2007, 09:46 AM:

Bruce@16: you don't need an expensive red contact lens, any more than Von Daniken(?)'s cavemen would have needed access to an x-ray source to illustrate skeletons. Red cellophane or a piece of red lighting gel should do the trick.

(I'm not sure that doing this would add to the colour experience of people who were already trichromats, though)

#37 ::: Ian ::: (view all by) ::: December 03, 2007, 10:11 AM:

"...wrote about the color-bland [sic] synesthete"

Is that someone who smells everything in shades of beige?


#38 ::: Faren Miller ::: (view all by) ::: December 03, 2007, 10:27 AM:

I have mild synaesthesia (mainly for numbers), but what interests me more is emotional responses to color. Some of this may be due to compatibility -- I have blue-greenish eyes, so I like blue-green shades, and now that my hair is graying I also like paler sage green -- but it can *feel* like it goes deeper.

My computer allows a color scheme set up, and I've got it in tones of "not as crude as turquoise" shading into something more tawny at the top bar, with utilitarian tan for the tool bar, and it pleases me greatly. I recently bought some coasters in similar shades (with varying prints) at Cost Plus, and I still feel delight every time I look at them. Is this a "woman" thing, or do some of the guys here have it too?

#39 ::: Monica Toth ::: (view all by) ::: December 03, 2007, 10:28 AM:

Greg London@33,

As I see it, the war between physicalism and dualism is rooted in religion, but it goes further than a turf war over mind ownership. If your mind is entirely made of atoms (so the theory goes), then there's nothing to salvage and send to heaven when you die. Therefore there must be an immaterial component of your consciousness that is preserved after the rest of your body decays.

And then there's the subject of free will vs. determinism, which I enjoy as a dilettante but am thoroughly unqualified to discuss.

#40 ::: individualfrog ::: (view all by) ::: December 03, 2007, 10:42 AM:

A lot of how we think of colors, I think, is probably based on the language we're using...Japanese people say the traffic light is "blue" when it's not red or yellow. Nowadays many people certainly will say "Well, it's green, but we call it blue" but it seems that in the past, blue and green were sort of mixed, or perhaps the dividing line between green and blue was different from where it is to Europeans. I certainly think it's possible for there to be more or fewer divisions between colors than we usually make, the number seems pretty arbitrary, especially considering how "purple" and "violet" overlap to most people.

I remember reading that some people...criticized? were disappointed in? Helen Keller's writing because she talked about sights and sounds, and they wanted to learn what it was like to live without them. But I think she said something like her whole consciousness was based on books she'd read, and every book, of course, describes sights and sounds. Her world wasn't made of physical sensations, it was made of words. (But I don't want to put words into Helen Keller's mouth, I may be misremembering completely.)

#41 ::: Zack Weinberg ::: (view all by) ::: December 03, 2007, 11:03 AM:

#40: Lots of languages don't have separate "basic color terms" for blue and green. This does not mean that people who speak one of those languages don't understand the difference; it just means that they have to say something longer if the difference matters. "grue, like new grass" versus "grue, like deep still water" for instance. (Yes, the word "grue" is used in the literature on this topic, and I still haven't gotten to the point where I don't think of Zork.)

There is a huge body of semi-controversial research on which languages have which basic color terms and why. I can't really do better than point you to Paul Kay's website: http://www.icsi.berkeley.edu/~kay/

#42 ::: C. Wingate ::: (view all by) ::: December 03, 2007, 11:16 AM:

Part of the problem I see in the Mary's Room situation is that it begs, very painfully, of wnat exactly is "knowledge" in the first place. I'm leaning extremely heavily on the "akido" theory.

Something that is bugging me here: If we succeed in making a totally computational (that is, AI-based) model of a particular human mind, doesn't that in fact give us a kind of dualism? If one can describe and model human thought without going to the trouble of modelling the biology of neurons, then it implies that if we say that the two minds (biological and computational) are the same (in the sense of one being an accurate copy of the other), the mind exists in some way other than the mere physicality of how it is represented.

Also, allow me to throw another thought experiment into the mix. Suppose one were to intercept the visual processing of a synesthete and feed it into the same stage of a normal person's visual processing. Will they see the non-existent colors? How about if we hook up our colorblind synesthete instead? It seems reasonable to anticipate that the normally sighted person would identify at least some of the "martian colors" as normal colors. If we reverse it, the color-blind synesthete might be able to give some of his "martian colors" names if his visual paths were stimulated with signals from normal eyes.

#43 ::: Carrie S. ::: (view all by) ::: December 03, 2007, 11:42 AM:

Lots of languages don't have separate "basic color terms" for blue and green.

So, do you guys know about the color term progression? The theory basically says that there's an order in which languages add color terms. It's not quite as absolute as the original theorist made it out to be, but it goes like this:

If a language has only two color terms, they'll be "white/light" and "black/dark". If it has three, the third one will be Red. If it has four, the fourth is either Green or Yellow, and five adds the other of Green or Yellow. Then Blue, then Brown, then Pink, Purple, Orange and Grey in any order until it has all 11.

There are languages that break this rule in one way or another (Russian has "light blue" and "dark blue", Hungarian has two terms for Red, e.g.), but it's a handy rule of thumb for conlangers. :)

#44 ::: John Stanning ::: (view all by) ::: December 03, 2007, 11:43 AM:

I think your AI-based model would be "RoboMary" as proposed by Daniel Dennett (see the Wikipedia article on qualia, linked by Avram above) to try to get round the Mary's Room argument.

If you could make an AI-based model that was functionally equivalent to a human brain, then it would exist in the same way as this blog exists, as an entity that has individuality, yet is dependent on a physical host, and is copiable (clonable) from one host to another. If you erase the physical host (body, server) without making a copy, then it's dead.

#46 ::: mjfgates ::: (view all by) ::: December 03, 2007, 12:22 PM:

There's a German camouflage color from WWII called dunkelgelb, "dark yellow." Americans invariably describe it as "tan," Englishmen as "grey" and Frenchmen call it "green." Germans insist that it's dark yellow, dammit, a whole separate color. A lot of camo colors are confusing this way, but I don't know of any others that are so, well, CULTURAL.

#47 ::: Diatryma ::: (view all by) ::: December 03, 2007, 12:32 PM:

I have maintained for years that indigo is not a real color. Show me something that is indigo and not dark blue. It's not in any of the rainbows my window prisms make (which is not a great criterion; more than one of them filters out green, but that leaves a space) and it doesn't make *sense* in the spectrum. Roy G Biv and I must disagree.

#48 ::: C. Wingate ::: (view all by) ::: December 03, 2007, 02:09 PM:

re 46: I'ts a bit more complicated than that. If you go prowling around websites you will find that non-military use of "dunkelgelb" means a very saturated, dark yellow-- what Grumbacher's sells in a tube as "Cadmium Yellow Deep". The WW II military usage is quite a mess; they seemed to have picked a color name out of the air. There are at least two different colors apparently intended by it. One is RAL 7028, which as the number shows is definitely well into the grey range. The other one is referred to as "Dunkelgel nach Muster" and appears to be pretty variable. It seems to be ochre-based, but like American/British khaki it often slides into a greenish cast. So it's definitely possible that a large component of the identification of it as different is because people are looking at differently colored objects.

#49 ::: Andrhia ::: (view all by) ::: December 03, 2007, 02:20 PM:

Carrie S. @35: Do you not have trouble distinguishing, say, a deep navy blue from black? Or light peach from light pink? I'd think that there's-a-difference-but-it's-hard-to-tell experience might be similar to those, which I had thought were common experiences.

Or maybe my color vision isn't as perfect as I had thought!

#50 ::: Julie L. ::: (view all by) ::: December 03, 2007, 02:20 PM:

Faren @38, wrt teals/turquoises: I recently bought some coasters in similar shades (with varying prints) at Cost Plus, and I still feel delight every time I look at them.

There's a certain range of blue-violets and violet-blues that somehow instinctively connects with something in me that produces a primal sense of "OoooOOOOOooooo". So yeah, it's not just you.

I'm tangentially reminded of the standard acceptable palette of men's clothing-- oh sure, there are nonce fashions for avocado green or pumpkin oranges etc., but almost all of the "classic" businesswear that can be worn for decades is in a limited range of navy blue, some pale blues, hunter green, burgundy, a lot of greys and browns, and black. I've heard rationalizations for these that vary from the seniority of those colors as natural wool dyes to the ethnocentric promulgation of colors which WASPs look good in; beats me.

individualfrog @40: Japanese people say the traffic light is "blue" when it's not red or yellow. Nowadays many people certainly will say "Well, it's green, but we call it blue" but it seems that in the past, blue and green were sort of mixed, or perhaps the dividing line between green and blue was different from where it is to Europeans.

That's to do with the aforementioned "grue" mixed green/blue category, which is really what the Berlin/Kay model's first invocation of "green" should be-- in Japanese (and Chinese), the old "grue" word is 青 ; when the new category 緑 split off as "green", it mainly left behind "blue" by process of elimination, but there's still some ambiguity. (IIRC there's a vaguely similar situation in Celtic languages about the color words gorm and glas, allowing for some inevitable spelling variations.)

If you go to this Chinese dictionary site (among other nifty features, it includes etymological info for many of the written characters) and enter "green" as an English search term, the latter character appears first, with sample phrases involving reasonably greenish stuff like peas, trees, and tea, but the former one still also indicates greenish stuff such as leaf vegetables and frogs, as well as the sky.

#51 ::: NelC ::: (view all by) ::: December 03, 2007, 02:21 PM:

I was going to mention Japanese noun ao, but I've been beaten to it. I'll note, however, that aoi, the adjective form, can also mean "pale".

I'll also mention that many birds have four types of cone in their retinae, being able to see into the ultraviolet, though it's not known how many separate colours they can perceive.

#52 ::: Andrhia ::: (view all by) ::: December 03, 2007, 02:23 PM:

C. Wingate @48: You're right, there's no separate indigo in a spectrum. It's my understanding that Sir Isaac Newton made it up because that made seven colors, not six, and seven was a more appropriately mystical number for his taste.

...sorry for the double post.

#53 ::: Carrie S. ::: (view all by) ::: December 03, 2007, 02:28 PM:

Do you not have trouble distinguishing, say, a deep navy blue from black? Or light peach from light pink? I'd think that there's-a-difference-but-it's-hard-to-tell experience might be similar to those, which I had thought were common experiences.

Well, sure, but I have a hard time imagining how red and green can be sufficiently close together to be hard to distinguish in that fashion, if they can be distinguished at all. This is because I am not colorblind, and it's surely a failure of my imagination. I'd think it was that the muddy color has a hint of red or a hint of green to it, but from Dad's description the colors are perfectly clear; he just can't tell them apart well.

