Just to qualify - just because I believe it is reducible to computation doesn't mean that I think that we're going to do so with perfect fidelity, or that I think a perfect simulation of my brain _is_ me.
Agreed. I think I would say something even weaker, in that I'm not convinced that hardware is irrelevant to computation.
So maybe we won't be able to replicate human consciousness on platforms radically different from the ones human consciousness is currently running on, any more than we can replicate the Internet on tin cans and string.
I just think the implication of that is non-human consciousness, rather than no consciousness at all.
(nods) And it might turn out to just be easier to replicate the hardware in its original configuration -- that is, replicate the person on carbon. Which would also have some built-in social-acceptance features.
Me too, at least this Friday; and in situations like that I find it particularly easy to believe that a computer could simulate what I've got without being overly taxed! Hell, a very small shell script might get more useful work done this afternoon than I will.
Yes, most mornings Julie can be easily simulated by a nose poking out from under a duvet and a tape recorder that says "Tea?" whenever a movement sensor is tripped.
Hm. When I first saw arguments like Penrose's I was comparatively young, and responded something like "wow, that's interesting. it's weird, but the guy is such a notable scientist, it must make some kind of sense, which is awesome".
Now, I think more like "I don't see why it should make any kind of sense." I suppose its possible that the brain does something a computer can't (although it seems unlikely to me) but there certainly doesn't seem to be any evidence other than wishful thinking that it does.
I don't like the idea that I could be written in Java either.
I like to imagine it'd need something like Haskell, or at least Lisp.
More realistically I worry that I am in fact made of a hideous and haphazard hash-up of horrible K&R-style C, some iffy FORTRAN IV, and a few incomprehensible bits of assembler. The evidence definitely suggests I'm not implemented on anything that effectively abstracts memory management away.
Always thought that one was sailing close to pure blasphemy. God never used Perl for His hacking. If He did, it'd have been sh scripts (definitely no bash), a bit of sed, and if absolutely necessary a touch of awk. That's Hacking As God Intended.
No Bash?! Can you really see a real deity writing scripts without tab-completion?! (Not to mention he'd want to use a Free Software solution rather than a proprietary, closed one).
Personally, though, I think the mere existence of perl is proof that God doesn't exist, or that if he does he hates me...
Maybe he'd give it us if anyone asked him for it - he's just not provided it as a default, to save disk space. But of course those who ask him stuff the most (the priesthood) make too much money from providing tech support to want anyone to be able to do anything by themselves.
Oh Penrose is cracked -- at least on that issue. At least the microtubules thing (Emperor's new mind) is just... well... hmm.. At the time I had not long written a bloody long essay on microtubules so I was pretty sure he was talking out of his backside rather than way, way cleverer than me (and , of course, all the sources I had used in my essay, some of whom most defintiely *were* way cleverer than me).
Yes, this. I think that our widespread confusion about consciousness is because we don't really have a good enough understanding of what we mean by consciousness to ask the question with sufficient clarity that it would admit of an empirical answer.
My ill-informed but strongly-held hunch is that consciousness is a contingent, transitory illusion of certain emphatically physical processes in the human brain. I could be wrong. In fact, I've read enough disturbing experiments related to consciousness that I'm certain I'm wrong about all sorts of things, but that doesn't stop me believing them firmly.
No toilet roll - bacteria Can't find the Apple TV remote - bacteria The duvet mysteriously disappearing from my side of the bed and appearing on your side - bacteria Disease - bacteria
I'm not even sure they exist! _I've_ never seen one!
"reducible" is the nearest, though I suppose someone like David Chalmers could be right without interactive dualism being true (which is the usual explanation preferred by people who believe in souls and whatnot).
Chalmers is a philosopher who thinks that consciousness relies on more than physical stuff, because there is a possible (for some value of "possible" which is itself the subject of much discussion, as far as I can tell) world in which everything is physically identical to this world, but people aren't conscious. The possible world is usually called the "zombie world", and the people in it "philosophical zombies".
If I remember rightly, Chalmers does think that the physical world is casually closed, so he doesn't think that consciousness pushes back on the world, if you like: on his view, consciousness is an epiphenomenon produced by special laws about consciousness which apply in our world (he calls these "bridging laws", which is a term from the philosophy of science for laws which map one scientific theory onto another, like mapping classical thermodynamics onto statistical mechanics, say).
