Hacker News Reader

Comment by bondarchuk | Hacker News Reader

If you are going to demand a precise, non-handwavey answer to the hard problem, then you must first give the hard problem in a precise, non-handwavey way, or at the very least prove in a precise, non-handwavey way that there even is such a thing as the hard problem. As far as I've seen nobody has really done this, it's always just "what breathes fire into the equations?" "what is it like to be a bat?" "I just feel it in my bones that there's something missing"... P-zombies are an attempt, except that "p-zombies are conceivable" is to be taken as axiomatic without any further explanation.

View on HN

I find the classical argument with color qualia very convincing (in fact, I thought about it before even knowing what qualia were...). Nothing you can find out about wavelengths, cones, rods, neurons and so on will tell you why you see red the specific way you do, and whether my perception of red corresponds to yours or my "red" is actually your "green". So there is a gap there, something that doesn't follow from the mechanical state of the system.

Of course, I won't claim to be able to define that gap in a precise way (I'm not even a philosopher), but I think it's clear that it exists, and in my view the burden on proof is on those that claim the problem doesn't exist because no one can come up with a believable methodology (even supposing infinite time, infinitely precise measuring instruments, etc.) that would tell us the answer to questions like why my "red" feels like it does or whether my experience of red is different from yours.

>Nothing you can find out about wavelengths, cones, rods, neurons and so on will tell you why you see red the specific way you do

Or maybe it will, but we just don't have the neurological and information-theoretical tools at this moment in time.

>and whether my perception of red corresponds to yours or my "red" is actually your "green".

And maybe (at least this is my personal guess right now) it will show that this question is incoherent.

I don't see why two people experiencing red differently causes a problem for consciousness. Can't we both be conscious even with very different qualia?

Of course we can be all conscious but with different qualia. But the thing is that if consciousness was merely a physical phenomenon, we should be able to determine qualia purely by looking at the physical system, just as we determine any other property. However, we could be in a situation where we have understood our behaviors down to particle level, and I would still have no idea how you perceive colors.

> However, we could be in a situation where we have understood our behaviors down to particle level, and I would still have no idea how you perceive colors.

I do not see how this is possible. If you can perfectly simulate me you can perfectly simulate any response I will give to any question about how I experience red. Moreover you will have better idea than I do because you could also simulate how my experiences of red will change with time as I gain more experience from seeing red or anything else really.

There is a very fruitful line in philosophy of mind and language that addresses the colour qualia question: anti-cartesianism and anti-skepticism.

Wittgenstein's arguments on seeing-as is the best example of this line. He argues, quite convincingly I think, that a lot of the kind of skepticism, of a cartesian sort, implicit in the colour qualia question arises out of conceptual confusions in language. The goal isn't to solve the debate, or directly refute it per se, but to dissolve the entire conceptual confusion upon which it relies by making clear how it is essentially a house-of-cards.

For instance, we know that humans have a relatively uniform perceptual landscape. Physiological realities in the eye limit the variations in human visual perception to a certain set wavelength band. With all the correct parts working correctly, we see and we all see in relatively the same way. We don't see our seeing. We don't perceive our perceptual experience from the inside out. We see. The variations in "what people see" are more in language, which is to say that people report seeing different things. As Wittgenstein would put it, we often "see-as" just as much as we "see-that", which are ways of "seeing" that intersect and mix in language.

Colour is interesting because in our language it is heavily dependent on ostensive definitions, which is the linguistic way of saying colour is reliant on pointing at examples. Often quite literally pointing with your hand, or directing someone to look at something in particular. So if there is a red car in front of you and both people say the car is "red" then they are seeing the same red because the colour is not the result of empirical validation of an internal private perceptual experience, but because the colour is in language. They are reporting the same colour.

Viewed in this line, the colour qualia question essentially dissolves away as irrelevant.

>the colour qualia question essentially dissolves away as irrelevant

And yet, I still wanna know.

The question is IMO as valid as "if I poke you with this pen, do you feel it different enough than what I feel if I poked myself?"

If we go through the exercise of breaking down the physiology, neurology and linguistic, I still wanna know if they feel it any different and how different.

> The question is IMO as valid as "if I poke you with this pen, do you feel it different enough than what I feel if I poked myself?"

Funnily enough, Wittgenstein also famously looked at pain as well [0] in the context of his famous private language argument, and he comes to the same conclusions as he does about perception and language.

It should be noted that Wittgenstein's central point is about the intelligibility of color, or perception, or pain, or a private language. He is not denying the existence of pain, or of what could be considered "first person priority", as a sensation. But he's commenting on the logical necessity of a public intelligibility for it. For him that intelligibility comes about through behavior or public criterions, that is, what we might call pain-behavior wrapped up in a public form of life.