#54 ::: elise ::: (view all by) ::: December 03, 2007, 02:41 PM:

C Wingate #42: Suppose one were to intercept the visual processing of a synesthete and feed it into the same stage of a normal person's visual processing. Will they see the non-existent colors?

I think you'd have to intercept something other than the visual processing of the synesthete (I keep wanting to write that "synaesthete," but maybe that's just me), because the colors the synesthete perceived happen as a result not of looking at colors, but of perceiving some other-than-color thing. (Colors as a byproduct of numbers, for example. Like tastes as the result of sounds, or why certain telephone rings taste like pickles. Yeah, I've got that one.) What you'd want to intercept is whatever strange neural firing in the synesthete's visual processing area(s) is happening, and route that to the non-synesthete's visual processing. (If you did it with the taste thing, and used my telephone ring experience, you'd just give somebody the taste of pickles, though, which they might well already have, at least if they'd had that kind of pickle already.)

Drat. Wait, no, I think I misunderstood what you said. Maybe that's what you meant: taking the synesthesia-generated stuff (if it in fact manifests as visual processing in the brain of the synesthete, and not in some other way) and giving it to the non-synesthete, while remembering that it is (for the synesthete) the product of something other than looking at colors.

Oh, bother. I keep getting Centipede's Dilemma with all this. But still....

How about if we hook up our colorblind synesthete instead? It seems reasonable to anticipate that the normally sighted person would identify at least some of the "martian colors" as normal colors. If we reverse it, the color-blind synesthete might be able to give some of his "martian colors" names if his visual paths were stimulated with signals from normal eyes.

Is it the eyes of your synesthete that are not normal? Or the visual processing (presumably in the brain)? Or something else? Also, synesthesia doesn't color inside the lines. if you stimulate my visual paths with signals from normal eyes, you'll probably get temperature/density/pressure sensations, which is not quite what you're aiming for. Then again, my synesthesia (if that's what it is) doesn't seem to map real well onto descriptions of it that I've seen. (It's darned useful in my work, though.)

Fascinating topic here. And I really want a sample of dunkelgelb some day.


P.S. Just now I think qualia, if they exist, don't exist in things; they exist in moments of juxtaposition. Relationship has a qualia, and that's why experience has a flavor.

P.P.S. Diatryma, I can show you the difference between indigo and dark blue right away. Indigo is the one that hums.

#55 ::: Andrhia ::: (view all by) ::: December 03, 2007, 02:52 PM:

Carrie S. @53: I see what you mean. :) I've always thought it must be like seeing via a CRT monitor with one of the colors gone. A dying metaphor, that.

And: 52 for me should have been @Diatryma in comment 47.

I'll just go back to playing by myself in the corner, now.

#56 ::: Gabriele Campbell ::: (view all by) ::: December 03, 2007, 03:07 PM:

I have full fledged snyaesthesia of the letters/numbers are in colour version. And not only for the Latin letters we use, but it works for Greek, Russian, Runes and Hebrew as well, which helps me a lot with languages. But the colours, and the aura of words shaped by the colours represented, are of so subtle shades that it's not easy to find words to describe some of them. What I hate is the question, 'what does my name look like to you.' ;)

Maybe some Martian words would help. :)

#57 ::: Lee ::: (view all by) ::: December 03, 2007, 03:11 PM:

mjfgates, #46: My partner used to work for Volkswagen, and he says they used that color on cars up until the early 70s. The comparisons he offers to describe the color are "just a little lighter than the outside of an overdone hard-boiled egg yolk," or (more succinctly) "baby-shit yellow".

#58 ::: Avram ::: (view all by) ::: December 03, 2007, 03:31 PM:

Heresiarch #28: I'm pretty sure that fuligin isn't a new color, but just evidence tha the people of Severian's culture separate what we call black into more then one color, possibly because they live in a more dimly-lit world than we do. Fuligin is used for things that are absolutely black, reflecting no detectible light at all, while black would presumably still be used for ordinarily black object that reflect enough light to have discernible surface contours.

Greg London #33 -- Can someone explain physicalism in such a way that it doesn't sound like a turf war with the religious types over the ownership of my mind? That may be like asking someone to explain the Cold War so that it doesn't sound like a turf war between the USA and the USSR.

Diatryma #47, you're right about indigo. Newton added it to the spectrum to get seven color for numerological reasons.

#59 ::: P J Evans ::: (view all by) ::: December 03, 2007, 03:37 PM:

Lee @ 57

Ah, close to the color my father referred to as 'muckledy-dun'. (Our VW was that sort of pale brownish-pearl they used in the early sixties - call it VW off-white.)

#60 ::: C. Wingate ::: (view all by) ::: December 03, 2007, 03:41 PM:

One notes from Wikipedia that there are a variety of different types of color blindness. In the most common sort, acto them, the green receptor is still functioning but is shifted strongly towards the red. If you transplanted one of these eyeballs into someone with normal vision, then presumably they would see red objects shift towards yellow and blue-green objects get darker and shift towards blue.

#61 ::: Julie L. ::: (view all by) ::: December 03, 2007, 03:44 PM:

I'm perfectly willing to believe that Newton arbitrarily labelled part of the normal spectrum as "indigo". But like Elise, I also have my own private indigo (as it were)-- a certain blue-violet range that goes *ping!*. Cobalt glass is toward the blue edge of "indigo"; iolite and tanzanite are also in the zone.

Perhaps within red/green color-blindness, the separation of red and green requires a similar process as a normal person trying to determine whether a nominally neutral color (e.g. brown or grey) is "warm" or "cool" wrt red/yellow or blue/green undertones?

#62 ::: Gabriele Campbell ::: (view all by) ::: December 03, 2007, 03:46 PM:

We had one of those, too.

#63 ::: kid bitzer ::: (view all by) ::: December 03, 2007, 03:52 PM:

did anyone else have trouble parsing this portion of avram's post?

"it refutes the premise of the Mary’s Room argument, by establishing the only way for Mary to have full knowledge of the color red is subjectively experienced would be for her to also have full knowledge of how that experience is physically generated by neural activity"

maybe "...of the color red *as* subjectively experienced"?

or "that the color red is subjectively experienced"?

some other typo? or is it perfectly grammatical, and my parser is missing an obvious construal?

#64 ::: Zack Weinberg ::: (view all by) ::: December 03, 2007, 04:18 PM:

#54 et preq.: I know a bunch about this because someone in my lab is doing research on how it works.

Color-number synaesthesia is thought to arise in a brain region pretty deep in the visual processing graph, the fusiform gyrus. (Most well known for recognizing faces and other shapes.) The best guess is that there is more crosswiring between there and the color areas than is usual. I don't think swapping such high-level brain areas between individuals (assuming an appropriate magic wand) would do anything interesting. The details of interconnection are a function of experience, not genes, and are wildly different between individuals. It seems to me that in both swappees, they would be so badly miswired as to be useless.

Some hallucinogens are reported to produce synaesthetic effects in "normal" perceivers -- this is presumably because they crank up the sensitivity on the crosswiring that does exist in those people. (You have to have at least some, for understanding objects with typical colors.)

#65 ::: Gabriele Campbell ::: (view all by) ::: December 03, 2007, 05:00 PM:

I have a very bad memory for faces, so maybe the synapses are too busy with that letter/colour thing. :(

#66 ::: Today Wendy ::: (view all by) ::: December 03, 2007, 05:11 PM:

On the topic of colour-blindness, one experience with my father (who is red-green colour-blind) really made things clear to me. He came home one day with a new suit. It was a very tweedy brown plaid fabric, with a hot pink thread running through the pattern. We all expressed great surprise at his selection and it became clear that he hadn't noticed the pink thread (but was able to distinguish it once it was pointed out to him) in the same way that you might imagine not noticing a forest-green thread against a variegated brown background. The red (or pink in this case) just didn't pop out.

#67 ::: Diatryma ::: (view all by) ::: December 03, 2007, 05:17 PM:

Part of me wishes I had some degree of synaesthesia. Sadly, no. I have interpreted some things... idiosyncratically, I guess; Dad never believed me when I said I could hear mice moving when I held them, and even I was never sure if I was making it up. Even a bit of UV or IR, or a broader spectrum of hearing or any other sense, would be interesting. I know it'd also be a pain and one more thing to deal with (I can't imagine what my childhood pickiness would be like were I a supertaster) but whenever I read about something like this, I feel like the world is a blog, and some of the posts are password-protected so I can see that they're there, but not read them. Information is available, but I cannot have it.

#68 ::: P J Evans ::: (view all by) ::: December 03, 2007, 05:18 PM:

I've seen red plastic sheets being sold for quilters, so they can tell if two fabrics are in the same range of values (light/dark). At least, that's my (possibly incorrect) understanding of the theory behind it.

#69 ::: C. Wingate ::: (view all by) ::: December 03, 2007, 05:29 PM:

Interesting site I found while looking for some stuff on color vision: a extensive discussion on color vision as it pertains to painting (in this case, water colors). Curious factoid along the way: the modern notion of primary colors seems to be a recovery of ancient philosophical notions. In between, primary colors were those that unmixed paints came in.

#70 ::: Lila ::: (view all by) ::: December 03, 2007, 06:43 PM:

I actually see colors slightly differently from my left eye vs. my right eye. (And no, my eyes aren't two different colors.) Unlike Bruce, however, I have no special UV-vision.

My husband is one of those partially-red/green-blind types. To him, the flowers and the leaves of flame azalea look the same color, unless he gets very close to them in strong light. (Also, avocado green and the color of Thousand Island dressing appear the same to him.)

#71 ::: C. Wingate ::: (view all by) ::: December 03, 2007, 06:52 PM:

re 45: I haven't read the whole thing yet but there's a huge gaping hole in the first case given (the blue banana). Knowing how your brain is supposed to react to color isn't the same as knowing how it is reacting now. Indeed, in terms of perception, that perception itself is the latter knowledge. Therefore Tricked Mary cannot tell that she is having an inappropriate reaction to the yellowness of the banana on her own, without something else (in her case, a mechanism for analyzing the reflected light from the blue banana, or alternately a device for analyzing what's going on in her brain in the terms of her existing understanding) to tell her what color the banana is. Without such a device or some other color standard to compare the banana to, I would suggest that she would see a "yellow" banana, but misunderstand "yellow" to signify what really should be labeled "blue". So she would learn what blue is, but tag that knowledge with the wrong name.

#72 ::: Tim Walters ::: (view all by) ::: December 03, 2007, 07:27 PM:

C. Wingate @ 71: By hypothesis, she knows every reaction that will occur in her brain as a result of seeing the color blue, and this includes her conscious reaction. Therefore, she can differentiate between seeing yellow and seeing blue without an external device, since they cause different conscious reactions.

