Ain't no spoon or fork or knife.

April 19, 2012 5:58 PM

“There are no images and no representations in our minds,” he insisted. “Our visual experience of the world is a continuum between see-er and seen united in a shared process of seeing.”

I was curious, if only because, as a novelist I’d always supposed I was dealing in images, imagery. This stuff might have implications. So we had a beer together.
posted by Brandon Blatcher (25 comments total)

SOCRATES: Then attend, and I will try to finish the story. The purport is that all these things are in motion, as I was saying, and that this motion is of two kinds, a slower and a quicker; and the slower elements have their motions in the same place and with reference to things near them, and so they beget; but what is begotten is swifter, for it is carried to fro, and moves from place to place. Apply this to sense:—When the eye and the appropriate object meet together and give birth to whiteness and the sensation connatural with it, which could not have been given by either of them going elsewhere, then, while the sight is flowing from the eye, whiteness proceeds from the object which combines in producing the colour; and so the eye is fulfilled with sight, and really sees, and becomes, not sight, but a seeing eye; and the object which combined to form the colour is fulfilled with whiteness, and becomes not whiteness but a white thing, whether wood or stone or whatever the object may be which happens to be coloured white. And this is true of all sensible objects, hard, warm, and the like, which are similarly to be regarded, as I was saying before, not as having any absolute existence, but as being all of them of whatever kind generated by motion in their intercourse with one another; for of the agent and patient, as existing in separation, no trustworthy conception, as they say, can be formed, for the agent has no existence until united with the patient, and the patient has no existence until united with the agent; and that which by uniting with something becomes an agent, by meeting with some other thing is converted into a patient. And from all these considerations, as I said at first, there arises a general reflection, that there is no one self-existent thing, but everything is becoming and in relation; and being must be altogether abolished, although from habit and ignorance we are compelled even in this discussion to retain the use of the term. But great philosophers tell us that we are not to allow either the word 'something,' or 'belonging to something,' or 'to me,' or 'this,' or 'that,' or any other detaining name to be used, in the language of nature all things are being created and destroyed, coming into being and passing into new forms; nor can any name fix or detain them; he who attempts to fix them is easily refuted. And this should be the way of speaking, not only of particulars but of aggregates; such aggregates as are expressed in the word 'man,' or 'stone,' or any name of an animal or of a class. O Theaetetus, are not these speculations sweet as honey? And do you not like the taste of them in the mouth?
posted by phrontist at 6:05 PM on April 19, 2012 [4 favorites]

I agree with the commenters on the article who feel that his views seem to strongly echo existential phenomenology (e.g., Merleau-Ponty). I also swear that the only way I can begin to wrap my mind around any aspect of Hegel's Phenomenology of Spirit is to posit that Hegel really is claiming that my consciousness of this or that object is first and foremost the object's consciousness of itself -- a notion (however bizarre on the surface) which also seems somewhat at home here. Neat stuff, but definitely not as radically new as the author seems to suppose.
posted by treepour at 6:50 PM on April 19, 2012 [1 favorite]

For the rainbow experience to happen we need sunshine, raindrops, and a spectator.
posted by jepler at 6:51 PM on April 19, 2012

Sounds like Zen to me.
posted by cmoj at 6:55 PM on April 19, 2012

…which is to say, how does this line of argument stand up to the apparent fact that a suitably-driven display can create something substantially like a "rainbow experience", and a suitably driven speaker can create something even more substantially like a "ringing bell" experience, while sharing basically no physical structure with an actual bell or rainbow? Except to the extent that the structure of the "external" part is actually perceived, it doesn't matter. And we'll enjoy soggy old potato chips, it turns out, if our chewing is accompanied (over head phones) by the satisfying sound of crunching.
posted by jepler at 7:05 PM on April 19, 2012

It's interesting that his theories came about from building robots. When I read Merleau-Ponty and Heidegger, I thought it would be interesting to build an existential robot. Robotics and AI is based on a rather behaviorist model of psychology. Input comes from a set of sensors, it is processed in a computer program, and behavior is output. The program is tweaked based on past stimuli, and so the robot "learns", much in the way that a lab rat learns to press a lever for food. Robots seem like an idealized model of rational stimulus response, but that model makes them resemble insects or rats more than people.

