Automatons

[westprog]

Yes I definitely fall on the side of Hofstadter and those that have followed him.

But that is irrelevant to the fact that Penrose's conclusion is simply the result of a fallacy. An error in logic.

If you don't believe me (but I suspect you do) simply google lucas + penrose + fallacy. Hell, if you just google lucas + penrose, something like half the links are to articles criticising the thesis/theory/whatever-one-should-call-it.

So it is not simply Penrose's "interpretation" versus Hofstadter's "interpretation", it is Penrose's "interpretation" versus the accepted standards for logical argument.

Fine - it's Penrose vs. Hofstadter and the proponents of strong AI, rather than Penrose vs. Godel.

However, even if Penrose is not correct in his views - that still does not imply that the case for strong AI has been proven.

Surely -- That is probably why I never suggested anything of the sort.

I merely asserted that the subjective nature of consciousness cannot be investigated at all by a third party, and cannot be investigated fully by yourself. Hence, it cannot be investigated fully.

You, because you are looking for any slight way to invalidate the computational model, grasp onto this and come back with a strawman claiming the computational model asserts that consciousness cannot be investigated at all. Which is of course utter nonsense.

If we regard consciousness as a subjective phenomenon, and assert that a subjective phenomenon cannot be investigated, then that would seem to me to be restricting the nature of the investigation.

We don't actually know that it is impossible to investigate subjective phenomena. We simply know that it has been impossible so far, and we have no means of doing so as yet. There is no fundamental physical law that says that in a thousand years we won't be able to wire two heads together and share experiences.

 

[Nick227]

You are hung up on this idea that subjective experience is somehow different from behavior. Yet, in the entire history of this argument, nobody has ever been able to argue why except "well, it just feels different.. so it must be..."

For me, one clear issue here is the way you're using language. There is simply no such thing as a "subjective experience" and to me it would be far more useful to point this out than to equate it to behaviour. "Experience" is simply thinking - the packaging and communication of data into dualistic soundbites. There is no actual subjective experience.

If you really want to equate it to behaviour, then I think it you also need to state the behaviour of what exactly.

Nick

 

[Nick227]

I merely asserted that the subjective nature of consciousness cannot be investigated at all by a third party, and cannot be investigated fully by yourself. Hence, it cannot be investigated fully.

As previously pointed out, this is simply not so. The subjective nature of consciousness can be investigated by third party techniques. See earlier link to heterophenomenology.

Nick

 

[lupus_in_fabula]

We know a little bit about consciousness.

Well, we might believe we know very little about consciousness, although it might also be revealed that we already knew most there is to know about it – i.e. consciousness not existing in the physical sense at all. I’m open to both suggestions.

We know it's not, in practice, divisible and recombinable. As experienced, each person has just one consciousness. Why assume that it is a combination of the experience of the bits?

Because we don’t know that it’s not divisible or recombinable. In fact, we might make the assumption that it is indeed divisible through empirical findings in the field of neurology and psychiatry. Moreover, it might not exist in the physical sense in the first place.

Example: Regarding lobotomy; person X has lobotomy (left hemisphere is disconnected from the right) and as a consequence person X seems to acquire the behaviour of two isolated persons. One could perhaps say that he behaves as if there are two consciousnesses. For instance, the right hemisphere believed in god whereas the left hemisphere didn’t. Thus, we have two sets of completely different ontologies where there previously was only one (see for instance Ramachandran for further details).

And what about multiple personalities? Can we say in a convincible manner that it’s the same consciousness which is underlying the different personalities, or is it a matter of different consciousnesses that seems to correlate to different personalities? Sometimes one personality cannot remember the experience of another, so perhaps there’s no “uber-conscious” reality that could keep a sense of continuity between the personalities.

 

[westprog]

Well, we might believe we know very little about consciousness, although it might also be revealed that we already knew most there is to know about it – i.e. consciousness not existing in the physical sense at all. I’m open to both suggestions.

I think being open to suggestions is the first prerequisite for this subject. Certainty deosn't seem to be appropriate.

Because we don’t know that it’s not divisible or recombinable. In fact, we might make the assumption that it is indeed divisible through empirical findings in the field of neurology and psychiatry. Moreover, it might not exist in the physical sense in the first place.

Example: Regarding lobotomy; person X has lobotomy (left hemisphere is disconnected from the right) and as a consequence person X seems to acquire the behaviour of two isolated persons. One could perhaps say that he behaves as if there are two consciousnesses. For instance, the right hemisphere believed in god whereas the left hemisphere didn’t. Thus, we have two sets of completely different ontologies where there previously was only one (see for instance Ramachandran for further details).