#74 ::: C. Wingate ::: (view all by) ::: December 03, 2007, 09:29 PM:

re 72: Two objections:

First, knowing what her conscious reaction would be is outside of Jackson's original version of the case.

Second, such knowledge simply begs the question. Of what would such knowledge consist? For real-world humans, knowledge of sensations consists of memories of those sensations. For example, the experience of touching wall current isn't natural to humans, yet is quite characteristic. Description of the sensation, or of how it comes about, is inferior to the experience itself. If a form of knowledge is to be postulated which consists of neither memory nor description, I for one am going to insist on a description of said knowledge that gives me some confidence that it even could exist.

As it stands, Jackson's original version with Dennett's modification is, I think, quite susceptible to my objection. The experience of analyzing something is not the same as experience of that thing. But I also object to Dennett's disquisition on how the scope of "understanding everything" is underestimated. Overestimation is also a problem, because one can easily posit enough understanding to require extra modes of thought and perception, thus requiring Mary to be superhuman. A superhuman Mary is not a suitable model for human perception; the understanding that Mary has must be kept within the bounds of what humans in general are able to understand for the thought experiment to have any validity.

#75 ::: Stefan Jones ::: (view all by) ::: December 03, 2007, 09:42 PM:

#73: Damn mutants! I bet they use that fourth color to send secret messages to each other plotting the downfall of us inferior normals.

#76 ::: Tim Walters ::: (view all by) ::: December 03, 2007, 10:19 PM:

First, knowing what her conscious reaction would be is outside of Jackson's original version of the case.

"She discovers, for example, just which wavelength combinations from the sky stimulate the retina, and exactly how this produces via the central nervous system the contraction of the vocal chords and expulsion of air from the lungs that results in the uttering of the sentence ‘The sky is blue’."

I would say that if you have this knowledge, you know how to associate the stimulus with a given response. In other words, Mary could lay a "trap" for the stimulus of blue light, causing it to trigger an association. At least, I don't see how we can assume that she can't do this.

A superhuman Mary is not a suitable model for human perception; the understanding that Mary has must be kept within the bounds of what humans in general are able to understand for the thought experiment to have any validity.

I don't think it has any validity anyway. To be an argument against physicalism, Mary has to be in exactly the same physical state before and after she sees a color for the first time; but to claim that, in that state, she couldn't recognize the color is simply to assume the consequent.

It's certainly true that a real human won't learn what red looks like without having seen it, but I don't see how that has any consequences for physicalism.

#77 ::: B. Durbin ::: (view all by) ::: December 03, 2007, 10:30 PM:

I work under the theory that the Japanese had a combined term for blue/green for a while because of the ocean. But it's only a theory.

For those people who were wondering about the Lafferty story, sorry it took me so long to answer. It's "Through Other Eyes" and I have it in the collection Nine Hundred Grandmothers, ISBN 0-441-58051-3 (undoubtedly out of print; this edition is from 1982.) It's actually the only Lafferty I have, though I do keep my eyes open in the used book stores.

I am definitely not synasthetic. But I had a friend who was and insisted that different notes had different colors, which is a pretty cool trait for a musician.

Incidentally, there's a shirt I own that I refer to as green while my husband insists that it's blue. It's not too far off the color of denim, actually, so I see his point, but it's just teal enough to make me think of it as green.

#78 ::: Anne Sheller ::: (view all by) ::: December 03, 2007, 10:48 PM:

Lee @ 57 - my parent's first VW microbus, Clarissa, was that greeny-yellow color. By the end of its existence, it was also liberally spattered with stick-on plastic flower cut-outs, mostly covering rust spots. I learned to drive in that underpowered, hell-in-a-crosswind, stick shift vehicle.

#79 ::: NelC ::: (view all by) ::: December 03, 2007, 10:53 PM:

I recall a memory of reading somewhere that the lens of the eye tends to colour with age and exposure to ultraviolet; yellow, I think. So the colours one sees in age aren't the same colours, quite, that one saw in youth.

I also recall reading elsewhere that this is supposed to be why some languages concatenate colours: in countries blessed with strong sunshine, older people are supposed to lose the ability to discriminate between blue and green. I can't say I'm convinced, but I thought I'd throw it out here.

#80 ::: Avram ::: (view all by) ::: December 03, 2007, 11:14 PM:

Kid bitzer #63, it should have (and now does) read "full knowledge of how the color red is subjectively experienced".

#81 ::: Marilee ::: (view all by) ::: December 03, 2007, 11:25 PM:

Winchell Chung, #73, I know a tetrachromat. She has twin daughters (fraternal) and one of them is also a tetrachromat.

#82 ::: Greg London ::: (view all by) ::: December 04, 2007, 12:08 AM:

Me@previously: Can someone explain physicalism in such a way that it doesn't sound like a turf war with the religious types over the
ownership of my mind?

avram@58: That may be like asking someone to explain the Cold War so that it doesn't sound like a turf war between the USA and the USSR.

I so don't get this. If a physicalist says that the "mind" doesn't exist, then why is he fighting a turf war over it?

Doesn't this boil down almost in exact parallel to the misconstrued notion that there is one question and the answer is either "God" or "Science"? Because I'm fairly certain that it's more accurate to say that "god" and "science" ask different questions.

#83 ::: Heresiarch ::: (view all by) ::: December 04, 2007, 01:50 AM:

Nix @ 29: "Heresiarch@#28, far be it from me to dispute Wolfe with someone with your nym, but I think it's `fuligin'."

How dare you disrepsect and quetsion my authotiry? This bevahior will not be torelated!

(I'm sure you're right; I wasn't totally sure and almost googled it but then I got lazy.)

There was a story a couple years back about a new fabric the military invented that absorbed something like 99% of all visible light. The picture of it looked like someone had photoshopped a square of black out of the middle. My first thought was, fuligin: check. On course to Urth; arrival in approximately a gazillion years.

Avram @ 58: "I'm pretty sure that fuligin isn't a new color, but just evidence tha the people of Severian's culture separate what we call black into more then one color, possibly because they live in a more dimly-lit world than we do."

Hm. That's an interesting theory. Where does that leave the whiter-than-white color?

#84 ::: Bruce Cohen (SpeakerToManagers) ::: (view all by) ::: December 04, 2007, 02:23 AM:

NelC @ 79

From personal experience, the lens of the eye does in fact get cloudy and discolors with age. You're right, it yellows, and so the colors shift, and get a little desaturated too. This is why everyone I know of, including myself, who has had one lens replaced because of cataracts, immediately starts pestering the surgeon to replace the other. It's like having the age of your eye rolled back to when you were in your twenties. In my case, the doctor said he could probably justify the second surgery to the medical insurance company after about six months, which means shortly after the beginning of the year. I'm anticipating it gleefully.

#85 ::: LLA ::: (view all by) ::: December 04, 2007, 06:04 AM:

My father is red-green color blind, but he likes to see me wearing red very much, just as he likes to see me wearing green. To his eyes, they both suit my complexion (very anglo-saxon; the architypical definition of a white person) more than shades of brown (which he mostly can see), and I and others agree.

He loves both of the popular red roses named "Mr. Lincoln" and "Chrystler Imperial", although he declares that he sees the deep red of their blooms and the deep greens of their foliage (exact opposites on the color wheel as I see them) as the same color. He is emphatic in his declarations that he sees neither red nor green as shades of grey -- he simply says that he sees opposites on what I see of the color wheel as the same shade.

Interestingly, while according to Mendelian genetics (although not according to more recent understandings), my uncle (my father's brother), who is also color blind, has a different degree of disability. I have listened to many discussions in which my father and uncle have conversed about such-and-such a traffic intersection where the shades of green, yellow, and red created difficulties for my uncle to discern driving clues (he uses position alone to determine when to go/slow down/stop -- a dangerous practice at particularly complex intersections and an even greater problem in rural areas where only red and green lights are given at intersections), whereas my father sees a different range of shades and has an easier time seeing the distinction between orange-yellows and reds, and therefore also has an easier time seeing the difference between orange-yellows and green.

As the only daughter between them, I am glad that my childlessness means that the trait (both according to Mendel and to modern understandings of genetics I am unquestionably a carrier) will die with me. My father has deeply regretted his visual handicap, as it not only limited his view of the world but has also prevented him from a number of professions he might have enjoyed (my maternal uncle, the chemist, relies on his color vision professionally; my father cannot distinguish even between blood and pus and is legally prohibited from flying a plane, to name just a few of the problems he faces; although he would never begin to compare his difficulties to other genetic diseases, color-blindness has deeply shaped his vision for his life).

The odd thing, given this genetic background, is that I consider myself to have an unusually heightened sense of color. I am a devoted crafter, so the ability to distinguish different shades and values is important to my appreciation of appropriate color combinations when I sew/knit/crochet/quilt/photograph/.... I did not inherit this ability from my mother; although I admire her color sense, she has no ability to make a match between thread and fabric unless she directly compares the two. I call my ability "color memory" because I am able to make such matches (for example, choosing between about 600 different values, tones, and shades among the several brands at my local store) for a piece of fabric I bought years ago.

I have often wondered if the X-chromosome I inherited from my father somehow gave me this ability, but I suppose I will never know if his color memory would have been equally powerful, had his Y-chromosome not defeated him, or if some other miscellaneous genetic bit gave me the advantage over my relatives. I suspect that this question is as unanswerable as whether or not the red/green my father sees matches either color as I remember it.

#86 ::: NelC ::: (view all by) ::: December 04, 2007, 06:25 AM:

Bruce @84: I was wondering if one actually notices any difference as one gets older, since the visual system ought to compensate as it does under differing light conditions, but a lack of saturation does sound likely. It might explain the occasional wistful remark one hears that the sky isn't the same blue as when one was young.

Now, here's an odd thing I've noticed. In the HSB colour system, the colours of the spectrum are arranged in a wheel, with the primary colours red, green and blue at 0°, 120°, and 240° respectively. The secondary colours yellow, cyan and magenta are at 60°, 180°, and 300°. I make (less than) a living as a designer, and whenever I have to choose a blue for something, I find myself gravitating towards 210°, midway between blue and cyan according to HSB, since that looks to me more like the platonic ideal of blue than the fully saturated blue at 240°; that position on the wheel looks a little purple to me. I'm not sure what's going on; possibly the RGB colourspace of CRT and LCD monitors don't quite match the colourspace of my visual system, or perhaps my lenses are yellowing.