Maybe what robotics is missing is the flimsier, more nebulous aspects of human consciousness. Could we program a robot to know it's going to die, to know that its individual existence is limited and will come to an end someday? Could we make it yearn for something that it can't get, and to know it can't have what it wants? In other words, is it possible to code the human condition?

Manzotti has a good point about perception. When I look at a tree, it seems reasonable to suppose that the brain isn't simply building a digital representation of a tree encoded in neurons. That's not what the experience of seeing a tree is like. For one thing, I can't actually focus on the whole object. My vision is fixed at one point, with a large radius around it where I can see foggy shapes of other objects within my visual range. I see that one point vividly, but I have to really concentrate to build an image of it in my head, and it's much harder to do that for a large object, like a tree, because doing so I'd have to rely on fuzzy peripheral vision. In normal perception, I don't think I see images at all. Rather, it's like my brain registers an assertion – "there's a tree a few feet in front of you now" – and cognition moves on with more pressing matters.

I'm sure that my experience isn't universal. Surely there are artists and photographers who see crisp, vivid landscapes in their heads; maybe their brains are building images. It's worth thinking about, if we want to build machines that act like people. (What kinds of people?)

I think I agree with Manzotti. Maybe this issue of encoding is a red herring. Human brains aren't computers. It seems like many theories in cognitive science and AI are attempts to apply computational models to the brain – computational models which were created by, and inspired by, the workings of the brain.

The amazing thing about the brain is that I, as an owner of one, can have a concept of something I have no first hand experience with, like Lagos, Nigeria. I have never been to Lagos. I will probably never go to Lagos, or meet anybody from Lagos. But there's a little model of Lagos in my head. Manzotti seems to be saying that somehow, the Lagos model in my brain is strangely related to the real Lagos, Nigeria, even though my ideas about the place are probably misinformed or downright wrong.

This gets even trickier when you think about fictional things. I read Tolkien as a kid, so I definitely have a model of Middle-Earth in my brain. But Middle-Earth doesn't exist. At all. It is not the case that there is a place called Middle-Earth, but it's model is in my brain. I suppose Manzotti might say that's my brain is communicating with Tolkien's brain across space and time, which, however romantic, seems suspect.

What if I make up my own fictional world? With little fictional people, who have fictional brains of their own, and put little fictional thoughts and ideas in their brains? What's the ontological status of my sense-perceptions about the fictional thoughts I just made up? Would Manzotti say that I can't imagine a non-existent thing, that everything I think of is somehow a reflection of the real world? If so, where do truly new ideas come from?

Hell, I can even have a totally wrong conception about my own brain. The human capacity for deception is vast.
posted by deathpanels at 8:39 PM on April 19, 2012 [3 favorites]

Having worked in the field (or tangential fields -- I've never built a robot), my (current) personal theory is that consciousness doesn't exist. That it's an emergent property of perception, memory and synthesis. The first and second are obvious, the third being the ability to combine records stored in the second to produce new records.

A lot of this depends on where you set the threshold for consciousness. A lot of people define consciousness as having some kind of self-awareness. But I'm not sure where the threshold is. A navigating robot is self-aware in so much that it locate itself in space. We ourselves lack self-awareness past a certain point in so much we can never truly know how others perceive us. Where do you draw the line?
posted by smidgen at 8:49 PM on April 19, 2012

A lot of this depends on where you set the threshold for consciousness. A lot of people define consciousness as having some kind of self-awareness. But I'm not sure where the threshold is. A navigating robot is self-aware in so much that it locate itself in space. We ourselves lack self-awareness past a certain point in so much we can never truly know how others perceive us. Where do you draw the line?

Which, I think, is the genius of the Turing test. It lets you skirt around the whole messy matter of self-awareness and puts a black-box around consciousness.
posted by deathpanels at 9:36 PM on April 19, 2012 [1 favorite]

When will we get rid of the silly, hubristic delusion that consciousness can be reduced to matter?
posted by shivohum at 9:47 PM on April 19, 2012 [2 favorites]

shivohum: "When will we get rid of the silly, hubristic delusion that consciousness can be reduced to matter"

When will we get rid of the silly, hubristic delusion that consciousness can't be reduced to matter?

Reductionist materialism has one hell of a better track record than non-reductionist theories like vitalism.
posted by Proofs and Refutations at 10:55 PM on April 19, 2012 [8 favorites]

Reductionist materialism has one hell of a better track record than non-reductionist theories like vitalism.