Given that we accept that a mind can be split in two to the extent of having two seperate consciousnesses - and while I am not disputing this, I would like to read more before I give full assent - I don't think that this applies granularity down to the molecular level. Human beings have bilateral symmetry, and it's reasonable enough to be able to make two brains out of one. Slicing 'em finer than that would be another matter.

And what about multiple personalities? Can we say in a convincible manner that it’s the same consciousness which is underlying the different personalities, or is it a matter of different consciousnesses that seems to correlate to different personalities? Sometimes one personality cannot remember the experience of another, so perhaps there’s no “uber-conscious” reality that could keep a sense of continuity between the personalities.

I'm a lot less convinced by multiple personalities. I don't see any sound evidence that they really exist beyond psychiatric convenience.

 

[rocketdodger]

Summing up zero millions of times adds up to zero. If a water molecule exhibits no "experience effects", then why should a more complex thing do so?

We know a little bit about consciousness. We know it's not, in practice, divisible and recombinable. As experienced, each person has just one consciousness. Why assume that it is a combination of the experience of the bits?

Who is assuming this?

If it's subjective, then it's different. Behaviour is objective. I think that explains why. How is it possible to claim that something that is subjective is the same thing as something which isn't?

It isn't. Which is why "subjective experience" is nothing more than an objective behavior -- from the standpoint of the one behaving. Subjectivity isn't an actual thing. It is simply the result of a unique point of view. Namely, the point of view of the entity in question.

 

[Nick227]

It isn't. Which is why "subjective experience" is nothing more than an objective behavior -- from the standpoint of the one behaving. Subjectivity isn't an actual thing. It is simply the result of a unique point of view. Namely, the point of view of the entity in question.

I'm pretty sure I understand what you're saying here, and I agree with it. However, you are still couching your point in, to my mind, excessively dualistic terminology (see underlinings). IMO this is likely to be confusing to anyone who does not already appreciate the point.

I would rather say... that which is termed "subjective experience" is actually just that which is internally represented by the organism over a certain time period, packaged through thinking as the dualistic proposition "I experience...[whatever]"

eta: when you are trying to convey a monist perspective I think it helps to avoid dualistic terminology where feasible.

Nick

 

[rocketdodger]

I'm pretty sure I understand what you're saying here, and I agree with it. However, you are still couching your point in, to my mind, excessively dualistic terminology (see underlinings). IMO this is likely to be confusing to anyone who does not already appreciate the point.

I would rather say... that which is termed "subjective experience" is actually just that which is internally represented by the organism over a certain time period, packaged through thinking as the dualistic proposition "I experience...[whatever]"

eta: when you are trying to convey a monist perspective I think it helps to avoid dualistic terminology where feasible.

Nick

Fair enough.

 

[rocketdodger]

We don't actually know that it is impossible to investigate subjective phenomena. We simply know that it has been impossible so far, and we have no means of doing so as yet. There is no fundamental physical law that says that in a thousand years we won't be able to wire two heads together and share experiences.

Yes, there is. It's called the identity axiom. A == A.

 

[rocketdodger]

I am not purposely misunderstanding you, RD. I happen to think you are completely a "Cartesian materialist," to use the term coined by Daniel Dennett. You like to project yourself as a materialist but your thinking reveals strong Cartesian elements.

I am a materialist that doesn't find it necessary to completely abandon commonly used terms outside of a formal context because there is no point other than to make one's brain hurt.

Especially because it leads to an impass in arguments like this. We say there is no subjective experience. The dualists say "wait.. what about qualia, and feeling, and all that jazz?" We say "it is just an illusion." They say "how can it be an illusion if it is real?" Back and forth, endlessly.

So it is more productive to say subjective experience does exist and then try to redefine what the term evokes in people's minds. That way we can say "ok, I grant you that qualia are real, but perhaps they aren't exactly what you thought they were... " and try to explain it from our point of view. At least that is my opinion.

 

[westprog]

Yes, there is. It's called the identity axiom. A == A.

And if it turns out that it is possible to wire two heads together and share experience, then we'll know that the "identity axiom" is false.

There's a bucketload of assumptions in presuming that experience cannot be shared. Unless it violates fundamental laws of physics, we have no reason to suppose it impossible. OTOH, until we can either do it, or plausibly demonstrate how it might be done, we have no reason to suppose it possible either.

 

[Nick227]

I am a materialist that doesn't find it necessary to completely abandon commonly used terms outside of a formal context because there is no point other than to make one's brain hurt.

Especially because it leads to an impass in arguments like this. We say there is no subjective experience. The dualists say "wait.. what about qualia, and feeling, and all that jazz?" We say "it is just an illusion." They say "how can it be an illusion if it is real?" Back and forth, endlessly.