> I still wanna know if they feel it any different and how different.

Wittgenstein's whole point, really, is that you can get an answer by asking them because, after all, you're speaking the same public language. That is also why pain, as it is intelligible, has a history.

[0]: https://plato.stanford.edu/entries/private-language/

I don't understand the point: we could both be seeing a red car (which is objectively red, in the sense that it has a wavelength of 625–740 nm) but we could be perceiving it radically differently (for example, maybe my perception of red is exactly like your perception of blue) and language would never help us find out.

In my world, that car, stop signs, fire extinguishers or chili peppers would look like the clear sky or sea water in yours. But we would never find out, because I would point at the car and say "red" (I have learned that all objects that look that way from my point of view are "red"), and you would agree that it's "red". There is no way to find out that our perception is different.

But I guess my lack of understanding is because I haven't read Wittgenstein, so I hope to do so when I get the time. Thanks!

> There is no way to find out that our perception is different.

Wittgenstein's point is more that there is no way in which a distinction of that kind could even be intelligible in the way that you have presented it. What is intelligible is our public behavior, namely our language use, so to find out if what you perceive is different you can ask someone if you're looking at the same thing, if it has the same characteristics, etc.

He would point to the fact that we don't have radical uncertainty in our day to day life and perception. The philosophical question is a question of sense and nonsense around the concept rather than a metaphysical question about an "inside" vs. an "outside".

I would really recommend reading Wittgenstein's Philosophical Investigations if you are interested in these kinds of questions, it's incredibly influential in philosophy precisely because of the way it approaches "age-old" problems in ways that lead to clarity and very interesting results.

> I find the classical argument with color qualia very convincing (in fact, I thought about it before even knowing what qualia were...). Nothing you can find out about wavelengths, cones, rods, neurons and so on will tell you why you see red the specific way you do, and whether my perception of red corresponds to yours or my "red" is actually your "green".

Red being green makes no sense. Let's simplify a bit and say you have a neuron for hearing the word "red", one for the visual stimuli "red" and then a sensory fusion/abstract concept of red. Several layers up you'll also have one for a "stop sign" (which also takes inputs like octagon shapes and some white letters). There are many other, more distant associations of course, we have many neurons after all.

The spoken words, physical wavelengths and stop sign design are more or less physically or socially fixed. If you magically swapped some red and green neurons then those physical stimuli wouldn't change. The brain would be forced to rewire all the associations until it arrived at the original connections again. Or it would suffer from observable mispredictions such as being more likely to respond to a green rather than a red stop sign. Why would there be a free-floating "qualia green" neuron connected to "visual red" that is yet somehow disconnected from "physical green" and why wouldn't it update once it became miswired?

Suppose that I see red the way you see green. Then, when I see a stop sign or a fire extinguisher, I'm seeing the color that you would call "green". But of course, I have been told since birth that fire extinguishers, stop signs, etc. are red, so I would naturally call that color red. And in the same way, if I see green the way you see red, when I look at grass I would be perceiving the same color that you see when you look at a stop sign. But I would call it green, because everyone knows grass is green, what else would I call it?

There would be no disagreement between us about what objects are red, no way to tell that we see them differently, and no real-world consequences that would force to rewire any neurons, because our ways of perceiving the world would both be coherent and equally valid.

By that sense, if qualia have no discernible impact on any part of the system, in what sense do they exist?

One can take this thought further. Red vs green light have real physical properties that affect one's subjective experience of them. For example different colors' physical properties change the way that they are perceived in low light conditions, the way they are perceived when placed next to other colors, etc. So at the end of the day even if hypothetically one person's qualia for red is swapped for green, you end up with a green that acts an awful lot like a red for all intents and purposes in the brain.

Edit: So my personal hunch is that qualia can't be meaningfully said to exist.

> Suppose that I see red the way you see green. Then, when I see a stop sign or a fire extinguisher, I'm seeing the color that you would call "green".

Are you sure it is possible to pluck the color I'm seeing out of my head, and the color you're seeing out of your head, and compare them side-by-side, like we could compare two different colors of paint on a piece of paper?

I'm not so sure about that. It seems to depend a lot on what phenomenon "how you see a color" actually refers to.

If "how you see a color" refers to the neural patterns excited by light of the relevant wavelengths, and you and I have the same neural patterns, then we see the color the same way. There is no metaphysical layer on top that can make one of those experiences green and the other red.

