There really might be an afterlife

[realpaladin]

I always like to say:

" I can't not exist, as far as I'm concerned. "

Descartes would love you for modernising his thoughts.

 

[realpaladin]

So the information needed for even just that carricature of a model is basically 2 to the power 1,000,000,000,000,000. Read that carefully. It's not just a million billions, it's two to the power a million billions! Well, 2^10 is approximately 10^3, so that's approximately 10^300,000,000,000,000. Yes, you write a one and then 300,000,000,000,000 zeroes behind it.

Even assuming that some particles assembled themselves in the equivalence of a brain (already improbable), it would be one chance in 2^1,000,000,000,000,000 that it's configured like yours.

My brain now has a bigger chance to exist... do you know how many neurons just gave up and went home just because I tried to imagine or put in perspective these numbers?

(All would be an obvious response, but then you probably make the infinite number of monkeys that are typing this post angry)

 

[rocketdodger]

He pointed out that consciousness, whatever it is, actually exists and is therefore a physical process. It arises somehow out of the interaction of atoms and energies and other particles.

Intelligence, which is not the same thing, does not require consciousness per se. But the phenomonon of the subjective perceptual experience, being real, thus requires a cause from real-world atoms and whatnot.

And therefore it is not based purely on information processing. In other words, we should not expect an electronic brain to have consciousness arise simply because we perfectly duplicate the information processing content of a wetware brain.

The third paragraph does not logically follow from the first two.

Saying something is a physical process is useless -- everything is a physical process. Information does not exist independent of a real physical substrate. Indeed, information is the physical substrate, just used a certain way.

So saying consciousness is purely a type of information processing doesn't mean it can exist independent of reality, or is not a real process, or anything of the sort. It simply means it is substrate independent at some level.

Whether or not something that behaves exactly like a biological neuron is a suitable substrate remains to be seen -- strong AI researchers don't think so, and neither do I.

But in any case, your argument here is defeated by the simple fact that both you and I exhibit consciousness, and we do not share the exact same particles. So at the very least there is that level of substrate independence.

Note that this goes for the entire class of things we label as a process -- the existence is independent of a specific system in which the process takes place. Any suitable system will do. The favorite example here on this forum is "running." Where is "running?" Nowhere -- it is a type of behavior that can be exhibited by many things. Does that make "running" non physical?

A perfect simulation of atoms and chemistry would easily simulate a fire. But that is not real fire. It cannot burn. A simulation of neurons would, at best, create a simulation of a conscious mind, but it would not be conscious (a real phenomenon) any more than the simulated fire would be a real fire, a physical phenomenon that burns real things.

The fire would be real in another frame. There is also no way for you to be sure that we are not in a simulation already.

I already had this argument with westprog. His absurd conclusion was that our consciousness is real if we are not in a simulation and not real if we are in a simulation. I happen to think that my consciousness is real regardless. What about you?

Note this also makes a side prediction that could, in theory, be tested some day: A perfect simulation of all known physics that had a "brain template" plugged into it, would, when "run", have one of three things happen:

1. It runs perfectly, and passes all Turing tests with flying colors. The consciousness is simulated ala the fire.

2. It doesn't run perfectly, or at all, because we've missed something in physics that's needed for true consciousness. Hence our simulation missed it, and no simulated consciousness arose*.

3. What won't happen is the simulation behaves perfectly because it generated a real-world consciousness to go hand-in-hand with its virtual data processing. This is also easier to understand when you realize that "information processing" has meaning only in terms of what the information represents. It's really just molecules and atoms and electrons and energies bouncing around in a complicated way that drives, somehow, creative outputs to varied inputs. But there's no "meaning" in reality beyond that.

You are missing the crucial point -- if you take this "simulated" consciousness out of the virtual world, and load it onto a substrate in our current world, it will behave the same way relative to the world it inhabits.

In fact, it would be enough to give the "simulated" consciousness mere access to our world. What happens when a virtual consciousness is able to access a non virtual world? The consciousness is no longer virtual, and since the property of being virtual is not dynamic, you can either say the consciousness is suddenly a different consciousness or else it was never virtual to begin with.

The same goes for all information. That is the great thing about information -- it doesn't matter where it comes from, because all information is frame independent.

 

[Gate2501]

Consiousness is biological, not mathematical.

Why could consciousness only emerge in a biological substratum? What would prevent other information processing mediums from giving rise to consciousness?

Is your argument something like: "I have only seen bananas on Palm Tree Island, therefore, it is impossible for bananas to grow anywhere else but Palm Tree Island."?

 

[rocketdodger]

Consiousness is biological, not mathematical.

What about running?

Do people with artificial legs no longer run, since those legs are no longer biological?