So it is more productive to say subjective experience does exist and then try to redefine what the term evokes in people's minds. That way we can say "ok, I grant you that qualia are real, but perhaps they aren't exactly what you thought they were... " and try to explain it from our point of view. At least that is my opinion.

OK.

I wouldn't personally consider "feeling" an illusion. Feeling is real. The illusion is that it is happening to someone. Though on your behaviourist level, this isn't illusion.

Nick

 

[Nick227]

And if it turns out that it is possible to wire two heads together and share experience, then we'll know that the "identity axiom" is false.

There's a bucketload of assumptions in presuming that experience cannot be shared. Unless it violates fundamental laws of physics, we have no reason to suppose it impossible. OTOH, until we can either do it, or plausibly demonstrate how it might be done, we have no reason to suppose it possible either.

Well, as I see it, the problem here is that if the strong AI theorists are correct, there isn't actually anyone to share experiences. "Someone" and "experience" are merely aspects of the same artificial duality. What is experienced includes "someone." You can't separate the person from the representations being created.

Nick

 

[westprog]

It isn't. Which is why "subjective experience" is nothing more than an objective behavior -- from the standpoint of the one behaving. Subjectivity isn't an actual thing. It is simply the result of a unique point of view. Namely, the point of view of the entity in question.

If that's the case, then there's nothing especially difficult in observing subjective experience. All scientific observation has to pass through the same filter.

It might be that there is a theoretical issue with the capacity of an organism to correctly interpret its own subjective experience. That doesn't imply that nothing useful can be inferred from reporting of subjective experience. It might be that there are aspects of subjective experience which are forever hidden by fundamental principles. Or it might turn out that as yet unknown physical processes will make subjective experience accessible, reproducible and understandable. Just another bit of physics.

 

[westprog]

Well, as I see it, the problem here is that if the strong AI theorists are correct, there isn't actually anyone to share experiences.

There's always the possibility that the strong AI theorists are wrong.

"Someone" and "experience" are merely aspects of the same artificial duality. What is experienced includes "someone." You can't separate the person from the representations being created.

If A is experiencing the sensation of being A, and also what B is looking at, then that will presumably continue to generate the illusion, or else confirm the fact, of A's identity. If A and B were merged entirely, then that might be a different matter.

 

[westprog]

Who is assuming this?

I'm trying to see some point to the whole "water molecule having experience" concept. If the experience of being part of a human being has nothing to do with the experience of being a human being, then there really doesn't seem to be anything useful in the concept.

 

[Nick227]

There's always the possibility that the strong AI theorists are wrong.

Absolutely.

If A is experiencing the sensation of being A, and also what B is looking at, then that will presumably continue to generate the illusion, or else confirm the fact, of A's identity. If A and B were merged entirely, then that might be a different matter.

Um, the duality radar is flashing. A does not experience. "A" and "experience" are two sides of a self-reinforcing duality created by thinking.

eta: continuing from my first comment - imo it could still be that all the theorising about consciousness is utterly off the mark. There are core issues around selfhood and thinking that are invariably overlooked in our desire to create some meaningful understanding of our world. Materialism can still legitimately be viewed as purely a self-referencing interpretation imo. Even Dan Dennett, who I've been reading recently, when addressing selfhood (eta in Consciousness Explained) very quickly leaps back behind the line that the "I" is a necessary centre even though it doesn't exist. There are places it's tough for any human to go.

Nick

 

[Nick227]

If that's the case, then there's nothing especially difficult in observing subjective experience. All scientific observation has to pass through the same filter.

It might be that there is a theoretical issue with the capacity of an organism to correctly interpret its own subjective experience. That doesn't imply that nothing useful can be inferred from reporting of subjective experience. It might be that there are aspects of subjective experience which are forever hidden by fundamental principles. Or it might turn out that as yet unknown physical processes will make subjective experience accessible, reproducible and understandable. Just another bit of physics.

Arguably, all that ever happens on an internet forum is individuals reporting their subjective experiences. That is what thinking and the expressing of thinking is.

Nick

 

[rocketdodger]

I'm trying to see some point to the whole "water molecule having experience" concept. If the experience of being part of a human being has nothing to do with the experience of being a human being, then there really doesn't seem to be anything useful in the concept.

Where else would you draw the line?

A person can experience. Ok.

What about a retarded person?

What about a really drunk person?

What about a monkey?

A dog?

A chipmunk?

A bird?

A tree?

A worm?

A thermostat?

 

[westprog]

Where else would you draw the line?

A person can experience. Ok.

What about a retarded person?

What about a really drunk person?

What about a monkey?

A dog?

A chipmunk?

A bird?

A tree?

A worm?

A thermostat?

We dont know. And saying that a thermostat can experience doesn't solve the problem, it's simply bypassing it.