On naturalism and consciousness

[bennyboy]

Because the individual parties are subsets of a greater whole. When I say the universe is idealistic, I'm not claiming that I made it with my mind, or that only what I personally can experience is able to exist. That's solipsism.

So you and I are figments of some larger mind. I find this even harder to swallow.

What's this "figment" stuff? You are comfortable with physical objects being a subset of a physical framework, and yet you think that ideas cannot be subsets of a framework of ideas? This seems like special pleading.

These mind subsets still do not explain consistency and repeatability between the subsets.

Why not? It is in fact the organizations of ideas in your mind that you call "consistency." How do you know any physical property or function is consistent? You remember it, and compare it with new experiences. But there's no special reason why the Matrix, or the Mind of God, or a BIJ-generated world couldn't have internal consistencies worth noting.

I don't know how a mind works; I'm not a neuroscientist. More importantly, I don't need to explain how a mind works to understand that there is more than one.

No, but if you want a physical model to be a sufficient account of human experience, you must be able to explain why there is more than zero. So far, there's not even a vaguely good account of how a purely mechanical system arrives at the ability to subjectively experience.

As far as a possible physical explanation of a mind arrising from physical mechanisms, read up on neural networks. It's a general algorithm that can learn and make predictions. It's used to buy and sell stock in the stock market. Also, it's based off how the neurons in our brains work.

I not only know of ANNs, I've done some programming with them myself. ANNs are interesting in that they can allow a system to adapt to its environment via "punishment" and "reward," and I think it's very likely that even a simple algorithm will be able to lead to Turing-passing machines, via brute force, in our lifetimes, especially if people on the internet can be drawn into providing feedback "punishment" and "reward." (for example by having a couple hundred million people rate iterations of a computer-generated song)

However, they say very little about the philosophical reality of subjective minds-- specifically, why there are subjective minds rather than an absence of them.

[The Grand Nudger]

I not only know of ANNs, I've done some programming with them myself. ANNs are interesting in that they can allow a system to adapt to its environment via "punishment" and "reward," and I think it's very likely that even a simple algorithm will be able to lead to Turing-passing machines, via brute force, in our lifetimes, especially if people on the internet can be drawn into providing feedback "punishment" and "reward." (for example by having a couple hundred million people rate iterations of a computer-generated song)

However, they say very little about the philosophical reality of subjective minds-- specifically, why there are subjective minds rather than an absence of them.

Are there?

I couldn't help myself. I have to ask, if you're certain that we'll see brute forced turing-passing machines in our lifetime, how will you overcome the problem of other minds (or do you not see that as a problem)? I think that a machine like that would have a hell of alot to say about "mind". We'd be coming face to face with a system that turns "possible theoretical interpretations" into cold hard reality. A turing passing system would give us no reason to doubt it's "mind" unless we were willing to doubt the minds of others - and possibly even ourselves.

[bennyboy]

I not only know of ANNs, I've done some programming with them myself. ANNs are interesting in that they can allow a system to adapt to its environment via "punishment" and "reward," and I think it's very likely that even a simple algorithm will be able to lead to Turing-passing machines, via brute force, in our lifetimes, especially if people on the internet can be drawn into providing feedback "punishment" and "reward." (for example by having a couple hundred million people rate iterations of a computer-generated song)

However, they say very little about the philosophical reality of subjective minds-- specifically, why there are subjective minds rather than an absence of them.

Are there?

I couldn't help myself. I have to ask, if you're certain that we'll see brute forced turing-passing machines in our lifetime, how will you overcome the problem of other minds (or do you not see that as a problem)? I think that a machine like that would have a hell of alot to say about "mind". We'd be coming face to face with a system that turns "possible theoretical interpretations" into cold hard reality. A turing passing system would give us no reason to doubt it's "mind" unless we were willing to doubt the minds of others - and possibly even ourselves.

