My take on why indeed the study of consciousness may not be as simple

[drkitten]

IOW, what defines information processing is the use made of the information by a conscious human being. If there's no human being using the information, it isn't being processed.

Not at all. It has nothing to do with the "conscious human being" and everything to do with whether or not the change in information is equivalent to the changes in a Turing Machine or equivalent.

As you can see by the fact that autonomous robots process information that they do not deliver to human beings, but rocks just sit there.

 

[drkitten]

Of course there are, because we can define the state transitions to be anything we want.

Well, if you can find a way to define a set of internal states and transitions (and so forth) that make a rock TM-equivalent, then you will have proven that a rock is indeed a computer.

That will, of course, be one of the biggest breakthroughs in computer science since the development of TM-equivalence itself, since it means that we need no longer pay huge sums of money to Intel; anything we want to do we can just do by picking up a rock from the garden.

 

[drkitten]

What we're looking for is a definition of computing (or computation, if you prefer) which is physical, and is defined for all possible forms of computing.

No. What you're looking for is a definition of computing which is physical.

Which is roughly equivalent to looking for a Honda Accord that is powered by magic pixies. You're taking a well-understood problem, and adding an unrelated and unreasonable requirement that every practicing engineer would reject in an attempt to prove that there is no such thing as a Honda Accord or that there exists no equivalence class "cars" under which a Honda Accord is equivalent to any other car on the road.

 

[drkitten]

I find it hilarious that DrKitten is proposing that mathematics is really physics

That's only because you still don't understand how mathematics describes physics.

at the same time insist that a mathematician can't be a physicist.

I don't. I know lots of mathematicians who are physicists; I knew someone in grad school who's now a dual PhD in both math and physics, for example.

I insist that Prof Penrose is not a physicist.

It's fairly impressive when you manage to pack two misrepresentations into the same post.

 

[rocketdodger]

Of course there are, because we can define the state transitions to be anything we want.

No, you can't.

You can define the states to be whatever you want. A "transition" is simply a change from one state to another.

For instance, you could say that a rock that is solid is one state, and a rock that is melted is another. And then you could say that a rock that melts computes because numerous external state transitions (numerous things can change external temperature) result in the same single transition within the rock -- the transition from solid to liquid.

Except that doesn't occur constantly. And your claim is that everything computes all the time, not just that everything can compute. Nobody disputes that a pile of parts does the same thing as a laptop that is not computing. The dispute is whether or not a pile of parts does the same thing as a laptop that is switched on and ... computing.

So... the rock, westprog? Examples of state transitions?

 

[westprog]

No. What you're looking for is a definition of computing which is physical.

No, I don't believe there is any such thing. Nor do I believe that it is necessary or appropriate for the field of computing. As far as I'm concerned, computing is a mathematical process, which people create devices to implement. It is not a physical process at all.

This only becomes relevant because of the claim that there is an emergent property of consciousness which happens when computation takes place. There have been different views about this. Some people say that consciousness only becomes relevant at a particular scale. Others that it is present even in a single switch, and it's a matter of aggregating the little bits of consciousness into one glorious whole.

So AFAICT it's implicit in this claim that computation is a physical process, and that it produces consciousness as a physical effect. If this is the claim, then the minimum, first requirement is a physical definition of what computing actually is.

Which is roughly equivalent to looking for a Honda Accord that is powered by magic pixies. You're taking a well-understood problem, and adding an unrelated and unreasonable requirement that every practicing engineer would reject in an attempt to prove that there is no such thing as a Honda Accord or that there exists no equivalence class "cars" under which a Honda Accord is equivalent to any other car on the road.

 

[drkitten]

No, you can't.

You can define the states to be whatever you want. A "transition" is simply a change from one state to another.

For instance, you could say that a rock that is solid is one state, and a rock that is melted is another. And then you could say that a rock that melts computes because numerous external state transitions (numerous things can change external temperature) result in the same single transition within the rock -- the transition from solid to liquid.

Except that doesn't occur constantly.

You're barking up the wrong tree. Even a Turing machine doesn't need to be computing "constantly"; it's perfectly legitimate to execute one transition every ten minutes and then just sit there.