Oh, and fuligin? Phase conjugate mirrors. These have the interesting property of reflecting light directly back to its source, which means that under most conditions an object coated in the material should appear the colour of one's pupil, i.e. deepest black. Since this would appear to be a useful defence against lasers — not just reflecting the beam away but back at the weapon itself — I could see PCM cloaks knocking around in Severian's world, a utilitarian artifact become fashion statement become tradition.

#87 ::: Niall McAuley ::: (view all by) ::: December 04, 2007, 06:43 AM:

Greg @82: A physicalist would not say that the mind does not exist. You sound like a physicalist yourself when you say @33: I'm certain my experience of emotions could be mapped into a collection of synaptic firings. If you believe your mind is a result of the physical properties and functions of your brain, you are some sort of physicalist.

It's the non-physicalist position which is harder for me to grasp. My mind is something else? What else is there?

#88 ::: Matt McIrvin ::: (view all by) ::: December 04, 2007, 08:01 AM:

In Searle's case I think it's pretty obvious that he doesn't want to believe in physicalism, and is grasping at whatever straw he can find to deny it. I take from his arguments that he finds the notion of a non-physical consciousness somehow ennobling, and the idea that the human mind and consciousness†† are physical is just degrading to human dignity. Do I need to point out that this is a religious argument, and not in any sense either a scientific or philosophical one?

If I recall correctly, Searle claimed it was the other way around: he was the true physicalist and strong-AI advocates were dualists, because they thought mind was the instantiation of an algorithm, independent of its physical substrate, whereas Searle insisted it was something that could only be done by a meat brain made of a particular kind of matter. But I never got what he thought these special "causal powers of the brain" consisted of.

#89 ::: Matt McIrvin ::: (view all by) ::: December 04, 2007, 08:05 AM:

...As for whiter-than-white, most white fabrics you see already are whiter-than-white to some extent, because they've got fluorescent pigments in them that convert ultraviolet light to visible. You'll discover this if you enter a room lit by ultraviolet "black light" while wearing a white T-shirt or white socks.

#90 ::: Carrie S. ::: (view all by) ::: December 04, 2007, 08:15 AM:

I've seen red plastic sheets being sold for quilters, so they can tell if two fabrics are in the same range of values (light/dark). At least, that's my (possibly incorrect) understanding of the theory behind it.

Another trick for that sort of thing is to put the fabrics on a copier. That way all the hue and intensity gets stripped out and you can just compare values. Very handy for planning Fair Isles patterns, where the values are the most important things.

#91 ::: Lila ::: (view all by) ::: December 04, 2007, 08:47 AM:

LLA @ #85, the first link in # 73 suggests that colorblind boys have tetrachromat mothers.

#92 ::: Greg London ::: (view all by) ::: December 04, 2007, 08:51 AM:

Niall@87: You sound like a physicalist yourself when you say @33: I'm certain my experience of emotions could be mapped into a collection of synaptic firings.

Except "mapping into" doesn't mean the same as "completely describe". It's like taking a three dimensional object and mapping it into 2D.

It's the non-physicalist position which is harder for me to grasp. My mind is something else? What else is there?

Subjective awareness or any other experience that cannot be measured empirically. You can look at some mapping of voltages at every synapse and you might be able to say, the person is feeling angry, or the person is feeling happy, but that's not the same as experiencing anger or happiness.

Taking the objective and subjective and pushing it down into nothing but objective creates a flatland, a multidimensional world compressed down into one dimension.

#93 ::: Bruce Cohen (SpeakerToManagers) ::: (view all by) ::: December 04, 2007, 09:15 AM:

Matt McIrvin @ 88

I never took that "special powers" idea of Searle's very seriously, because it just sounded like yet another special plea for the uniqueness and supernaturalness of human consciousness. Maybe I'm not giving him the respect he deserves, but his arguments always seem to boil down to "I can't be made just of matter, that would be too gross!"

#94 ::: Bruce Cohen (SpeakerToManagers) ::: (view all by) ::: December 04, 2007, 09:29 AM:

NelC @ 86

Actually, I didn't notice it until about a year ago, because the change was very gradual until then, and my vision system corrected for it. But when the cataracts started, the cloudiness increased quickly, and I noticed all the changes at once. If the cataracts hadn't occurred*, I might not have noticed until it became difficult to distinguish colors that were easy for me before.

Phase conjugate mirrors were proposed in the 80s as a way of guiding high power laser beams to kill incoming ICBM warheads. The idea is to use a low-power laser to illuminate the target and have the mirror track the returned beam. Then fire the intense beam into the mirror and it automatically hits the target with optimum energy transfer. So what happens when an offensive PCM is used against a defensive PCM? I bet something burns out and throws sparks all over the bridge of the Enterprise.

* The doctors tell me that eventually everybody gets them, so this is a very hypothetical idea.

#95 ::: Niall McAuley ::: (view all by) ::: December 04, 2007, 09:41 AM:

Greg @92: Except "mapping into" doesn't mean the same as "completely describe". It's like taking a three dimensional object and mapping it into 2D.

I think it is probably the other way around. Our nevous systems are doing colossal amounts of sense data processing, pattern-matching, memory access and the like all the time, and the thing we call our minds is a relatively simple summary abstracted from all that activity.

Going from the brain-state to the mind-state would involve enormous simplification, throwing away gigantic amounts of detailed information to give a much simpler result.

#96 ::: Niall McAuley ::: (view all by) ::: December 04, 2007, 09:52 AM:

On the subject of unearthly colours in fiction, Gutenberg has A Princess of Mars, ERB from 1912:

Then a door opened at the far side of the chamber and a strange, dried up, little mummy of a man came toward me. He wore but a single article of clothing or adornment, a small collar of gold from which depended upon his chest a great ornament as large as a dinner plate set solid with huge diamonds, except for the exact center which was occupied by a strange stone, an inch in diameter, that scintillated nine different and distinct rays; the seven colors of our earthly prism and two beautiful rays which, to me, were new and nameless. I cannot describe them any more than you could describe red to a blind man. I only know that they were beautiful in the extreme.

#97 ::: Individ-ewe-al ::: (view all by) ::: December 04, 2007, 10:16 AM:

LLA at #85, that's a fascinating account. You have a minor misconception about the genetics of colourblindness, though: your father isn't colourblind because his Y chromosome "defeated" him. He's colourblind because he has only one X chromosome, and that defective. You aren't colourblind because you benefit from the redundancy provided by your maternal X chromosome (ie you have one defective X, which does nothing, and one which works correctly and gives you full colour vision). Colourblindness, in common with many other genetic conditions that primarily affect men, is nothing to do with the Y chromosome.

I'm surprised you feel so strongly that your childlessness is a genetic benefit. It goes without saying that it's up to each individual whether to have children or not. But everybody has some genes with negative effects in some circumstances. A person who does want to have children might hope not to pass on a particular trait, that I could understand, but you almost seem to be arguing that the world is a better place for not having your hypothetical colourblind child in it.

#98 ::: Niall McAuley ::: (view all by) ::: December 04, 2007, 10:24 AM:

Further to Individ-ewe-al's point, unless a defective gene in LLA is a very recent mutation, it certainly exists in multiple copies out in the population at large. Refusing to pass it on won't bother that gene much.

#100 ::: John Stanning ::: (view all by) ::: December 04, 2007, 11:06 AM:

Niall #95 et prev* : I suppose the mind is an abstraction of the brain.

Or put it another way: the brain is the host (hardware**) on which the mind (software) runs. If your consciounsness and personality could be replicated with AI, it would be your mind that would be replicated, not your brain.

* what's the opposite of "et seq."?
** if you've handled a brain, "hardware" is an odd word, but it fits the computing analogy.

#101 ::: Niall McAuley ::: (view all by) ::: December 04, 2007, 11:25 AM:

John @100: If we're using computer analogies (which aren't necessarily any more accurate than steam engine or radio analogies, imo) I would say there's a whole operating system of software running below the level of the mind, doing all the aforementioned data processing, memory retrieval and pattern matching, before you get to the wetware.

#102 ::: C. Wingate ::: (view all by) ::: December 04, 2007, 11:33 AM:

re 88: As I argued quite a ways back in this, the strong AI guys are dualists.

re 76: The passage you quote first is still standing on the "knowing about" side rather than the "knowing". I don't think such a "trap" is acutally possible without it being an external mechanism that tells her she should be having a certain physiological reacion, but in any case the issue of how she can have the knowledge of the experience itself can be gotten without having the experience.

I would note that this all relies on one element of ignorance and one of ignoring. The ignorance, obviously, is that we don't know that much of what's going on in the brain. The ignoring is that we already know that the brain can function (defectively) in states where it is aware (that is, responsive to the outside world) but not conscious (that is, not self-aware in the normal way).

#103 ::: John Stanning ::: (view all by) ::: December 04, 2007, 12:12 PM:

Niall #101 : yes, that's better - the operating system gets on with routine functions without bothering the 'mind' level.

And yes, the AI analogy is just that, an analogy, not necessarily accurate and certainly not to be taken too far or too literally.

"wetware", yes, that's the word.

#104 ::: Tim Walters ::: (view all by) ::: December 04, 2007, 12:17 PM:

As I argued quite a ways back in this, the strong AI guys are dualists.

That's like saying that because both humans and boats can propel themselves through the water that both must rely on some mystical essence of swimminess.

The passage you quote first is still standing on the "knowing about" side rather than the "knowing".

Well, one of the things she "knows about" is exactly what set of thoughts to think to see the color red in her mind's eye. This distinction simply isn't as clear as Jackson wishes to make it, at least given his dictum that she knows "all about" color.

But either way Jackson loses. If all she has is "knowing about," then fine, but clearly she won't have the brain state of someone who has seen the color red and been told that it was "red", so there's no issue. If he would rather specify that her course of research does give her this brain state, then Dennett and I say that she will recognize the color red, or at least that he can't say she won't without circular reasoning.

the issue of how she can have the knowledge of the experience itself can be gotten without having the experience.

That's not the issue at all, at least from the point of view of questioning physicalism. Either some training course for recognizing red without having actually being exposed to red light exists, or (more likely, I think, barring SF scenarios about directly twiddling brain parts) it doesn't; but the idea that this has some relevance to the independent existence of qualia is a red (or green) herring.

#105 ::: Zack Weinberg ::: (view all by) ::: December 04, 2007, 01:21 PM:

#104: Gleep! I hadn't thought about it much, but I was assuiming that the Mary's room thought experiment had some deep flaw in its premises that rendered it not-even-wrong, because that's my default assumption about all such thought experiments. But what you said there caused me to realize just how utterly broken it is.