 

[HansMustermann]

My brain now has a bigger chance to exist... do you know how many neurons just gave up and went home just because I tried to imagine or put in perspective these numbers?

Actually, it's still the same probability. Just in that binary number with 10^15 digits, different bits are set, but it's still one of the combinations I was talking about

 

[Cosmic Roy]

First of all, I'd like to apologise in advance for the following post. I've read the thread, but I'm on the beer right now, and I'm finding it hard to concentrate.

rocketdodger, I really liked your post. I've thought about this a lot myself, though I've never tried (I wouldn't know how!) putting numbers to the idea. If I'm understanding you correctly, I think this concept was explored in a novel by Greg Egan: Permutation City - you may enjoy it if you haven't read it already.

To make sure I'm not misunderstanding you, I'd like to quickly state in my own words what I think you're driving at. Please correct me if I'm barking up the wrong tree.

You think that, if consciousness is nothing more than the processing of information, the interactions that cause it to arise can occur just as well in one medium as they can in another. So conscious experience would feel the same whether the computational substrate was a biological brain, a network of silicon chips, or the chance interactions of particles in space. So, given a universe extending an infinite time into the future, all possible computations will eventually be performed, and all possible conscious states will therefore be manifested - including those consistent with experience of an afterlife.

Now, I like this idea, but like I said, I wouldn't know how to estimate any given state's probability of occurring within a specified period of time. I appreciate HansMustermann's efforts in this direction.

I've read somewhere (I won't look it up now, sorry - my mind is too cloudy for that) that, if the universe was to end in a Big Crunch, the rate of computation would increase as the universe became hotter and denser, meaning any consciousness manifested by those computations could subjectively live an infinite amount of time (or a near-infinite amount - if that means anything!) even as their computational substrate collapsed rapidly to a singularity. Could the opposite also work? That is, could it be the case that, in the hot, dense state that existed soon after the Big Bang, random interactions occurred at such a rate that the likelihood of a configuration that simulated a consciousness would be much increased?

I'm going to stop writing now. Please remember that the contents of this post do not necessarily reflect the views of the poster.

 

[realpaladin]

First of all, I'd like to apologise in advance for the following post. I've read the thread, but I'm on the beer right now, and I'm finding it hard to concentrate.
<snip>
I'm going to stop writing now. Please remember that the contents of this post do not necessarily reflect the views of the poster.

*BAM* Now the extended families of aforementioned neurons have slammed the door whilst shouting 'I quit!'.

But a cool notion though.

 

[realpaladin]

Actually, it's still the same probability. Just in that binary number with 10^15 digits, different bits are set, but it's still one of the combinations I was talking about

You are trying to make me think about them numbers again, don't you?

 

[Thabiguy]

The question is, from the moment the universe was born to the moment it dies, will there be any set of information isomorphic to the set on your hard drive (other than the drive itself, of course) ?

I'm afraid you're misusing the word "isomorphic" here. Isomorphism is something that applies to algebraic structures. The contents of a hard drive is not an algebraic structure, it's simply a natural number, X.

If you meant to ask, "will there be anything else that represents the number X in some encoded form", the answer is, of course. Everything represents the number X in some encoded form. It's just a question of choosing a proper decoding function.

 

[Cosmic Roy]

I'm afraid you're misusing the word "isomorphic" here. Isomorphism is something that applies to algebraic structures. The contents of a hard drive is not an algebraic structure, it's simply a natural number, X.

If you meant to ask, "will there be anything else that represents the number X in some encoded form", the answer is, of course. Everything represents the number X in some encoded form. It's just a question of choosing a proper decoding function.

You may be accustomed to using the term 'isomorphism' in the context of algebraic structures. But surely, 'isomorphic' means something along the lines of 'of the same shape' (I don't know Latin, sorry), and is thus more flexible than you claim.

Edited to add:
Yes, I see that the term 'isomorphism' is used in biology and geology as well as mathematics.

 

[Thabiguy]

But surely, 'isomorphic' means something along the lines of 'of the same shape' (I don't know Latin, sorry), and is thus more flexible than you claim.

The only relevant issue here is what rocketdodger meant by "isomorphic to the set of information on a hard drive".

If you rotate the data one bit to the left, is the resulting image "isomorphic" to the original (since you can always restore the original by rotating it back)?

If you flip the first bit of the data, is the resulting image "isomorphic" to the original (since you can always restore the original by flipping the bit back)?

If the answer to both question is yes, and this "isomorphism" is transitive (if A is "isomorphic" to B, and B to C, then A is "isomorphic" to C), then any hard drive contents is "isomorphic" to any other contents of that hard drive.

 

[Gate2501]

The contents of a hard drive is not an algebraic structure, it's simply a natural number, X.