I would say it has exactly no track record in showing how qualities of conscious experience can ever come from matter. None. It shows correlations, nothing more.
posted by shivohum at 11:03 PM on April 19, 2012

shivohum: "I would say it has exactly no track record in showing how qualities of conscious experience can ever come from matter. None. It shows correlations, nothing more"

We have evidence that damage to specific regions of the brain prevents conscious vision but still allows a person to navigate a room visually. It's called blindsight in the literature. That's a direct causal link, if it's there you can consciously see, after it's damaged you can't. If you want to rule out that as evidence it seems to me you have to rule out the very possibility of evidence (although I'm happy to be wrong on this point, feel free to describe what would count as evidence that you're wrong), at which point you've got no basis for accusing others of hubris.
posted by Proofs and Refutations at 11:18 PM on April 19, 2012

Yeah, blindsight's not evidence of how consciousness comes from matter. It shows how intelligent processes may become inaccessible to self-awareness; it does not show how qualia arise from matter.

Though apart from that, even the possibility of scientific evidence is in fact ruled out. It's really quite simple.

I find the inverted spectrum arguments really clinch it. Suppose that where you see the color blue I see the color green and vice-versa, but we call the same things by the same names (we both call the sky "blue" and grass "green").

How would we ever tell this was or was not the case? We couldn't. Therefore the perception of color is not reducible to matter. That breaks materialism. QED.
posted by shivohum at 11:31 PM on April 19, 2012

Therefore the perception of color is not reducible to matter.

What are we going to build observers out of then?

I don't see how the fact that we can look at the same thing and have different conscious experiences "breaks materialism". After all, I don't have your brain and you don't have mine. Sure, the mapping from stimulus to conscious perception may be different, but we are different observers with different eyes and different brains. How does the fact that our conscious experience may differ rule out a materialistic explanation of how it arose?
posted by Avelwood at 12:39 AM on April 20, 2012

A consistent materialist should not admit the possibility of an inverted experience of colour (without physical differences), just as they shouldn't admit the possibility of a philosophical zombie.

I think that since we can (sort-of) reproduce colours in our imagination, we have to reduce colours to structures that are literally reproduced in the brain. The fact that brain surgeons don't see colours when they look in a brain may just entail that they aren't seeing the brain from the right perspective (don't ask me what the right perspective is though).
posted by leibniz at 2:07 AM on April 20, 2012

The trouble with non-materialistic theories of mind is that they don't exist. You get as far as 'materialism cannot explain THIS!' (as noted, a common argument through the ages) and then run into the buffers. It's a position, certainly, but it's not usable.

On the other side? Many decades of observation and experimenting on ourselves and other animals show we've been peeling away the onion with some relish. While there is no neat metaphor for how consciousness works, in the way that the equations for a photon are a pretty good metaphor for what a photon 'actually is' (sorry for the scare quotes, those two words are a placeholder), we haven't found any aspect of it that isn't affected by material changes. Even if the non-materialists are right, they can't show where consciousness isn't very, very tightly bound to material phenomena - and experience shows us that it makes sense to consider a system tightly bound to a mechanism to be caused by that mechanism.

If you want to cite external variables, factors or systems, then what or where might they be? You have to have something to say there - otherwise, it's not a theory, it's a wish powered by incredulity. The history of science - hell, history full stop - is full of those, so there's a lot of heritage for that approach, but their track record is abysmal.
posted by Devonian at 2:44 AM on April 20, 2012 [1 favorite]

shivohum: "I find the inverted spectrum arguments really clinch it. Suppose that where you see the color blue I see the color green and vice-versa, but we call the same things by the same names (we both call the sky "blue" and grass "green").

How would we ever tell this was or was not the case? We couldn't. Therefore the perception of color is not reducible to matter. That breaks materialism. QED"

That's just begging the question. If you take the irreducibility of conscious experience as a premise it's neither surprising nor compelling that you come to the conclusion that materialism is false.
posted by Proofs and Refutations at 3:57 AM on April 20, 2012

deathpanels, and others interested in the synthesis of existential thought and artificial intelligence --- I enjoyed Hubert Dreyfus's paper Why Heideggerian AI failed and how fixing it would require making it more Heideggerian.
posted by mbrock at 4:27 AM on April 20, 2012

The trouble with non-materialistic theories of mind is that they don't exist. You get as far as 'materialism cannot explain THIS!' (as noted, a common argument through the ages) and then run into the buffers. It's a position, certainly, but it's not usable.