The short answer is, I don't know. I can imagine the possibility of whole social revolutions going on-- android rights, etc., and the universe's big joke on us would be that we've programmed them to stimulate emotional responses in ourselves, and then responded emotionally, perhaps at great inconvenience to ourselves-- all the while without the androids actually being able to experience anything.

But if I couldn't tell the difference between a real human and a borg, then I would indeed have a serious philosophical issue: I'd have to reexamine my criteria for being willing to accept that another physical entity is in fact a thinking being.

[The Grand Nudger]

Any chance you might speculate on where the cutoff may be for creating a machine capable of handling that much data in such a dynamic way? Lets say we benchmark cpus at about 2.6 billion transistors (SPARC T3). What if they had 32x that many transistors, or were 32x as dense? That would be about 15years from today if Moores Law holds. Maybe a little more (you never know what sort of wall we might run into, eh?).

Would a machine with 32x the brute force power (imagine the programming options) of something like SPARC T3 be in the running for a pass on a Turing test, judging by our benchmark of today-in your estimation? Or would you put it further out in time or scale?

[Whateverist]

And yet that's the same demand which you made of physicalism above. That's special pleading.

None of your declarations about what experience is establish that it is what you say. They're just bare assertions.

What I said had nothing all to do with what experience is on a metaphysical level. All that matters is what can be known. Any thing beyond what can be known is unfounded speculation.

But is it really "what can be known" or "what is assumed"?

Declarations based on interpretations of subjective experience are not automatically valid. That one has subjective experience is a given. But the only thing that can be assumed to be valid is the phenomenology of the experience, not its interpretation.

---

Would a machine with 32x the brute force power (imagine the programming options) of something like SPARC T3 be in the running for a pass on a Turing test, judging by our benchmark of today-in your estimation?

That bar is way too low. Hell, when I was a kid at Disneyland the robotic parrots near the jungle ride passed with flying colors.

[Surgenator]

These mind subsets still do not explain consistency and repeatability between the subsets.

Why not? It is in fact the organizations of ideas in your mind that you call "consistency." How do you know any physical property or function is consistent? You remember it, and compare it with new experiences. But there's no special reason why the Matrix, or the Mind of God, or a BIJ-generated world couldn't have internal consistencies worth noting.

There is no reason why different subsets would have similiar experiences unless the subsets are similiar. In your monism world view, what guarantee's similiar subsets?

Lets me you ask you this, where does the mind store its experiences in your world view? We know a mind interacts with experiences, and we know we forget and remember experiences. Taking that into account, the mind has to store the experiences somewhere and it's retrieval is not perfect. In a world of purely minds, where are the storage devices? And if your gonna say it is in the mind itself, then please explain how the mind can forget and remember.

ANNs are interesting in that they can allow a system to adapt to its environment via "punishment" and "reward," and I think it's very likely that even a simple algorithm will be able to lead to Turing-passing machines, via brute force, in our lifetimes, especially if people on the internet can be drawn into providing feedback "punishment" and "reward." (for example by having a couple hundred million people rate iterations of a computer-generated song)

However, they say very little about the philosophical reality of subjective minds-- specifically, why there are subjective minds rather than an absence of them.

Great, you agree that neural networks can pass the Turning test in the future. So they are a viable candidate for the explanation of the mind from pure physicsal processes.

We have subjective minds because everybodies experiences are slightly different. Using the ANN example, you should know that for the same training tasks the hidden nodes values would be different (if your neural network is complex enough) if the starting conditions are slightly different. Nevertheless, the end result would be very very similiar.

... if you want a physical model to be a sufficient account of human experience, you must be able to explain why there is more than zero.

Abiogenesis + evolution + biological neural networks, ta da.

[Neo-Scholastic]

…Declarations based on interpretations of subjective experience are not automatically valid…the only thing that can be assumed to be valid is the phenomenology of the experience, not its interpretation.

Here I am making a distinction between knowledge and true belief as it relates to speculative ideas for which no one can offer any justification .

