Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

Jack comments on How to think like a quantum monadologist - Less Wrong

-14 Post author: Mitchell_Porter 15 October 2009 09:37AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (266)

You are viewing a single comment's thread. Show more comments above.

Comment author: Mitchell_Porter 19 October 2009 11:58:33AM -2 points [-]

We can skip the details of the story and come straight to the point. Let us suppose that what I see is indeed a two-dimensional data structure in my brain. It has pixels and they are neurally encoded somehow, e.g. by spiking rates.

Now suppose I'm seeing something colored - a blue sky, a red apple, anything. By our hypothesis, what actually exists is nothing but a two-dimensional sheet of neurons all firing in different rhythms - ions moving across membranes, and so forth.

Where is the blueness or the redness, if this and only this is the reality?

It seems to me you have two choices. Either blueness is nothing but ions plunging back and forth across a membrane with a certain regularity; or, there is no such thing as blueness - only talk of blueness, neural dispositions to categorize as blue, and so on.

I think that what people usually imagine is that the ions-crossing-the-membrane-at-a-certain-rate "is" the blueness, but they do this by mentally juxtaposing the physical picture (if they think it through that far) with the blueness that they actually see and experience. But in that case they've gone beyond the nothing-but-the-atoms provision. Or, the troublesome color-word will be buried in a larger phrase, and so all those neural firings are identified with "seeing blue" or "the experience of seeing blue". But I don't see how that solves anything, although psychologically it has the effect of directing your attention away from the blueness itself, towards the more abstract states in which it features. And being more abstract, perhaps it is easier - perhaps it is subjectively more plausible - to imagine that they are nothing but neural computations. However, that's just a trick that you play on yourself.

So I bite the bullet and say, the blueness is obviously there, somewhere in reality; it is obviously not there in a physics which consists of nothing but point masses moving back and forth, or any of the other, slightly more complex physical ontologies on offer; so, I had better seek a perspective on physics in which there is a place where it might be. This is the point of a monadic interpretation of entanglement. I don't say it's the only way to do it, but it is a way to create the necessary room.

Comment author: Jack 19 October 2009 12:33:02PM 5 points [-]

Or, the troublesome color-word will be buried in a larger phrase, and so all those neural firings are identified with "seeing blue" or "the experience of seeing blue".

This is a linguistic fact, not a phenomenological one. Our language happens to distinguish verbs by referring to their objects, subjects and using adverbial modifiers. But the language could have just as easily had a one-word verb phrase than means "experience blueness". Say this word was "bluep". If this were the case we wouldn't be asking where the object of blueping was. Rather, we'd see blueping as fundamental and would easily identify blueping with a particular configuration of neural firings.

Since blueness is a phenomenological quality I can't imagine finding it anywhere except as an object modifier of experience. I don't see how a monadic interpretation of entanglement changes that fact-- you're just associating the configurations (or whatever) of monads with the subjective experience of blueness. Blueness itself is meaningless.

Comment author: Amanojack 12 March 2010 11:26:39PM *  1 point [-]

Rather, we'd see blueping as fundamental and would easily identify blueping with a particular configuration of neural firings.

Yes, I'd say blueping is one of your "fundamental concepts that can't be dissolved." But to me it's no surprise that sensations of the five senses are the fundamental units of experience. The only reason I think people posit that there is more to experience than mere sensation is that they say, "What about thinking and emotions? Those are experience but not really sensations of the five senses."

My theory - developed for entirely separate reasons - is that all thinking is done in the (imagined) five senses, but that we don't notice because it's usually happening too fast, or is auto-ignored because the background sensory processes aren't generally relevant and would overwhelm our conscious mind. The sensory thought processes can be noticed in some special situations, though, which is how I found out about them. (In fact, I'm heading to an isolation tank this afternoon to try to "see" more of my own thoughts.) My theory says that emotions are primarily (or perhaps only) imagined physical sensations. I hope to write a main post on the theory once it's more developed.

Anyway, that would indicate that all experience is sensation, hence it would be natural to consider blueping as fundamental (or if you like, undissolvable).

Comment author: jimrandomh 13 March 2010 12:22:40AM 2 points [-]

My theory - developed for entirely separate reasons - is that all thinking is done in the (imagined) five senses, but that we don't notice because it's usually happening too fast, or is auto-ignored because the background sensory processes aren't generally relevant and would overwhelm our conscious mind.

Be warned: thoughts have properties and connections that appear only when summoned. For example, suppose I take a sentence or two from my inner monologue and try to analyze the voice. I will imagine hearing it in the voice of a person or character who might say that sort of thing. But the thought wasn't originally in that voice, or any voice at all; trying to inspect details like tone and inflection caused my mind to create those details, more or less at random, where they would not have been created if I hadn't gone looking for them. I can connect any thought to one of the senses in this way, but that doesn't mean that the connection was there before I summoned it. Words are imported through hearing, but there's no reason the brain has to maintain that connection.