Where are the contents of a hard drive expressed as a natural number, X? I mean, technically you could take ANY non variable phenomena, and quantify it as a number, X, right?

 

[Cosmic Roy]

The only relevant issue here is what rocketdodger meant by "isomorphic to the set of information on a hard drive".

If you rotate the data one bit to the left, is the resulting image "isomorphic" to the original (since you can always restore the original by rotating it back)?

If you flip the first bit of the data, is the resulting image "isomorphic" to the original (since you can always restore the original by flipping the bit back)?

If the answer to both question is yes, and this "isomorphism" is transitive (if A is "isomorphic" to B, and B to C, then A is "isomorphic" to C), then any hard drive contents is "isomorphic" to any other contents of that hard drive.

I concede, I feel blinded by technical terminology. My understanding of rocketdodger's use of the word is that he meant something like 'embodying the same information'. I don't know nearly enough about algebra or computer science to say whether his use of the word was justified. But I can say that, given that the function of language is to convey concepts, rocketdodger's language reached at least one person (if I've interpreted his concept correctly, of course).

 

[Gate2501]

I concede, I feel blinded by technical terminology. My understanding of rocketdodger's use of the word is that he meant something like 'embodying the same information'. I don't know nearly enough about algebra or computer science to say whether his use of the word was justified. But I can say that, given that the function of language is to convey concepts, rocketdodger's language reached at least one person (if I've interpreted his concept correctly, of course).

I think that RD was talking about the more general definition of Isomorphism that deals with 2 different objects being identical on some level of abstraction. You could apply this more general definition to an object like a hard drive, which you could ALSO apply the strict mathematical definition to in a different way.

Thus, a semantic argument ensues.

 

[Cosmic Roy]

I think that RD was talking about the more general definition of Isomorphism that deals with 2 different objects being identical on some level of abstraction. You could apply this more general definition to an object like a hard drive, which you could ALSO apply the strict mathematical definition to in a different way.

Thus, a semantic argument ensues.

Indeed! Let's not start a semantic argument, and instead let rocketdodger come and tell us what he meant by it. In the meantime, I will go to the pub.

 

[Thabiguy]

Where are the contents of a hard drive expressed as a natural number, X? I mean, technically you could take ANY non variable phenomena, and quantify it as a number, X, right?

I guess you could, but that's not what I was getting at. I was merely referring to the simple fact that a hard drive is a device designed to store a very long binary number - these days, usually on the order of hundreds of billions of binary digits (a.k.a. bits) - and allow the drive controller to read and modify portions of its binary representation. A binary number is a natural number.

I concede, I feel blinded by technical terminology.

Sorry about that. What I wrote was really aimed at rocketdodger, whom I expect to understand the terminology.

 

[Alkatran]

Premise 2) -- if another pattern of information P is isomorphic to E(t) under R, then C(t) will be instantiated by P.

I don't think this premise is valid. It certainly goes against how we perceive consciousness. I certainly perceive it as localized to my own body, so how could some other body existing at the same time have the same consciousness just because they were thinking the same thing?

Also, you haven't explicitly defined what isomorphism you want to use. Same state? Same operations to be performed on states? Same state and operations to be performed on states? These make a big difference! There's also the issue that you can turn operations to be performed into state (ie. a computer program instead of a hard-wired computer).

Premise 3) -- for every t, there may be some n, at any given time, such that P is isomorphic to E(t) under R. Extended to infinity, we can assume that there will be some such n.

This is just plain wrong. You wave away how unlikely is that two processes will be identical, and then you make a classic "infinity -> whatever the hell I want" mistake.

Just because a system can continue indefinitely does not mean that it will visit every possible state. For example it might get stuck in a small loop, or converge on some state (ie. the heat death of the universe).

 

[AkuManiMani]

I think its a bit premature to propose a likely mechanism for continuing life after death. Even if some part of us does continue on after we die we've no way of knowing right now if we'll just be some persistent memory trace or if whats left will still even carry our self identity >_<

 

[Gate2501]

I don't think this premise is valid. It certainly goes against how we perceive consciousness. I certainly perceive it as localized to my own body, so how could some other body existing at the same time have the same consciousness just because they were thinking the same thing?

Let's examine this idea.

Our brains are physical, and the information processing that goes on within them, causes consciousness and a sense of self awareness to emerge. Any consciousness *C* then, should be able to be duplicated given sufficient understanding of what is going on in the brain, and the ability to mirror it. Consciousness C could logically exist in two places at one time.

The problem with this, is that C has come about as a result of the evolution of a biological organism over a buttrillion years. C is hard wired to think that locality is VERY important, and to see itself as a unique identity. Each instance of C, would think that it was the "real" C. However, both would be the real C.