Mary has, by assumption, spent her entire life in a world with no color information. But if you take an animal and deprive it of some aspect of sensory experience from birth, the processing areas for that aspect don't develop. For example, if you take a kitten and raise it for the first six weeks of its life in a box painted with vertical stripes, so it never sees a horizontal line, it is ever after unable to recognize long horizontal objects. (See e.g. http://scicom.ucsc.edu/SciNotes/9502/Lines.html -- mostly about something else, but contains the most coherent short explanation I can find, and a reference to the original paper.)

Mary can't have an experience of red upon leaving her room, even if she knows exactly what pattern of neuron activation corresponds to that experience in a normal brain, because her brain's color-processing areas have not developed normally. (They won't have totally atrophied; they'll have been reassigned to texture recognition or something like that.) I think it is not too farfetched to say that Mary will not have any sort of new experience upon leaving the room, and that therefore the thought experiment tells us nothing interesting.

#106 ::: C. Wingate ::: (view all by) ::: December 04, 2007, 01:52 PM:

re 104: Hmmmm, I think independent existence is a red herring. For some value of independence, anyway. Or "mystical", for that matter.

As far as physicalism meaning that we can describe everything solely using the terms of physics, the fact is, we can't. One is prompted to ask, can we describe Hamlet solely in terms of physics? Even if potentially we could produce such a description, I don't feel any compunction to believe in it until it is produced. There's nothing mystical about this. It's simply the assertion that we will always need other languages to describe things besides physics, as is the case now.

I think you missed the point of the strong AI case. If one makes a simulation of an automobile, there is no level of detail at which the simulation becomes real. If one models a triangle with lines on a paper or with bits in a computer image, however, within the context of the image, both are equally real. What the strongest AI position asserts is that mind is metaphysical in exactly the way that the triangle is, and that a sufficiently detailed "simulation" of a mind is in fact real. This seems to me to be quite dualist, but not supernatural.

As far as the supernatural is concerned: while Christianity classically took a dualist position, a lot of theologians have abandoned that degree of Neoplatonism. One would hope, at least, that there is some test that could be made as to whether people have souls in a supernatural, as opposed to metaphysical, sense (I can conceive of a metaphysical soul that is nonetheless contained entirely in the physical). But very plainly we have no such test, other than through appeal to religion.

#107 ::: C. Wingate ::: (view all by) ::: December 04, 2007, 02:05 PM:

re 105: I think that aspect can be gotten around by taking a different case. For example, we could work with the electric shock case I mentioned earlier. It doesn't require any sort of preconditioning, unlike (ostensibly) visual development. But Mary can be made to learn about it. If one could establish that everyone has the same underlying nervous system reaction to the stimulus, could you put Mary in such a state to where the sensation is not novel to her without essentially dropping a memory of the experience into her head? And (the more important point, I think) could one describe her reaction entirely phyiscally? At present, one could not; one has no choice but to use mental categories in the description.

#108 ::: Tim Walters ::: (view all by) ::: December 04, 2007, 02:53 PM:

C. Wingate @ 106: As far as physicalism meaning that we can describe everything solely using the terms of physics

I don't think it does mean that. The Wikipedia link defines it as "the view that the universe, including all that is mental, is entirely physical"--which is rather different. In Dennett's own example, to understand why a particular copper atom is located thirty feet above Trafalgar square, you're going to need to know about British naval history; but that doesn't mean you need new chemistry.

If one makes a simulation of an automobile, there is no level of detail at which the simulation becomes real. If one models a triangle with lines on a paper or with bits in a computer image, however, within the context of the image, both are equally real.

Either the simulation is virtual--in which case the simulated automobile is as real or unreal as the triangle--or not, in which case the printout or screen showing the triangle is as real as the capable-of-transporting-humans pseudo-auto.

Similarly, when a simulated mind uses its speaker to talk to you, it moves real air. There's nothing more or less metaphysical about it than about any other device.

What the strongest AI position asserts is that mind is metaphysical in exactly the way that the triangle is, and that a sufficiently detailed "simulation" of a mind is in fact real.

The strongest AI position I know of is that it's possible in principle to design a non-organic entity whose behavior will give us just as much reason to regard it as conscious as we have for other human beings.

Searle's position is that while a computer "really" computes prime numbers, its consciousness, even if developed to the point where it would be completely indistinguishable in behavior from that of a human, would only be "as-if" consciousness, and that some further ineffable ingredient would be needed for "true" consciousness. That is dualism. If one wishes to regard strong AI as dualist because you feel that algorithms have a metaphysical existence, then Searle is a... triplist? triunist? Anyway, he always believes in one more type of thing than the strong AI proponent does.

#109 ::: Mary Dell ::: (view all by) ::: December 04, 2007, 03:50 PM:

Nix @#29: right, fuligin. I was trying to remember the right spelling and accidentally googled my way to a poem about Terminus Est.

It's...um.

#110 ::: Zack Weinberg ::: (view all by) ::: December 04, 2007, 07:14 PM:

C. Wingate #107:

If one could establish that everyone has the same underlying nervous system reaction to [an electric shock], could you put Mary in such a state to where the sensation is not novel to her without essentially dropping a memory of the experience into her head?

Even in a thought experiment, I cannot suspend disbelief in the hypothesis - a neurologically normal adult who has, somehow, never experienced physical pain in her entire life? No way. I'm not just whingeing; if we have to stretch the hypothetical that far, I doubt it can tell us anything useful about how things are in real life.

(There is such a thing as congenital insensitivity to pain - see e.g. SCN9A channelopathy causes congenital inability to experience pain (may require Nature subscription) - but that would be the same category as the horizontal-stripe-deprived cats.)

But really, we don't have to go so far afield to make a sane hypothetical; we don't even have to go as far afield as the original. I'm going to point all the way back at Greg's comment in #2: "knowing about" is not the same as "knowing". Or, in the terms we cognitive scientists like to use, declarative memory is not the same thing as skill memory. Consider the first time you ever drove a car, for instance; you probably had substantial book education in how to do that before you ever sat down behind the wheel, but you didn't have the physical skill, so the experience was novel even though you knew what to expect would happen.

But here's the thing: we don't need qualia or anything to explain that. In the case of driving, can point at the entirely separate brain areas that have been shown to be involved (motor skills in the cerebellum, declarative knowledge in high-level sensory cortex). You can come up with hypotheticals that, while still vaguely plausible, involve no motor skills - imagine Mary has never heard a piece of music in her life, but has otherwise heard a normal range of sounds. The "visceral" response that a piece of music evokes is also skill memory, and can't be trained by anything but the sounds themselves; all the music theory in the world is declarative memory, and they do not meet.

#111 ::: Zack Weinberg ::: (view all by) ::: December 04, 2007, 07:15 PM:

(Also addressed to C.Wingate:) I'd like to understand the distinction you're making between a supernatural soul and a metaphysical one, but I have no idea what you are talking about. Can you clarify please?

#112 ::: elise ::: (view all by) ::: December 04, 2007, 08:05 PM:

Zack @110: An electric shock is not the only kind of physical pain. It is, however, a very specific kind, one with known effects in most people.

#113 ::: elise ::: (view all by) ::: December 04, 2007, 08:07 PM:

Sorry, posted before I should have. What I mean is that having some exposure to basic physical pain (accidental bumps, bruises, cuts, and so on) is the equivalent of seeing things on the black and white screen. An electric shock would be the red, the new experience.

#114 ::: Zack Weinberg ::: (view all by) ::: December 04, 2007, 08:19 PM:

In that case, I don't see it as different from the "never heard a piece of music" scenario. The visceral response will be the source of the novelty, is not trainable with words, and arises from entirely different brain areas...

#115 ::: Avram ::: (view all by) ::: December 04, 2007, 09:01 PM:

From an episode of Futurama: (Fry is a 20th century person living in the 30th century. Amy is a 30th century person.)

FRY: I just saw something incredibly cool. A big floating ball that lit up with every color of the rainbow, plus some new ones that were so beautiful I fell to my knees and cried.

AMY: Was it out in front of Discount Shoe Outlet?

FRY: Yeah.

AMY: They have a college kid wear that to attract customers.


#116 ::: Julie L. ::: (view all by) ::: December 04, 2007, 09:16 PM:

Avram @58: Fuligin is used for things that are absolutely black, reflecting no detectible light at all, while black would presumably still be used for ordinarily black object that reflect enough light to have discernible surface contours.

Latin has two different words each for black (ater and niger) and white (albus and candidus), iirc for shiny vs. matte distinctions though I don't remember now which is which.

#117 ::: Lee ::: (view all by) ::: December 04, 2007, 09:46 PM:

Mary Dell, #109: The word you're looking for is "poor".

#118 ::: Nicole J. LeBoeuf-Little ::: (view all by) ::: December 04, 2007, 10:13 PM:

This probably has less to do with the topic at hand than I think, but, some years ago, my flight instructor was ground-schooling me in advance of my first night flight, telling me all about the different kind of airport beacons. "So the Aeronoutical Information Manual [AIM] will tell you that lighted land airports use a beacon that alternates green and white pulses. But, you know what? It always looks blue to me."

So we went up in the air and flew our night cross-country from Boulder to Cheyenne, and damn straight that beacon's "green" pulse looks blue.

More: It looks like the same blue, to me, as the taxiway edge lights--which the AIM describes as "blue."*

So, as far as my color-sense goes (and I'm not, to my knowledge, colorblind; they let me fly, after all) the AIM is apparently using the words "blue" and "green" interchangeably. Surely this can't have anything to do with some languages only having one word for "grue"? Airport beacons are pretty darn recent! What's up with this?

*Yes, I have my copy of the 2007 FAR/AIM open to AIM 2-1-8 as I write this. Yes, I'm a geek. Yes, I am also sufficiently intimidated by the collective erudition of You People™ that I would go dig up my copy of the 2007 FAR/AIM and find the section dealing with Airport/Heliport Beacons before hitting the big scary "post" button.

#119 ::: P J Evans ::: (view all by) ::: December 04, 2007, 10:19 PM:

Nicole @ 118

Well, from the ground, they look different to me. (My train goes past two airports, so I see both.) I will say that the beacons are sort of a bluinsh green, though. (Not that this is too strange: I remember that the pale green paint we had in one house looked blue in some kinds of light, and deeper green in others.)

#120 ::: Lindra ::: (view all by) ::: December 04, 2007, 10:21 PM:

Whenever I hear about colour-blindness, I think of a conversation I had with my Chemistry teacher, who was quite colour-blind. One day he told us he had his wife pick out clothes for him and demostrated how hard it was for him to tell the red and green whiteboard markers apart from each other. I used an FM system to hear him in class and I had to wait for him to give me my microphone back, which invariably occured to him only after the rest of the students left, so we had a lot of short conversations while he untangled all the wiring. After that particular class he asked me about my deafness, if it caused me similiar kinds of trouble in being unable to discriminate things other people could, and I said yes. He then told me that he thought it would be more awful to be deaf, and he'd much rather be colour-blind. I told him I thought it would be more awful to be colour-blind, and I'd much rather be deaf. Each to their own.