It's a more general problem. For example, if you've defined the rock as having two states (molten and solid), those two states are not, by themselves, sufficient to act as a Turing machine.
You would need some way of defining input strings, output strings, and a suitable storage mechanism (tape or tape-equivalent). And in particular, you'd need to show how a particular set of transitions from molten to solid and back acted in a way to operate on the storage mechanism in a manner equivalent to TM storage.

Simply saying "a rock has states, a TM has states, therefore they are equivalent" is not valid. A TM is actually a rather precisely-defined set of (mathematical) properties, properties that a rock appears not to have.

So... the rock, westprog? Examples of state transitions?

 

[drkitten]

No, I don't believe there is any such thing. Nor do I believe that it is necessary or appropriate for the field of computing. As far as I'm concerned, computing is a mathematical process, which people create devices to implement. It is not a physical process at all.

And as a mathematical process, it is instantiated by any physical process that obeys the mathematical strictures. Just like the mathematical concept of "two" is instantiated by the moons of Mars or a pair of apples, without any necessary physical relationship between the two.

This makes "two apples" a physical instantiation of the mathematical concept of "two."

Just as a (physical) Intel Pentium (as well as a Univac) is a physical instantiation of the mathematical concept of "computing." When properly configured and powered, of course. Without power, the Pentium isn't doing what a "computer" needs to do to be Turing-equivalent. (And hence the separation between "a computer" and "a pile of computer parts".)

This only becomes relevant because of the claim that there is an emergent property of consciousness which happens when computation takes place.

Exactly. The "emergent property" is the mathematical property of being Turing equivalent, emergent because no component has that property, any more than either of the apples instantiates "two" by itself.

So AFAICT it's implicit in this claim that computation is a physical process,

Well, then you can't tell very well. Which is what you've been told since you first posted on this thread. You don't understand how physical systems can instantiate mathematical abstractions, or how mathematical abstractions can have physical consequences, or how computation is defined as a mathematical abstraction despite being instantiated in an underlying physical substrate.

With that much lack of understanding, it's hardly surprising that you're completely and utterly wrong in your writings on AI and consciousness.

 

[westprog]

That's only because you still don't understand how mathematics describes physics.

I've explained how mathematics describes physics. The idea that mathematics is physics is entirely yours.

I don't. I know lots of mathematicians who are physicists; I knew someone in grad school who's now a dual PhD in both math and physics, for example.

I insist that Prof Penrose is not a physicist.

And you continue to do so in spite of the fact that it's been pointed out to you that

• Penrose has produced theories in physics
• he has published in journals of physics
• he has been responsible for the discovery of physical laws
• he teaches physics
• he is a professor of physics
• he has published a textbook of physics

So your nonsensical insistence that he is not a physicist is clearly without any sensible basis. If you want to point out that you are only being inconsistent in the specific case of Professor (Distinguished Professor of Physics and Mathematics at Syracuse University, and Distinguished Professor of Physics and Mathematics at Pennsylvania State University) Penrose, without giving any justification for your ridiculous contention, then I hardly think it's surprising if I find it difficult to figure out just what it is you do think.

It's fairly impressive when you manage to pack two misrepresentations into the same post.

 

[drkitten]

I've explained how mathematics describes physics.

Wrongly.

You can also describe how fairies make bread rise. That doesn't mean your explanation is correct.

The idea that mathematics is physics is entirely yours.

And you continue to do so

I do. Because he is not a physicist.

 

[Robin]

Another 2 observation about consciousness right there.
We as conscious minds talk about not meaningful as if it "is" something meaningful to have no meaning and saying which has no content but was said.

You are confusing a statement about language with a statement about consciousness.

In any case are you certain that this is an observation about consciousness? Are you certain that associating meaning with a word is a conscious process?

If not then why do you think it is an observation about consciousness?

No , but I have yet to understand a proof whilst unconscious.

I have yet to understand how consciousness can prove something.

So what does one thing have to do with the other then?

You brought up using circular definitions as proof, not me.

No, I brought up that circular definitions do not prove anything.

As I said before, I made my understanding of your words clear.

If you did not correct me then that is not my fault.

Metaphysical conclusions can only be drawn from metaphysical assumptions.
I do not see how thinking can be regarded as a metaphysical assumption as any metaphysics pre-supposes thinking.