I soo disagree. It all emergent properties. Energy->matter->autopoesis->cognition->consciousness.

Matter is necessary, but it is all in the motion, interaction and complexity. And it is not reducible to it's parts.

Someday we will have machines with temporal and spatial resolution to observe consciousness directly, but qualia require participation and from the outside we will only be able to see the firing of billions of neurons, the changes in concentrations of various chemicals and the formation and destruction of various neural pathways. It is the combination and interaction of all this and more that is the experience of consciousness as we know it.
posted by psycho-alchemy at 4:34 AM on April 20, 2012

What are we going to build observers out of then?

Thought, clearly. That's what the universe is made of.

I don't see how the fact that we can look at the same thing and have different conscious experiences "breaks materialism".

It's not the fact that we have different conscious experiences but the fact that we can never, ever access each other's conscious experiences directly that's the problem.
---
A consistent materialist should not admit the possibility of an inverted experience of colour (without physical differences), just as they shouldn't admit the possibility of a philosophical zombie.

Oh, they can decline to admit what they like, but there's no proof. You don't even have to go so far as to say that inverted spectra are actually possible. All that's necessary to show is that we can't know if they're possible. Ever.

If we can't know something that might or might not be the case, there exists a realm permanently outside of material knowledge. That's what breaks materialism, because the whole point of materialism is that everything's in theory accessible to the senses.
---
It's a position, certainly, but it's not usable.

You mean it's not usable for positivistic science. It's perfectly usable for art. It's perfectly usable as a philosophy. Heck, even it's even usable for a completely individualistic, subjective science, one where each individual experiments on themselves in their own private domain of experience.

Even if the non-materialists are right, they can't show where consciousness isn't very, very tightly bound to material phenomena - and experience shows us that it makes sense to consider a system tightly bound to a mechanism to be caused by that mechanism.

Well that's certainly not true. Why not equally well say that a system that's tightly bound to a mechanism causes that mechanism?

Moreover, consciousness has in fact been tied only at an extremely blocky level to matter. Look at a poet. Then look at your typical fMRI study. There's a mindblowing difference in the specificity of the quality of consciousness predicted. You're not getting there from here, because poetry shows how individual experiences are, how unscientificalizable in the conventional sense they are.
---
If you take the irreducibility of conscious experience as a premise it's neither surprising nor compelling that you come to the conclusion that materialism is false.

I don't have to take it as a premise (though it is clearly true). I merely have to show that we can never know. And how can we? Anything we see happens within our field of qualia, and we cannot be sure about our own color inversion.

Do you have a compelling argument for how you could ever, even in theory, know whether someone had different qualia than yourself? I find only purely religious arguments that "science will somehow solve it because science solves everything," even though science has not in fact solved everything.

Science, for example, has nothing directly to say about what we ought to do; it cannot directly dictate the axioms of ethics. There is no reason to believe it ever, ever will have anything to say about it, either. It's simply a different category, untouchable. This is no different.
posted by shivohum at 6:03 AM on April 20, 2012 [3 favorites]

Why does blindsight prove anything other than dissimulation on the part of the blindsighted subject?

1. they demonstrate the faculty of sight by successful navigation
2. they deny having the faculty of sight
3. they fail to complete certain other tasks requiring the use of sight

Moving from these observations to the conclusion that the person is or is not conscious of something seems impossible. Likewise, any sort of behavioral-empirical approach to a neuron, a neural network, or an entire brain will have to similarly side-step the question of consciousness (in the interest of careful reasoning), because consciousness isn't something you can observe as an empirical object.

As Karl Jaspers put it:

If, for example a materialist explains the external world as a creation of our physiological organization, more particularly our brain, still the brain, including his own, is a part of the external world which can be observed under local anesthesia, with a trepanation and mirrors. Thus the brain becomes a creation of the brain: formally the same mode of thought which describes God as causa sui.

So how does one "get out" of consciousness and look at it, study it, figure it out?