The results of the study mentioned in the blogged article seem obvious to me because the same thing happens to me aurally. I hear sounds as a result of some tricky (possibly synaesthestic) brain wiring that are functionally impossible for me to hear physically. They're also qualitatively different from anything else I hear with or without hearing aids. I call them 'my personal symphony', but 'Martian sounds' is admittedly much cooler.

My quibble with Mary's Room and other such philosophical arguments posed in attempts to prove this or that hypothesis is how nonsensical most of them are. It's a nice theory that Mary could learn all about the firing of this or that neuron and the impression this or that collective firing made upon the brain and this or that area was stimulated and therefore this or that connected with this or that to link this or that concept stored in this or that area, but I think we just don't know enough about the physical working of the brain to make this hypothesis in the first place.

#121 ::: vian ::: (view all by) ::: December 04, 2007, 10:34 PM:

Mary Dell, #109

Thanks - I'd been missing Really Bad poetry, but I feel all caught up now.

#122 ::: elise ::: (view all by) ::: December 05, 2007, 12:19 AM:

Lindra @120: The results of the study mentioned in the blogged article seem obvious to me because the same thing happens to me aurally. I hear sounds as a result of some tricky (possibly synaesthestic) brain wiring that are functionally impossible for me to hear physically. They're also qualitatively different from anything else I hear with or without hearing aids. I call them 'my personal symphony', but 'Martian sounds' is admittedly much cooler.

Dang. Wish I could listen in to your brain, there.

Hi, by the way. I'm Elise. I'm moderately hearing impaired, for reasons having to do with delayed nerve impulses in the brain stem, as near as they can figure out. I have stuff going on that seems pretty darned synesthetic -- though nobody's officially called it that -- but the one place I don't get it is with hearing. I mean, no other stuff produces sounds. Except one thing: when I am falling asleep and sometimes have that myclonic jerk thing happen, the dang thing always has a huge airhorn-volume brass fanfare with it. Anyhow, pleased to meet you! And the Martian sounds thing is fascinating.

Here's to learning more about the brain. (Maybe someday they can figure out why my hearing, after 20 years of going away, suddenly took a jump upward again about two years ago... and recreate that effect elsewhere.)

#123 ::: Greg London ::: (view all by) ::: December 05, 2007, 12:35 AM:

Tim@108: The Wikipedia link defines it as "the view that the universe, including all that is mental, is entirely physical"

Isn't that the same as materialism? So many isms. So little time.

If it is, then the whole Mary's Room thing is just stupid because it's talking about a hypothetical situation like in the Matrix where Trinity learns to fly helicopters with a simple download, without ever having to touch a helicopter.*

And that (If that's what Mary's Room is) is just nonsense because it's using a hypothetical situation beyond current technology to (dis)prove the existence of some nonempirical, currently nonmeasurable, level of experience that we can only call the "subjective".

Yeah, sure, IF we could demonstrate the hypothetical situation of downloadable helicopter training, THEN we could say that it's all in the download bits, and there's nothing in the experience. But until we can actually do that, it's all science fiction, not proof.

(*)Damn her.

#124 ::: Margaret Organ-Kean ::: (view all by) ::: December 05, 2007, 01:51 AM:

After an experience this summer I can no longer think of 'color-blindness', at least of the red-green sort, as a defect.

I went to the San Diego Zoo in the early evening (a good time to go, many of the animals are active). Many of the animals are in natural exhibits, and while it's pretty easy to spot the elephants, giraffes, and zebras, some of the smaller ones can be quite difficult to spot.

Now, I'm an artist. I'm supposed to be trained to observe things and see small, telling differences. I should be able to spot these difficult to see animals. But in every instance, my red-green color-blind companion spotted them first; spotted all of them; and saw them in enclosures everyone else thought empty. And did it very quickly.

So now I wonder if at some time red-green color-blindness was a useful thing to have - if you wanted meat for dinner.

#125 ::: Lee ::: (view all by) ::: December 05, 2007, 02:26 AM:

Mary Dell and Vian -- here, try this one instead. There's a fanvid from the CD recording here, and it's definitely worth a listen.

#126 ::: Mez ::: (view all by) ::: December 05, 2007, 06:11 AM:

Margaret #124, or if you wanted to avoid becoming dinner. For a very long time, I suspect, human ancestors were more prey than predator, or at least equally; something 2001: A Space Odyssey acknowledged.
We are so used to the advantages we've inherited and the changes humanity has made in the world that that's easy to forget. There are a few SF&F stories around that try to recapture that situation.

#127 ::: Matt McIrvin ::: (view all by) ::: December 05, 2007, 09:44 AM:

The strongest AI position I know of is that it's possible in principle to design a non-organic entity whose behavior will give us just as much reason to regard it as conscious as we have for other human beings.

The positions get much stronger than that. For instance, I think Roger Penrose might agree with you, and he's anti-strong-AI. It's just that he'd say this non-organic entity would have to possess some apparatus for doing the same weird quantum whatsis that he says goes on in microtubules; and the strong-AI folks (along with a lot of other people) would disagree with that.

My impression of the strongest strong-AI position is that the essence of conscious thought is an algorithm, or, put another way, the evaluation of some enormous function; and any computing machine that could instantiate such an algorithm, regardless of its physical construction, would be as conscious as a human while doing so.

Personally, I think that this may be true but may also be moot, for practical reasons of performance. In other words, if you ran the consciousness-simulating algorithm on any general-purpose computer of conventional construction, it might run so slowly that you'd wait your whole life for it to grind out part of a thought, and doing better would require tailor-made architectures bringing the machine closer to an analogue of a physical brain (though maybe not so close that you'd need to build it out of meat).

#128 ::: Bruce Cohen (SpeakerToManagers) ::: (view all by) ::: December 05, 2007, 10:01 AM:

Lindra @ 120, Greg London @ 123

That's precisely the point I was trying to make when I mentioned the "zombie" thought experiments* upthread. How useful is an argument that assumes the existence of people without consciousness who are different from us, in order to prove the existence of consciousness?

There's a truly monumental amount of silliness exhibited in discussions of the philosophy of consciousness. I think there wouldn't be anywhere near as much nonsense if we weren't conscious :-)

The key point is that hypotheses about the way the mind works, and how connected it is to the way the brain works, are highly suspect, and certainly premature, in our present woeful ignorance of how the brain works. Give neuro-imaging a couple of decades to work on it and we'll be in a much better position to talk about these things without making up ridiculous Just-Ain't-So stories.

* I'm tempted to put "thought" in quotes too, because I think this sort of speculation is brain-damaged. And I use that phrase in all irony, remembering that much of what we know about the brain and mind comes from examining the behavior of ones that don't work right.

#129 ::: Heresiarch ::: (view all by) ::: December 05, 2007, 10:07 AM:

Tim Walters @ 105: "That's like saying that because both humans and boats can propel themselves through the water that both must rely on some mystical essence of swimminess."

Or that because a sound wave can propagate through both air and water it must have an independent existence from them. I see consciousness primarily as a relationship of things; what those things are (cells, chips) isn't as important as how they are arranged and how they interact.

C. Wingate @ 106: "As far as physicalism meaning that we can describe everything solely using the terms of physics, the fact is, we can't. One is prompted to ask, can we describe Hamlet solely in terms of physics?"

As a physicalist, I would say Hamlet has already been described by physics, in the sense that it exists already within our entirely physical universe. The idea is that the material world taken as a whole can explain everything within it, not that any individual person can do so.* Clearly the description the world itself offers, i.e. reality, will be superior to any that a human being could offer.

*I feel like this is a parallel argument to one medieval mathematicians might have had about whether it was possible to calculate the area under a curve. Simply because their mathematics couldn't do it yet doesn't mean mathematics can't do it at all.

#130 ::: Bruce Cohen (SpeakerToManagers) ::: (view all by) ::: December 05, 2007, 10:20 AM:

C Wingate and Tim Walters

Much of the discussion of strong AI is hampered by the fact that its opponents don't really know what it is. They seem to think it's a strongly materialist philosophical position; it's nothing of the sort. It's a statement about the feasibility of a complex engineering task: the construction of a system that a human being could reasonably attribute human or human-like intelligence and consiousness to.

Very similar to what the unintelligent Intelligent Design arguers have done, opponents of strong AI* have tried to re-cast its definition so as to make it easier to deny. That's part of the whole consciousness debate: materialism treads on a lot of religious and emotional toes in ways that strong AI does as well, so they get conflated even in the minds of people who don't care particularly about the debate.

The distinction between materialism and dualism is not that materialism insists on a single level of description for all things, i.e., that I can write differential equations that describe human emotions in the language of subatomic particle interactions. Materialists recognize that descriptions of the world require many levels of organization and many languages to describe. However, materialists insist that it is at least in principle possible to describe the language used to talk about a domain with the languages of "neigboring" domains. For example, the rules of the operation of living cell can be explained in terms of chemistry and nano- and micro-scale physical interactions. Important note: it is not the case, as far as anyone has proved, that any one of those languages causes the nature of another to be exactly what it is. There are constraints, certainly, but we don't know for instance that biological reproduction can only use DNA as an information substrate because of the way the laws of physics work.

Dualism, on the other hand, asserts that there is a fundamental break between the physical and the mental, which most people interpret in spiritual terms. Dualism says that mental states can cause physical brain states, but that brain states can't in any way affect mental states. Materialism says that cause and effect can and do run both ways.


* disclaimer: I personally believe that strong AI in some sense is possible; I know enough about how hard it is to prove this that I'm agnostic on whether it will ever in fact be done, or even can with the level of intelligence that humans can muster.

#131 ::: Mr. Chris ::: (view all by) ::: December 05, 2007, 12:22 PM:

In response to Bruce Cohen @ 128:

Your objection to the "philosophical zombie" thought experiment reminds me of hearing some of my classmates, many year ago, complain about how terrible it would be to live in the society described by Borges in "The Lottery of Babylon." My classmates were horrified by the idea that the a random lottery conducted by a secretive, unknowable organization could direct the course of the lives; the narrator of the story is horrified by the idea such a lottery doesn't exist.

It occurred to me that the reality described by the narrator was indistinguishable from the one my classmates and I inhabited, the only difference being the existence of undetectable human agents orchestrating those events which we can observe. I expressed the opinion that any comments about how awful it would be to live in the world described in the story were premature until it could be established that we weren't living there ourselves.