Suppose you believe in deceived by a demon, a brain in a vat, the dream of a sleeping god, or something else of that nature. None of these can serve as the basis for knowledge since nothing about them can be confirmed by either logic or inductive inquiry, or as I like to say reason applied to experience. That is just the human condition. However, the lack of omniscience is not reason to believe that no knowledge is possible. Those who think about these things either recognize first principles of reason (like the law of non-contradiction) or fall into self-refuting absurdity. As it specifically relates to philosophy of mind, there are some problems, like the problem of other minds that cannot be properly known since it is amenable to neither deduction nor direct experience.

Those people who do not have a prior commitment to ontological naturalism find the results of contemporary psi research compelling. If a machine were able to produce psi effects of equal magnitude to that of humans I would consider that as satisfactory evidence that they are indeed conscious.

[The Grand Nudger]

Whahahahat? Psi effects? Goddamit- Don't all computers of today pass the bar for you then, being that they all express precisely as much "psi effect" as a human being - that amount being nil?

[bennyboy]

There is no reason why different subsets would have similiar experiences unless the subsets are similiar. In your monism world view, what guarantee's similiar subsets?

I don't think I understand what you're saying. There's a large idealistic universe, and it has minds in it. Basically, it's the same as the physical universe you imagine, except it doesn't reduce down to mechanical elements-- it reduces down only to concepts: mathematical concepts, for example.

Lets me you ask you this, where does the mind store its experiences in your world view?

If it's a human mind, in the brain (at least as far as I can tell).

Great, you agree that neural networks can pass the Turning test in the future. So they are a viable candidate for the explanation of the mind from pure physicsal processes.

The Turing test isn't a test of consciousness. It's a test of the ability of a machine to simulate the behaviors of a conscious person. To claim otherwise wouldn't be a conclusion based on observation, it would be an assumption-- one most people probably aren't willing to make.

We have subjective minds because everybodies experiences are slightly different. Using the ANN example, you should know that for the same training tasks the hidden nodes values would be different (if your neural network is complex enough) if the starting conditions are slightly different. Nevertheless, the end result would be very very similiar.

That's a narrative, not a mechanical explanation.

... if you want a physical model to be a sufficient account of human experience, you must be able to explain why there is more than zero.

Abiogenesis + evolution + biological neural networks, ta da.

That's like saying "The Father, the Son, the Holy Ghost, ta da." You've just put a bunch of words from your world view together. It neither confirms the validity of your model, nor explains how in your model, mind exists.

[ManMachine]

I remember doing a thread like this over a year ago now, but I thought it might be best to start with a clean slate, and see how we go. Also, I really like discussing this topic especially on an atheist forum because being in the minority is more fun. Here we go:

I don't believe consciousness can be explained by way of a naturalistic account. Why? Because I don't think particles have it in them to act in such a way as to recreate what we mean by consciousness i.e. our thoughts, beliefs, attitudes etc.

Let's use an example; my belief that spoons are curved. So to make things easier, let's call this belief p. Now, how can we possibly arrange particles in such a way that they would express p? How could some physical arrangement *ever* describe p? I don't think it's possible to physically arrange particles in such a way that would then inherently possess the belief that other sets of particles - aka spoons - have the property of being curved.

Maybe we could place a spoon on the kitchen counter, and put beside it a piece of paper pointing to it, saying "curved". But I wouldn't say the particles of lead forming the word "curved" are arranged in a way that make them hold p. The reason being that it takes an already conscious being *to make that connection*. The already conscious being has the ability to give meaning to such an arrangement of lead, and mind you, it has to be a being that speaks English. Therefore, in no way is the arrangement of lead inherently a beholder of p, since not even all conscious beings can arrive at that conclusion to begin with.

As we can see, for a particle to be "about" another particle, consciousness is a prerequisite... almost as if it weren't made up of particles in the first place

Or, our understanding of 'consciousness' is flawed/wrong. Which I'm more inclined to believe.

MM