On the other hand, minds vary. People with synesthesia definitely have stronger connections between their senses and other thoughts than normal. Someone with a slightly more active auditory cortex might hear their inner monologue in a specific voice even if they weren't trying to. There're many possibilities, and spending some time in a tank or meditative state to figure out how your mind works is a great idea.

Comment author: Amanojack 13 March 2010 02:01:02AM 0 points [-]

I believe you're cautioning that maybe what I'm noticing are sensory patterns my mind attaches to the thoughts after the fact, when I go in and try to analyze them, rather than the thoughts themselves. I actually had a few false starts this way, but later on I found what I think are my internal representations of the structure of logical reasoning. They are like fundamental thought-widgets (visual and physical, àla Einstein's claim) that fit together with a certain set of rules to form meaning. These could be after-the-fact sensory patterns as well, but since they are mostly visual I've started using them to create a picture-only language as a sort of proof-of-concept. I'll share more as it develops.

Also, I try not to think in words. Habitual word thinkers and non-word thinkers will probably have very different reactions to what I've been writing.

I grant that I might just have a particularly strong case of synesthesia. Hopefully the tank will bring answers.

Comment author: Mitchell_Porter 19 October 2009 12:44:20PM -1 points [-]

the language could have just as easily had a one-word verb phrase than means "experience blueness".

Are you sure about that? Do you think the meaning of "bluep" could be conveyed to young humans without having blue objects to point at and without those humans at least forming a concept of blue? I also doubt that this would make physicalism any easier subjectively. Whether it's the experience or the object of the experience which is regarded as blue, something's blue.

I have to go offline now, right in the middle of some heated real-time exchanges. Don't anyone get too steamed if you don't hear back from me for a day or so. :-)

Comment author: Jack 19 October 2009 02:09:36PM *  2 points [-]

Are you sure about that? Do you think the meaning of "bluep" could be conveyed to young humans without having blue objects to point at and without those humans at least forming a concept of blue?

You're conflating two things. There is the property some objects have of emitting photons with a wavelength of about 475 nanometers. Then there is the phenomenology of seeing such things which is "blueing" or experiencing blue. Teaching children the word blue just involves teaching them how to describe the reflective properties of certain objects.

Now people tend to divide their experiences along similar lines. Thus, the set of all blue objects is that same for most people. We don't disagree that there is a difference between red things and blue things, unless we're color blind. This lets us share a vocabulary of qualia but we have no reason to think that what the words actually refer to are the same for everyone except insofar as we find correspondence between the physical configuration of the brain and reports of subjective experience.

I also doubt that this would make physicalism any easier subjectively. Whether it's the experience or the object of the experience which is regarded as blue, something's blue.

No. Experience isn't blue... not at all in the same way objects are. This is a HUGE category mistake. The verification conditions for claims about "blue experience" are TOTALLY DIFFERENT then verification conditions for claims about blue objects. We might as well have different words.

Comment author: Mitchell_Porter 20 October 2009 07:27:43AM -1 points [-]

So, if I understand you correctly, the world is as follows: There are objects that emit "blue" light. And there are nervous systems which respond differently to blue and non-blue light. And some aspect of this differential response is "experiencing blue".

It seems to me that this functions as a way to avoid directly mentioning the problematic entity, i.e. shades of blue. There is a concession that, yes, objects in the external world aren't actually blue. One might suppose, then, that the thing which is actually blue is somewhere in the brain. But instead, by talking about "experiencing blue" as a unit, we get to focus on language ("how to describe the reflective properties of certain objects", "vocabulary of qualia", "verification conditions for claims about 'blue experience'", "reports of subjective experience"), cause and effect, information processing, anything but phenomenal blueness itself.

Comment author: Jack 20 October 2009 04:16:16PM *  2 points [-]

I'm fine with saying, "There are qualia. They are part of our experience and we need to account for them." The way to do that is to find the brain-state that corresponds to the experience of seeing blue and then (with other information about other brain-states) posit rules that relate subjective experience with brain-states. If we develop computers that report qualia then we can do the same and then generalize the theories and come up with a universal theory of qualia. We want to get to a point where we can say: P1, P2, P3, P4... (which are any set of empirical descriptions of a brain or computer) AND L1, L2, L3...(which are our laws) and then output S1, S2, S3 (descriptions of subjective experience). Once we can do this both ways I would take us to understand qualia and phenomenal experience.

But you keep using language which makes it sound like you're looking for real "blueness" or something. But there is no such thing as blueness except as a way of distinguishing a certain kind of sight experience. I'm not denying that the phenomenal experience of it isn't a real experience just that "experience being blue" is anything like jeans being blue. Jeans are blue in virtue of the fact that when a subject looks at them they experience blueness. So how could that kind of blueness be found within experience? Whatever experiencing-blue is it is likely very different from the property objects have of being blue.

This is all to deny the motivation of looking past neuron firings to find qualia.