This is essentially how I feel about the "zombie" thought experiment. My response to being told "imagine a world just like ours, except that no one has consciousness," is "how do I know the people in our world don't already lack consciousness?" Why imagine how such a world would be different if I can't know it actually is different? For all I know, a world without consciousness could be the same as a world without unicorns: we might already live there.

I also have a problem with the Mary's Room thought experiment, in that it combines the impossible (someone with complete knowledge of the physical facts of people's experience of color) with the mundane (leaving a black and white room and seeing the color red). It's possible that the ease with which most people can imagine the mundane experience of leaving the room makes them more likely to assume that Mary obviously learns something new upon leaving the room.

I think that this represents a critical failure of the imagination, however. Most people do have the experience of the color red, so it's easy to assume that that experience can't be replicated by merely knowing all the physical facts about how people perceive color. What happens, though, if Mary isn't studying how people ordinarily perceive color? What if she's studying the experiences of someone who is completely colorblind but, for whatever reason, experiences a particular taste sensation when observing the wavelength that we perceive as "red?"

If Mary can somehow rewire her own brain to replicate that sensory experience, can we say that upon looking at a red object and "tasting" it for the first time, she has really learned something about how this one other person experiences the color red, something she couldn't have learned from studying how the other person's perceptions of color stimulate the parts of the brain responsible for creating the sensation of taste? How can we even be certain she really is having the same experience? If abnormal perceptions can have purely physical causes, why can't normal perceptions also have purely physical causes?

I'm not quite sure how to express the point I'm trying to get at, but it seems to me that the pro-qualia elements of the thought experiment are based on the fact that most people CAN perceive redness on a visual level, so it's relatively easy for people to come to the (in my opinion probably fallacious) conclusion that sure, there's something about one's perception of the color red that goes beyond mere physical information.

I guess this is just a very long-winded way of saying that I think I agree with Niall McAuley @ 95 that "our mind is a relatively simple summary abstracted from all that activity." Considering the sheer complexity of the human nervous system, I don't really see any reason to assume the existence of perceptual experiences independent of physical information. To say that such experiences must exist (or even are likely to exist) would require a much more intimate knowledge of the human nervous system than we presently have.

#132 ::: Nicole J. LeBoeuf-Little ::: (view all by) ::: December 05, 2007, 12:24 PM:

P J - OK, well, then, maybe it's just me... or maybe it's been too long since I've been in a position to compare the lights' colors closely. In my imagination they're the same, but memories do have a way of morphing. (On my drive home, I see the Rocky Mountain Metro taxi lights but not its beacon as I pass Broomfield; I see the Boulder beacon but not its taxi lights when I crest the hill coming into Boulder. I haven't flown at night in some years, and I can't really remember looking too hard last commercial flight I took.)

Maybe the gel / filter on that beacon is very much green when you get up close to perform maintenance on the bulb, but the light it produces blues to some extent in action?

#133 ::: NelC ::: (view all by) ::: December 05, 2007, 01:19 PM:

Greg @123: It's possible that Trinity doesn't download the helicopter pilot module into her own brain, but just uses a Matrix subroutine to interface between her and the simulated helicopter. She thinks "hover" and the module does the stuff which causes the helicopter to hover in the simulation. It would be like flying a simple PC flight sim, or perhaps a complicated flight sim with the easy options dialled to the max.

For added verisimilitude, the module could then translate its control actions into those of Trinity's simulated body, so that when the module adds collective her arm moves just so. Since Trinity would be feeling the sensations that would cause a real-life pilot to move the collective just so, and finding herself doing it, it's possible that it would seem to her that she was actually piloting the helicopter in the same way that a real pilot is conscious of it. That part of her consciousness actually resides in a helicopter pilot module that exists solely in the Matrix only becomes relevant if she tries to pilot a helicopter in the real world and finds she doesn't quite remember how.

Which I guess means that Trinity's identity might be slightly more fluid than we find comfortable, Trinity in the real world being less than Trinity plus wire-fu gun-fighting helicopter pilot modules in the Matrix. Same for Neo, I guess, but more so.

On the other hand, we've been putting part of our consciousness external to our brains since the invention of writing, which I recall one of the ancient greeks grouched about because it meant that people's memories weren't quite as good as before.

#134 ::: Diatryma ::: (view all by) ::: December 05, 2007, 01:24 PM:

Nicole, maybe it's something like the green dye that turns blue under gaslight. I've had a couple shirts do that-- turquoise in sunlight and fluorescent, blue in the school gym.

#135 ::: Syd ::: (view all by) ::: December 05, 2007, 04:52 PM:

Margaret @ 124:

In Sex, Time and Power, Leonard Shlain lists a series of characteristics (including color-blindness) carried by 8-10% of the population and gives the possible evolutionary advantages of each. Whether you buy into his premise or not, it's an interesting discussion.

#136 ::: Edward Oleander ::: (view all by) ::: December 05, 2007, 07:19 PM:

I think the reason that the dualist view is so often associated with the religious/spiritual community is that the dualies use an argument very close to that of the Intelligent Design community.

I call it the Fallacy of Impatience. It boils down to saying that since Science hasn't yet managed to explain the entirety of Theorum X, then Theorum X will NEVER be explainable by natural processes, and therefore a supernatural explanation is the only possible conclusion.

I can see where this idea would be enticing to those who want to separate the mind from the brain. I guess it all boils down to where you're comfortable placing your unsupportable assumptions, because ALL major theorums have a couple of these hiding somewhere.

I would rather believe that since there are several additional dimensions (according to Hawking and his crowd, up to 9 more than the ones we can directly experience), we might find someday that the seemingly metaphysical/mystical connection between mind/soul and brain have a connection we just can't see yet.

Under Occam's Razor, ANY physical explanation is more likely than a pure thought solution, even if we cannot see the relationship yet.

~Ed~

#137 ::: Vicki ::: (view all by) ::: December 05, 2007, 10:08 PM:

Margaret @ 124: I read years ago (maybe something by Oliver Sacks) about a small Pacific Island where everyone is color-blind, because of founder effects. And they are, indeed, better at hunting, or spotting anything moving in the jungle, than people with normal color vision. (Human and other great ape color vision is hypothesized to be at least in part an adaptation for finding ripe fruit and tasty new leaves.) ISTR that color-blind people are better than the rest of us at detecting people in camouflage uniforms, jeeps painted in that color pattern, and the like.

#138 ::: C. Wingate ::: (view all by) ::: December 05, 2007, 10:37 PM:

My distinction between "metaphysical" and "supernatural" (which I admit is a little non-standard) is intended to make a distinction which I see being obscured if not outright fudged.

The other thing which "supernatural" signifies is the realm of the spirits-- gods, angels, kamis, ghosts, etc. It almost by definition lies outside of the purview of science, but by the same token the kind of discussion we're having can say next to nothing about it.

We are really talking here about what I'm terming the metaphysical, which is to say, things whose existence isn't utterly bound to physical existence. Obviously they can be presumed to be so represented as far as they require a physical manifestation to "touched" in this world. But let's consider Hamlet again. When we talk about representing it purely physically we need to make more of a commitment to what the representation actually is. Words on a page, for example, do not cut it, for they are simply representations of the abstract words and can be re-presented in an array of equivalent media. At least to that level, the words are Hamlet; the ink and paper are only Hamlet insofar as they represent the words.

Now the words: one could assign them as symbols of mental states. Well, OK, except for two things. First, I'm not inclined to take claims about them very seriously until they can be described rather than speculated about. Much of the more bogus dualism is plainly enabled by not knowing much about what's going on inside, and it seems to me only fair to hold all the possibilities to the same standard. Second, since the words are symbolic abstractions, it is conceivable that to achieve consciousness it is only necessary to move the abstractions around the right way, and that the underlying representation doesn't matter. SF is full of stories in which it is possible to load a "person" into a computer without having to actually simulate the neurological basis for thought.

It seems to me that a lot of "classic" dualism stems from getting too attached to neoplatonic language. OTOH, I'm increasingly unconvinced by the presumption that representation in brain-state settles the matter. The question that leads to is, can it be otherwise represented and still work? And I'm pretty sure the answer to that is, we will never know.

#139 ::: Tim Walters ::: (view all by) ::: December 05, 2007, 10:50 PM:

Matt McIrvin @ 127: The positions get much stronger than that.

You're right. I wasn't thinking too good when I wrote that.

For instance, I think Roger Penrose might agree with you, and he's anti-strong-AI.

I'm somewhat sympathetic to the idea that consciousness might depend on quantum behavior--it has a certain intuitive appeal--but reading his book makes me rather less so. His actual argument boils down to about twenty pages of hand-waving. Still, just because his "proof" is weak doesn't mean he's wrong...

Personally, I think that this may be true but may also be moot, for practical reasons of performance.

Ray Kurzweil claims that, due to Moore's Law, desktop computers will each have the power of a human brain by June 24 2019 around tea-time. Aside from the obvious problem of relying on extrapolation, though, this assumes that we have a good idea about how much computing power it takes to simulate a neuron "well enough," that simulating neurons is good enough in the first place, and that we can get a lot better at writing code.

C. Wingate @ 138: My question would be: what's the difference in metaphysical status between the design of a thinking machine and the design of a medieval waterwheel, or for that matter the design of a flint axe, or any human artifact?

If we grant that this is dualism, at least it's a proven dualism; I can calculate the same thing* (e.g. the first 10 prime numbers) in my mind or with a computer, so in some sense they both really do instantiate the same algorithm.

*And here I think is the rub; both are only the same thing from a human point of view. From an universal point of view both the piece of paper I wrote on and the computer screen are slightly different flavors of quark soup; from a bacterium's point of view, they're completely different, since one is edible and one is not; etc. So I'm not convinced that Hamlet has any existence outside of human brains and artifacts.

#140 ::: C. Wingate ::: (view all by) ::: December 05, 2007, 11:18 PM:

re 139: Well, the metaphysical difference wouldn't be in what the design is, but what the designed thing does. Which is to say, lots of difference. And part of the issue is that, lacking such a machine, we don't know what those differences would be in their entirety.

I think it would be more accurate to say that Hamlet, as far as we know, only exists through brains and artifacts. Which leads to a "tree in the quad" question: Would Hamlet still exist if there wasn't anyone left who could read it?

#141 ::: Peter Erwin ::: (view all by) ::: December 06, 2007, 06:41 AM:

NelC @ 133:
Greg @123: It's possible that Trinity doesn't download the helicopter pilot module into her own brain, but just uses a Matrix subroutine to interface between her and the simulated helicopter. She thinks "hover" and the module does the stuff which causes the helicopter to hover in the simulation. It would be like flying a simple PC flight sim, or perhaps a complicated flight sim with the easy options dialled to the max.