The assumption "metaphysics pre-supposes thinking" is in itself a metaphysical assumption.

Now I know you are going to take this as further evidence that thinking is self-referential, but it is not. If anything it is evidence that metaphysics is self-referential.

 

[Robin]

Choice.

If you are going to make the claim that choice is a method of study then you should justify it.

Because you are beginning to sound very evasive on this subject.

 

[rocketdodger]

You're barking up the wrong tree. Even a Turing machine doesn't need to be computing "constantly"; it's perfectly legitimate to execute one transition every ten minutes and then just sit there.

That is fine, because we can just say that if a system is between two transitions, it is also computing. So as long as our TM executes more than a single transition, the definition is consistent.

On the other hand, if westprog wants to say that a rock computes (not in the TM sense, "switches" would be a better word) because it *might* satisfy the definition, some time or another, then he can do so.

But that is just semantic games, and if he wants to play that game all we have to do is come up with yet another word for "active" computation or somthing, and so on and so forth.

It's a more general problem. For example, if you've defined the rock as having two states (molten and solid), those two states are not, by themselves, sufficient to act as a Turing machine.
You would need some way of defining input strings, output strings, and a suitable storage mechanism (tape or tape-equivalent). And in particular, you'd need to show how a particular set of transitions from molten to solid and back acted in a way to operate on the storage mechanism in a manner equivalent to TM storage.

Simply saying "a rock has states, a TM has states, therefore they are equivalent" is not valid. A TM is actually a rather precisely-defined set of (mathematical) properties, properties that a rock appears not to have.

Yeah I know, and I am not trying to show that.

I was trying to get at the fundamental physical difference between systems that we consider necessary for computing (informally, switches) and all other systems.

You could, if you were very clever, build a TM equivalent computer from a bunch of switches consisting of rocks + other stuff.

If I can show that switching (I have been calling that "computation," when I mention "computation" I am not speaking of the formal TM equivalent sense) has a physical definition then it is trivial to show that all the rest of computation theory can be physical as well -- all one needs to do is instantiate the abstract. But westprog wants that first step, so I have been trying to provide it for him.

 

[Dancing David]

OK. How are you defining computing? Arithmetic? The "Principles of Operations" manual for z/OS describes how arithmetic operations work for the z/OS architecture. In physical terms there are special areas in a CPU called the PSW (where the current instruction is located) and there are registers which point to other areas of memory. These are treated as hexadecimal values which are added for example using AND, XOR etc gates in the hardware.
Any better?
How about "It's complicated"?

Good luck, Westprog seems to want to use private idiomatic language which makes discussion difficult.

 

[RandFan]

Then again you could just ignore consciousness as Randfan/Pixy Misa/Joe TheJuggler/Rocketdodge/PingofPong do and announce that science will give us the proper definition of consciousness once we have created it outside ourselves.

Then again you could simply create straw men.

This is the metaphysical proposition that science exists a priori to consciousness. The metaphysics only becomes apparent when you start asking how much of "something" do we need to create this consciousness and you are told consciousness does not have to do with "something".

This is the metaphysical proposition that science exists a priori to flight. The metaphysics only becomes apparent when you start asking how much "something" do we need to create flight?

!Kaggen, your proposition is utterly and demonstrably absurd.

There may very well be a threshold for consciousness. I suspect there is but our ignorance is no basis for any conclusion. You ARE arguing from ignorance. IOW: You ARE drawing conclusions from that which we do not know.

 

[Robin]

Uh-uh. The act of shopping might involve actual transfer of physical material, but it might just involve a conceptual exchange of rights.

And is an "exchange of rights" not a process built from other, more primitive physical processes?

Try to define it in physical terms and it will evaporate from you.

So in other words you are saying that some processes which are built from other more primitive physical processes are not physical?

If so that would change your definition.

Because you cannot narrow it down to anything specifically physical beyond the fact that some physical action probably needs to take place.

I am not narrowing it down to the fact that some physical action probably needs to take place.

I am narrowing it down to the fact that some physical action definitely needs to take place.

If I can't identify anything else that needs to take place then I have narrowed it down to something specifically physical and it meets your definition.

So it all reduced to "matter, mass, forces, etc..".

If we only knew what those things are.