Comment author: Mitchell_Porter 22 October 2009 08:08:29AM -1 points [-]

you keep using language which makes it sound like you're looking for real "blueness" or something.

Of course I am. I emphasize again that the really undeniable reality (though you are apparently denying it) is the individual shade of color. Color names like blue are fuzzy in scope. But the concrete instances of color which they are intended to categorize very definitely exist.

Your second paragraph is a fascinating exercise in constructing a way to keep colors out of the realm of the real. First, assert that there is no such thing as blueness, except something not actually blue which has the functional properties of blueness:

a way of distinguishing a certain kind of sight experience

Second, say that a thing is blue only if it has the property of causing the experience of blueness:

Jeans are blue in virtue of the fact that when a subject looks at them they experience blueness

Finally, observe that the cognitively relevant physical properties of the brain are very different from the reflectively relevant physical properties of surfaces, and triumphantly conclude that the blueness in the external world is nothing like the experienced blueness in the brain,

so how could that kind of blueness be found within experience?

The physicalist vision of a world made solely of quantity, space, and causality has a very strong grip on the imaginations of those who can wield the formalism. But arguments like this really are an exercise in denying reality. The physicalist ontology is a subset of the real ontology and you can see some of what's missing whenever you open your eyes. Just because you don't yet know how to think about it precisely is no excuse for denying that it's there.

Comment author: Jack 22 October 2009 05:20:54PM 0 points [-]

Look, an annotated repetition of my argument followed by

The physicalist vision of a world made solely of quantity, space, and causality has a very strong grip on the imaginations of those who can wield the formalism. But arguments like this really are an exercise in denying reality.

amounts to a kind of circumstantial ad hominem fallacy. You don't actually dispute any claim I make you just 'diagnosis' it. It is mildly annoying and throughly unhelpful. Are you really denying that there is a difference between blueness as a phenomenal quality and blueness understood as the reflective quality of an object? Even if you want to say that the experience of looking at the sky is "blue" experience do you actually hold that experience is blue in the same way that the sky is? Do think that experiences have reflective properties? Do the electrons in the atoms of experience drop out of higher energy levels and release photons of different wavelengths?

When we talk about any phenomenal entity, quality or event are we not talking about subjective experience? Isn't the definition phenomenology the study of things as we experience them and not the things in themselves? If so, when we talk about the phenomenon of blueness are we not talking about a kind of experience?

Anyway, I don't even understand how I'm the dogmatic physicalist in this discussion. I'm the one willing to posit fundamental laws that relate brain states to subjective experience. You're the one positing a physical entity with no empirical evidence, which somehow, through the handwaving magic of quantum physics is subjective experience. This is a big point: even if it is the case that a substantial part of subjective experience is accounted for by something other than neuron firings we will still need a separate set of laws to relate the brain state to subjective experience.

Comment author: Mitchell_Porter 23 October 2009 07:17:24AM *  0 points [-]

Are you really denying that there is a difference between blueness as a phenomenal quality and blueness understood as the reflective quality of an object? Even if you want to say that the experience of looking at the sky is "blue" experience do you actually hold that experience is blue in the same way that the sky is?

The appropriate use of the words has changed along with our ontology. In a mode of naive realism, in which appearances are not distinguished from their external causes, then the blueness of the sky is the blueness of the experience of the sky, because no distinction is being made between sky and experience of sky. However, once you get to the point of distinguishing between the experienced sky and the physical sky, then blueness in the original sense is only a property of the experienced sky, and the new "blueness" of reflective physics is only a property of the physical sky.

The problem now is that in the attempt to reduce experience to physics as well, the original sense of blueness is being banished entirely from discussion, solely in order to achieve the reduction. While it may be annoying to be lectured about how you're evading the question, you say outright

there's no such thing as blueness except as a way of distinguishing a certain kind of sight experience

which I take to be an explicit repudiation of the naive concept of blueness as applying to anything, physical sky or experienced sky. And you also said, two steps back,

This is all to deny the motivation of looking past neuron firings to find qualia

which suggests that you do understand this motivation, and are deliberately trying to route around it.

Comment author: RobinZ 20 October 2009 01:40:06PM 0 points [-]

What was your reaction to How An Algorithm Feels From Inside, incidentally?

Comment author: SilasBarta 19 October 2009 01:15:23PM *  2 points [-]

Are you sure about that? Do you think the meaning of "bluep" could be conveyed to young humans without having blue objects to point at and without those humans at least forming a concept of blue?

Who modded this up? Is this the standard now for what words a language could have? Whether the concept could be explained to a child? You might as well dismiss the word "oxygen" on the grounds that you can't explain the full chemical model to a child, in a way that allows you to make use of the concept.

In any case, you could explain it to a child: "When you're dreaming about blue, you're just blueping. But when you see blue for real, you're blueping and seeing something blue."

Why is it that "It's impossible to explain ..." always seems to say so much more about the speaker than the concept?