The strong AI people think Chalmers's idea is pretty silly, hence Yudkowsky's Philosophical Zombie Movie, in which the line "I'm Daniel Dennett, bitches" wins the Internet.
However, supernaturalists typically won't find Chalmers's ideas satisfactory either: there's a popular argument for theism which more or less runs "Fucking consciousness, how does that work?", based on the idea that atheists must also be materialists. Chalmers is an atheist non-materialist.
Interactive dualism is the view that there's some separate sort of mind stuff that causes physical stuff to happen. While Googling that, I found a useful summary of mind body relations.
Aaaah. I've read the Philosophical Zombie Movie before, and found it terribly amusing.
Interactive Dualism I could have guessed at - just hadn't bumped into Chalmers before. I remember chunks of this stuff from the philosophy I did at uni, but generally not the names of the people associated with it.
I don't think consciousness is reducible purely to computation, I think it's probably an emergent property that results from a combination of computation, the hardware it runs on, and possibly other stuff we're not yet aware of, like maybe when and where the computation is instantiated on that hardware. I don't think the hardware necessarily has to be biological, however. So my answer is somewhere in between options 1 and 2.
The Bekenstein Bound puts an upper limit on the amount of different states a quantum system can be in. Given that, every finite system is, literally, a finite-state machine.
In the case of an average human being, any computer capable of processing data on the order of 2.5072178×10^38 megabytes could therefore *perfectly* emulate that person, unless:
1) The Bekenstein Bound doesn't hold. The Bekenstein Bound is equivalent to a restatement of Heisenberg's Uncertainty Principle, one of the cornerstones of modern physics. Or:
2) The Church-Turing thesis is false. This is the cornerstone of modern computer science.
These two, the Uncertainty Principle and the Church-Turing thesis, are two of the most tested, most reliable scientific findings of the last century.
If the world was classical, not quantum, then because classical physics deals in continua it would be theoretically possible to have a non-computable physical system. But at a fine enough granularity quantum physics seems to hold.
So I'm pretty certain that all physical processes are computable, because the opposite would require me to have a better explanation than the best current knowledge in two different sciences...
Yep, it seems plausible to me that all physical processes can be described by computation. What I'm not so sure of is whether they're reducible to that computation.
I would say it lies somewhere between can be simulated by a computer, but the simulation couldn't produce "real understanding" and can't even be simulated by computer, but nevertheless has a scientific explanation . One day it will probably be possible to simulate true consciousness (hell, given sufficient time, computing power and imaginative soft- and hardware architecture, it may even be possible to recreate a true consciousness) but I think it takes more than just raw computation to do this (though admittedly I would be at a loss to even start trying to explain why this is...)
Rephrase: could be simulated by computer in theory, possibly, given a powerful enough computer, the likes of which we're not yet capable of building. Which may mean my real vote should be for the first option.
(I'm firmly in the "Do I believe science can explain everything? Yes. So I believe we know all there is to know about science? No." camp)
Yes, I was assuming this. I don't understand when people forget progress happens. Or indeed supposed scientists who operate on "if it's not proven true it must be false", rather than just not known yet.
I think people tend to confuse consciousness with being human. Humans also hunger, thirst, lust, love, and seek shelter. These things are motivated by our core animal nature, not by any sort of abstract reasoning.
Either 1 or 4, and I'm not sure which. I wouldn't be surprised if it turned out consciousness doesn't have a scientific explanation, but Hofstadter's reductionist arguments are persuasive.
I'm the one vote for "doesn't have a scientific explanation.."
Consciousness is part of Spirit. Like many things in the Mind and a few things in the Body, Consciousness (being part of the Spirit) cannot be seen or measured, (aside from extensive behavioral observation.)
[x] I'm an AI researcher, and I'm not sure that I can answer this question. [x] Sometimes I worry that I'm a zombie, on the grounds that I cant prove that I have qualia.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
andrewducker) wrote2010-07-23 03:43 pm
no subject
no subject
So maybe we won't be able to replicate human consciousness on platforms radically different from the ones human consciousness is currently running on, any more than we can replicate the Internet on tin cans and string.