 

[quarky]

To study one's own consciousness has no scientific validity, even though one might study their hand in an acceptable manner. There is accumulated data concerning the anecdotal claims from various spiritual or religious disciplines regarding the inner study of consciousness. Bhuddists come to mind. Can we disregard all of that, as if it isn't a valid experiment? What takes place in the mind of someone skilled at sitting, that hasn't moved in hours, yet remains alert?

The feeling is that consciousness is the background state. Awareness is the default position. Our brains intervene like radios or i-phones, interpreting the perceptual data in a fitting way for our survival and reproduction. There is no evidence, but there is evidence that our reality is becoming more infused with invizable waves carrying data.

The metaphor is there, anyway, and it wasn't 100 years ago.

 

[PixyMisa]

Uh-uh. The act of shopping might involve actual transfer of physical material, but it might just involve a conceptual exchange of rights. Try to define it in physical terms and it will evaporate from you.

Wrong.

Everything that happens is physical.

That's not to say that shopping doesn't involve physical action, but delimiting what those actions might be is impossible.

Proof?

Because you cannot narrow it down to anything specifically physical beyond the fact that some physical action probably needs to take place.

Probably? Probably? How can it happen without any physical action taking place? How can anything happen without any physical action taking place?

 

[PixyMisa]

To study one's own consciousness has no scientific validity, even though one might study their hand in an acceptable manner.

I think you mean, to study one's own consciousness through introspection (and introspection alone). Certainly you can climb into an FMRI or run TMS experiments on yourself; it's just kind of awkward.

There is accumulated data concerning the anecdotal claims from various spiritual or religious disciplines regarding the inner study of consciousness. Bhuddists come to mind. Can we disregard all of that, as if it isn't a valid experiment? What takes place in the mind of someone skilled at sitting, that hasn't moved in hours, yet remains alert?

What you're talking about there is the distinction between attention and arousal, something that is found well beyond the sphere of buddhism. Introspection is, as you say, anecdotal; it can provide a starting point for real research, but certainly not an ending point.

The feeling is that consciousness is the background state.

Which is clearly not the case.

Awareness is the default position.

Which is clearly not the case.

Our brains intervene like radios or i-phones, interpreting the perceptual data in a fitting way for our survival and reproduction.

I'm not sure what that is supposed to mean.

There is no evidence, but there is evidence that our reality is becoming more infused with invizable waves carrying data.

True, but certainly not relevant.

The metaphor is there, anyway, and it wasn't 100 years ago.

New metaphors don't always contribute to our understanding of that which they are a metaphor for.

 

[!Kaggen]

Then again you could simply create straw men.

This is the metaphysical proposition that science exists a priori to flight. The metaphysics only becomes apparent when you start asking how much "something" do we need to create flight?

!Kaggen, your proposition is utterly and demonstrably absurd.

There may very well be a threshold for consciousness. I suspect there is but our ignorance is no basis for any conclusion. You ARE arguing from ignorance. IOW: You ARE drawing conclusions from that which we do not know.

Randfan, I do realize this is a skeptics forum and I appreciate criticisms, as any skeptic should. However my approach is specifically aimed at drawing attention to observations not conclusions using a phenomenological approach. I prefer to leave the conclusions to further scrutiny by the scientific method of (observation), induction, deduction and testing. From my experiences in the academic side and the practical side of agriculture I have concluded that the gap between the two is frightening (sorry cannot think of a justified non-emotional description). Further thought has led me to believe that the reason for this gap is purely down to observational skills on both sides and hence my interest in phenomenology as a tool to develop these skills.

Academics ask only what conclusions can we draw and farmers ask what practical relevance are these conclusions. The approach of both is very much goal driven and I am not critical of having a goal only I have seen how this can blind one to the process of achieving that goal resulting in unnecessary and unforeseen consequences. The argument that science is self correcting and therefore all we need to do is keep at it is true, but improving our observational skills as scientists is not inconsistent with this view. After all we do have history of recognizing better "scientists". We have learn't how to make better athletes by developing "sports psychology" why not "scientist psychology".

The problem is that our language is too abstract to describe the observations resulting from the phenomenological approach the result being that these observations might read like conclusions, especially by a scientist and hence your scrutiny.