Jaspers talks about 'The Encompassing' (Das Umgreifende) which is a kind of horizon of our experience, serving as an absolute context, but which, like the horizon, recedes as we approach it.

Any observation capable of changing implies, or has as a necessary prerequisite, something not observed, without which change would be impossible, as all possibilities would already be present in observation.

It follows from that distinction between the observed and the not-observed that all science takes place on one side of that divide: the observed side.

Because our observation is delimited (or umgegriffen) by that which we do not observe, all objects of our consideration take place in a larger context of our whole present observation.

If one equates this whole present observation with consciousness, then it comes as no surprise that it constantly eludes definition and scientific study. It is necessarily outside our consideration, as it is the context and foundation of all consideration.

This seems like it would apply to all finite conscious beings, hypothetical AI included.
posted by edguardo at 9:01 AM on April 20, 2012 [1 favorite]

Wiki on Quantum mind–body problem: Parallels between quantum mechanics and mind/body dualism were first drawn by the founders of quantum mechanics including Erwin Schrödinger,[1] Werner Heisenberg,[2] Wolfgang Pauli,[3] Niels Bohr,[4] and Eugene Wigner[5]

Aren't there examples of biological uses of quantum effects? In, like, narwhal tusk environment sensing or something? It wouldn't surprise me if it comes down to physical processes that we don't understand.
posted by StickyCarpet at 10:49 AM on April 20, 2012

Instead of an essential definition of consciousness, which is what a lot of people seem to be grappling with here, it looks like he's taking strict operational definitions and applying them to consciousness. It worked for Skinner when he was coming up with his stuff for behavior and decided to "black box" the mind process. Right up until Bandura came along with the Bobo doll. It's fine as a starting point, but as a scientist I would hope he doesn't think that's an end point.
posted by Rocket Surgeon at 8:32 PM on April 20, 2012

It's not the fact that we have different conscious experiences but the fact that we can never, ever access each other's conscious experiences directly that's the problem.

Why not? If the materialist hypothesis is correct, then in theory I might be able to access another person's conscious experiences by connecting their consciousness-producing apparatus to mine. Obviously this would require science-fictional technology, but for a materialist it's not ruled out. I imagine the result wouldn't be "seeing into another person's qualia"; more likely, since their consciousness-producing apparatus and mine are now part of a single system, our formerly-distinct consciousnesses would merge together and there would be a single "I" (or no integrated self at all) experiencing a single set of qualia.

Moreover, consciousness has in fact been tied only at an extremely blocky level to matter. Look at a poet. Then look at your typical fMRI study. There's a mindblowing difference in the specificity of the quality of consciousness predicted. You're not getting there from here, because poetry shows how individual experiences are, how unscientificalizable in the conventional sense they are.

Your assertion that we're "not getting there from here" doesn't follow from that distinction. The fact that our present understanding is coarse-grained doesn't mean it can't ever be fine-grained; it just means we're still refining our tools.

Anyway, I think poetry is a poor example for the individuality of experience. A poem is a communicative act, a way of sharing information between consciousnesses. If anything, the fact that I can understand your poem indicates that our experiences aren't uniquely, irreducibly individual. I can look at faces in a crowd and see something there that is like petals on a wet, black bough, just like Ezra Pound said. A materialist could even investigate how different brains respond to the image.
posted by twirlip at 2:33 PM on April 21, 2012 [1 favorite]

Why is an emergent property of a material process non-materialist? Isnt that just splitting hairs? Isnt the precise refutation of mystical non-materialist thought that, yes, if you mangle the brain, what we think of as human conciousness will disappear? Youd have to be willfully blind to doubt that. Ithink it's well estblished, that yes, the nervous system is the substrate upon which human conciousness exists.

It's like saying a bell curve is not materialist. I can take a pile of marbles and a pegboard and generate one, and I can even follow each marble's path if I want. However, In order to truly understand whatis going on, I have to analyze the machine in aggregate. This doesnt make it magic, it just makes it an emergent property of lots of small yes/no decisions.

Like I hinted earlier, I think we eventually have to accept that there are levels of conciousness, that we exist of a contiuum defined not only by fairly strict dimensions, but those criteria will have a direct physical analogue .... just like a larger pegboard and more marbles will give you a smoother curve...
posted by smidgen at 11:03 AM on April 22, 2012 [1 favorite]