My impression from the movie is that she's using the same system that Neo used to "learn" the various different martial arts: something separate from the Matrix, built into the Nebuchadnezzar. So, just as Neo got his brain rewired in some fashion so that he acquired the training and reflexes of a martial artist, Trinity's brain is rewired so that she gets the training of a helicopter pilot.[*]

(Besides, if Trinity & Co. could control portions of the Matrix to the extent of telling helicopters how to behave, I think they'd have a lot less trouble getting away from the Agents...)

[*] At least to the extent of being able to control the simulated physics of their bodies in the Matrix or the "Construct" (the training space Neo and Morpheus fight in).

#142 ::: NelC ::: (view all by) ::: December 06, 2007, 08:29 AM:

Peter @141: Well, I didn't go back to the movie to check my thesis, I was more reinterpreting what I could recall in a way that made sense to me. Kind of, "How could I achieve similar results, using the little bit more I understand about computers and neurobiology than the Matrix script-writers do?"

But even so, I'm not suggesting a direct control of the simulation, suspending the simulated rules of physics for the simulated helicopter (which is more what Neo does for his "Superman thing") but an interfacing with an unintelligent subroutine which has the skillset of a helicopter pilot, such that when Trinity thinks of hovering outside Agent Smith's office suite, the subroutine picks up her intention and flies the helicopter so that it does that thing. In the same way that if she actually had those skills in her own brain by virtue of learning them the usual meat way, she'd think consciously, "I'll hover here", and her unconscious skills would cause her to move the collective and cyclic controls so that she'd hover there.

Where the subroutine resides isn't that important; it could be in the Matrix hardware, in the hardware onboard the Nebuchanezer, or even borrowed from some sleeping helicopter pilot's brain plugged into the Matrix. In fact, that latter idea holds some attraction to me, since I'm of the opinion that the Matrix hardware is actually using all those wetware brains as part of itself, and the Matrix simulation is a kind of consensual dream that the humans share when their brains aren't being used as computing modules. The Nebuchanezer would have a way of finding a human with the required skills who is sleeping in the Matrix simulation and co-opting part of their brain from the tasks that the Matrix hardware has for it, borrowing it to supply the helicopter flying skills Trinity requires.

Which makes Trinity's identity even more fluid, since Matrix Trinity would be meatspace Trinity plus a bit of helicopter pilot plus a bit of nth-dan black belt master plus a bit of sniper plus whatever, all of whom have their own consciousness when they awake in the Matrix simulation. I wonder if they dream of Trinity?

#143 ::: NelC ::: (view all by) ::: December 06, 2007, 08:46 AM:

Me @142: The problem I have with "rewiring" the brain to acquire skills instantaneously is that the human brain isn't designed to do that. Unless you can directly manipulate neurons via magic nanomachines or extremely localised magnetic fields or the like, learning probably has to occur at meatspace speeds. Even with direct manipulation, you'd essentially have to force neurons to grow connections at their maximum pace according to a predetermined plan. And the problem with that is that the connections of a brain, while very broadly similar, are unique to each brain. Plonking a helicopter skillset directly into a brain would be like forcing a set of pieces from one jigsaw puzzle into the middle of another, an ugly mess that doesn't quite fit.

The scheme I have above is possibly no less of a handwave, but at least I've added a glamourous assistant to distract the audience.

#144 ::: Bruce E. Durocher II ::: (view all by) ::: December 06, 2007, 09:49 AM:

steve:

Bruce@16: you don't need an expensive red contact lens, any more than Von Daniken(?)'s cavemen would have needed access to an x-ray source to illustrate skeletons. Red cellophane or a piece of red lighting gel should do the trick.

It has to be a situation where only red-shifted light is reaching the eye. Red cellophane or lighting gel wouldn't do it unless they were taped over the socket to prevent light leak: you'd need to use a set of goggles with one red lens. (Hey! I should run this by Kaja! Just think--a special edition of Girl Genius goggles...) Something like the goggles they have for people who only have rods, not cones. (And the report from them is that rods are color sensitive, thank you.)

#145 ::: NelC ::: (view all by) ::: December 06, 2007, 10:30 AM:

Bruce @144: I think I detect a possible Girl Genius/Transmetropolitan crossover, featuring Spider Heterodyne. Or possibly Agatha Jerusalem.

#146 ::: Carrie S. ::: (view all by) ::: December 06, 2007, 10:44 AM:

The problem I have with "rewiring" the brain to acquire skills instantaneously is that the human brain isn't designed to do that. Unless you can directly manipulate neurons via magic nanomachines or extremely localised magnetic fields or the like, learning probably has to occur at meatspace speeds.

Consider, however, that in the Matrix Trinity doesn't have any physical neurons; her digital-representations-of-neurons can be manipulated in any way the Neb. has programs for. We have no information (in the first movie; I haven't see the other two) on whether her Matrix-abilities translate into the real world, and in fact a far amount of implication that they don't.

#147 ::: Peter Erwin ::: (view all by) ::: December 06, 2007, 12:16 PM:

NelC,

Those are good points -- and the idea that they're sucking specialized knowledge out of Matrix natives would explain how they get these skill sets in the first place (it's not like anyone is flying helicopters around in Zion...).

What I was getting at was more what the movie seemed to portray the process as, rather than what would be the most physically/neurologically plausible way of doing it.

#148 ::: ethan ::: (view all by) ::: December 06, 2007, 12:47 PM:

NelC #142: We see here that fanfic is sometimes far better than its source material.

#149 ::: NelC ::: (view all by) ::: December 06, 2007, 01:25 PM:

Carrie @146: I don't believe that anybody's brains are being simulated in the Matrix, but instead the sensations from the simulated bodies are being played into the sensory nerves of their real bodies. Otherwise there would be no need to keep the meat around. Also, note that simulating their brains would require dumping the simulated memory into the real brain whenever they were disconnected from the Matrix, to maintain continuity. Maybe not a problem if the bodies are never disconnected, but without building a memory dump into the system it would leave escapees with no memory whatsoever when they left the Matrix, and Trinity, Neo and Morpheus would retain no knowledge of their missions there, which would rather muck up the plots of the movies.

#150 ::: Carrie S. ::: (view all by) ::: December 06, 2007, 01:53 PM:

I don't believe that anybody's brains are being simulated in the Matrix, but instead the sensations from the simulated bodies are being played into the sensory nerves of their real bodies. Otherwise there would be no need to keep the meat around. Also, note that simulating their brains would require dumping the simulated memory into the real brain whenever they were disconnected from the Matrix, to maintain continuity.

Their brains have to be simulated, because they can be acted upon in the Matrix. You can shoot a gun at someone's head and it explodes, complete with brain matter. As for the downloading, who says it's not a continual process?

My mental picture of how the Matrix works is like this: Trinity has a meat body, with a meat brain, that generates her mind. When she plugs into the Matrix, it generates for her a Matrix body that is her mental picture of what she looks like and how her meat body feels and all that, and her mind takes control of that body through the conduit of the plug--rather like the soldierboys in The Forever Peace. The conduit, intercepts all impulses for moving or whatever* and transfers them to the Matrix body instead of the meat body.

So when she needs to know how to fly a helicopter, the computer back in the real world takes a program that knows how to do that (I like the idea that it gets that program from someone who's a pilot by training, that's awesome) and adds it to the simulation of her Matrix body. Since her mind is working with the Matrix body, she can now use that knowledge as if it were real, but when she gets unplugs she won't be able to use it anymore (except possibly in the very gross sense of "well, I remember that pushing forward on the stick made us go up" or whatever).

Trinity's mind is still in her meat body, it's just that all her impulses are being transferred to her simulated Matrix body. So she's still forming memories in her own neurons. Would've been much cleverer for the machines to come up with a system where the mind actually moved into the Matrix body, because then there'd be no rebellion problem at all, just as you said.

*: One hopes autonomic impulses stay in the meat body, or maybe the Neb's computers handle all that or something.

#151 ::: Marilee ::: (view all by) ::: December 06, 2007, 10:21 PM:

NelC, #143, I don't know about the Matrix, but it only took my brain six weeks to rewire enough for me to wake up out of a coma. I didn't get nouns back for another six weeks, and then had to learn to read and walk again, but that's pretty good compared to the original 32 years.

#152 ::: David Bain ::: (view all by) ::: December 08, 2007, 05:18 AM:

I disagree with Avram that the colour blind synaesthete upsets the Mary argument against physicalism, though I think other replies do. The Mary argument is roughly this: given you cannot know what it is like to experience a given colour until you do, no matter how much you are told about the physical universe, then what it is like to experience a colour must be a non-physical fact about the universe. The plausibility of the premise that, without having done so, you cannot know what it is like to experience a given colour no matter how much you
are told about the physical universe is not undermined, as far as I can tell, by the possibility of people suffering from any or all of the following conditions: (i) when seeing an object x, they experience a colour x does not have; (ii) when seeing an object x, they experience a more coarsely individuated colour than normal humans would when seeing
x; (iii) when hearing a sound, they are caused to experience some colour. (Some, by the way, think colour blindness involves (i); I think
it involves (ii). (iii) is the synaesthesia case.)

Avram's reply seems to be to quote a neuroscientist who says, probably
rightly, that the casual mechanisms behind synaethesia involve fairly central brain processing. But, though I agree, I don't see how this cuts any ice at all regarding Mary.

Welcome to Making Light's comment section. The moderators are Avram Grumer, Teresa & Patrick Nielsen Hayden, and Abi Sutherland. Abi is the moderator most frequently onsite. She's also the kindest. Teresa is the theoretician. Are you feeling lucky?

Comments containing more than seven URLs will be held for approval. If you want to comment on a thread that's been closed, please post to the most recent "Open Thread" discussion.

You can subscribe (via RSS) to this particular comment thread. (If this option is baffling, here's a quick introduction.)

Post a comment.
(Real e-mail addresses and URLs only, please.)

HTML Tags:
<strong>Strong</strong> = Strong
<em>Emphasized</em> = Emphasized
<a href="http://www.url.com">Linked text</a> = Linked text

Spelling reference:
Tolkien. Minuscule. Gandhi. Millennium. Delany. Embarrassment. Publishers Weekly. Occurrence. Asimov. Weird. Connoisseur. Accommodate. Hierarchy. Deity. Etiquette. Pharaoh. Teresa. Its. Macdonald. Nielsen Hayden. It's. Fluorosphere. Barack. More here.















(You must preview before posting.)

Dire legal notice
Making Light copyright 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017 by Patrick & Teresa Nielsen Hayden. All rights reserved.