I just think the implication of that is non-human consciousness, rather than no consciousness at all.
no subject
no subject
no subject
Me too, at least this Friday; and in situations like that I find it particularly easy to believe that a computer could simulate what I've got without being overly taxed! Hell, a very small shell script might get more useful work done this afternoon than I will.
no subject
no subject
Now, I think more like "I don't see why it should make any kind of sense." I suppose its possible that the brain does something a computer can't (although it seems unlikely to me) but there certainly doesn't seem to be any evidence other than wishful thinking that it does.
no subject
no subject
I like to imagine it'd need something like Haskell, or at least Lisp.
More realistically I worry that I am in fact made of a hideous and haphazard hash-up of horrible K&R-style C, some iffy FORTRAN IV, and a few incomprehensible bits of assembler. The evidence definitely suggests I'm not implemented on anything that effectively abstracts memory management away.
no subject
http://xkcd.com/224/
no subject
no subject
Personally, though, I think the mere existence of perl is proof that God doesn't exist, or that if he does he hates me...
no subject
no subject
apt-get install universe-src
Re: apt-get install universe-src
no subject
no subject
no subject
no subject
no subject
My ill-informed but strongly-held hunch is that consciousness is a contingent, transitory illusion of certain emphatically physical processes in the human brain. I could be wrong. In fact, I've read enough disturbing experiments related to consciousness that I'm certain I'm wrong about all sorts of things, but that doesn't stop me believing them firmly.
Consciousness...
Re: Consciousness...
No toilet roll - bacteria
Can't find the Apple TV remote - bacteria
The duvet mysteriously disappearing from my side of the bed and appearing on your side - bacteria
Disease - bacteria
I'm not even sure they exist! _I've_ never seen one!
no subject
no subject
no subject
If I remember rightly, Chalmers does think that the physical world is casually closed, so he doesn't think that consciousness pushes back on the world, if you like: on his view, consciousness is an epiphenomenon produced by special laws about consciousness which apply in our world (he calls these "bridging laws", which is a term from the philosophy of science for laws which map one scientific theory onto another, like mapping classical thermodynamics onto statistical mechanics, say).
The strong AI people think Chalmers's idea is pretty silly, hence Yudkowsky's Philosophical Zombie Movie, in which the line "I'm Daniel Dennett, bitches" wins the Internet.
However, supernaturalists typically won't find Chalmers's ideas satisfactory either: there's a popular argument for theism which more or less runs "Fucking consciousness, how does that work?", based on the idea that atheists must also be materialists. Chalmers is an atheist non-materialist.
Interactive dualism is the view that there's some separate sort of mind stuff that causes physical stuff to happen. While Googling that, I found a useful summary of mind body relations.
no subject
Interactive Dualism I could have guessed at - just hadn't bumped into Chalmers before. I remember chunks of this stuff from the philosophy I did at uni, but generally not the names of the people associated with it.
no subject
no subject
no subject
no subject
no subject
In the case of an average human being, any computer capable of processing data on the order of 2.5072178×10^38 megabytes could therefore *perfectly* emulate that person, unless:
1) The Bekenstein Bound doesn't hold. The Bekenstein Bound is equivalent to a restatement of Heisenberg's Uncertainty Principle, one of the cornerstones of modern physics. Or:
2) The Church-Turing thesis is false. This is the cornerstone of modern computer science.
These two, the Uncertainty Principle and the Church-Turing thesis, are two of the most tested, most reliable scientific findings of the last century.
If the world was classical, not quantum, then because classical physics deals in continua it would be theoretically possible to have a non-computable physical system. But at a fine enough granularity quantum physics seems to hold.
So I'm pretty certain that all physical processes are computable, because the opposite would require me to have a better explanation than the best current knowledge in two different sciences...
no subject
no subject
no subject
.
One day it will probably be possible to simulate true consciousness (hell, given sufficient time, computing power and imaginative soft- and hardware architecture, it may even be possible to recreate a true consciousness) but I think it takes more than just raw computation to do this (though admittedly I would be at a loss to even start trying to explain why this is...)
no subject
(I'm firmly in the "Do I believe science can explain everything? Yes. So I believe we know all there is to know about science? No." camp)
no subject
no subject
no subject
no subject
no subject
no subject
Consciousness is part of Spirit. Like many things in the Mind and a few things in the Body, Consciousness (being part of the Spirit) cannot be seen or measured, (aside from extensive behavioral observation.)
no subject
[x] Sometimes I worry that I'm a zombie, on the grounds that I cant prove that I have qualia.
no subject
b) Have you seen the Philosphical Zombie Movie (link below)?