> Nothing you can find out about wavelengths, cones, rods, neurons and so on will tell you why you see red the specific way you do, and whether my perception of red corresponds to yours or my "red" is actually your "green".

This is begging the question. You're assuming that color qualia exist in the way that objects do. That "my red" and "your green" are not just properties of personal experience, but refer to objects that could be identified, drawn out, moved around and compared like the ordinary objects in our world.

The basic mistake of qualia theory is to assume that qualia share any of the properties of physical objects or even abstract concepts. That qualia are entities distinct and stable enough to be placed side by side, compared and contrasted, the way you could compare rocks or different software architectures.

I suspect that this makes about as much sense as asserting that square circles exist. You can say it, you can even kinda-sorta imagine it, but you can do that with a lot of things that fall apart on closer inspection.

For all we know, p-zombies are square circles. We think that they're a coherent concept because we can kinda-sorta imagine them, but we're nowhere near knowing that they're a coherent concept.

There is such a thing as trusting feelings and intuitions and imaginations too much, especially in the incredibly abstract context of somehow reifying your experiences and attempting to understand what they "are".

There are a lot of things that people used to feel strongly about, like "heavier objects fall faster". Many people still didn't believe Galileo was right when he proved it false with his experiments. Even when he did the experiment right before their eyes. The feelings and intuitions were too strong.

Such is the case with qualia, I think. I don't know if we'll ever be able to prove the intuitions wrong and illuminate the true nature of consciousness, like Galileo, but that doesn't mean our prescientific intuitions are correct by default. P-zombies and qualia are embellishments of our prescientific intuitions, not rigorous or scientific versions of them.

That's the problem and exactly why I say I don't think it can't be explained via a vague feeling. I've never seen someone who says there is a hard problem of consciousness adequately explain to someone why there isn't. The poster is now making what seem more like appeals to emotion throughout the thread now: "are you saying that sufficiently complicated blocks of matter are conscious?" and the like. This is what always happens.

While I personally believe there is a problem, I cannot explain why, and for a long time I didn't think there was a problem

The so-called "hard problem of consciousness" may be a problem through which one may find a suitable definition for their own humanity, but it absolutely is not a scientific problem.

If one wishes to gain concrete data, one needs to define the problem before defining the answer. Any current "problems" relating to consciousness are explainable without needing a consciousness component. I have not encountered a single problem which couldn't be explained by pure cognitive computation.

Consciousness becomes a problem suitable for scientific endeavor the moment someone can actually define the problem.

Yes, just like the various interpretations of quantum mechanics, it's not a scientific question because it doesn't allow for any testable hypothesis. Science is bound by the scientific method, questions of philosophy exist outside these bounds. Is math invented or discovered? Does the wave function collapse upon observation, or do observers exist in superposition? Is my "red" the same as your "red"?

Even if we could have exact answers to these questions, they wouldn't allow use to make predictions, and so they are not science. I think it's reasonable to say that means they aren't "useful", we can't go to space or build skyscrapers with these answers, but I think it equally makes them the more interesting questions. Our usual tools at rational inquiry fail to penetrate these sorts of questions.

Actually most of the problems you described are relevant and do exist in the objective world. The problem of the problem of consciousness is that there is no aspect in the definition of consciousness which couldn't be explained by objective mechanisms.

Currently, asking whether there is a consciousness is like asking if there is glaxtimbo. What is glaxtimbo, you may ask? I can't explain it, but I feel like it is there. That's the extent to which the definition of consciousness in its unexplainable attributes currently reaches.

As a doctor in quantum physics, we probably will be able to test between interpretations in the future, so you should not use this as an argument.

My question is simply "why am I me?" Why and how is my personal point of view fixed to this particular lump of meat and I never wake up one day as you?

Because if you woke up as me, you'd be me.

Let's assume that consciousness works in roughly the way implied by your question: it exists, but it exists entirely separately from both the body and from mental contents.

Now, let's further assume that the scenario you're proposing actually happened. Last night, in our sleep, you and I switched souls. Yesterday's me is actually you now, and yesterday's you is actually me now.

How can either of us tell the difference? Clearly the contents of our consciousness - our memories, knowledge, skills, perceptions, emotions, etc. - are attached to the body, particularly the brain. This is just the assumption we started with. What, then, makes you you and me me? Clearly our souls aren't doing a very good job at telling the difference!

Agreed, I somewhat address this in my follow on comment. In my mind rejecting the hard problem is the same as accepting the "consciousness is an illusion" answer to the problem. But I can see the point of view that views the entire conversation as a category error.

Reply on news.ycombinator.com