Has consciousness been fully explained?

[westprog]

Obviously it's not the thing itself, it's a simulation...

And a simulation is not expected to produce real effects.

Thought experiment, remember?

There's a big difference between a thought experiment which says "what if we had the capacity to build a great big machine that could do this" and one that says "what if the laws of nature were not as they are, and we could do this."

I've no problem with postulating a computer of potentially unlimited power - but it should be recognised that as far as we now understand the universe, it's impossible for any computational device to simulate the universe, even in principle. That limits the usefulness of the thought experiment considerably. The universe is not computational, as far as we are able to detect. There's no reason to suppose that it will turn out to be even if we learn more.

 

[Ichneumonwasp]

And a simulation is not expected to produce real effects.

Yet, is a simulated person performing a calculation calculating or not? If we simulate someone adding two plus two, doesn't the action of adding actually occur?

There's a big difference between a thought experiment which says "what if we had the capacity to build a great big machine that could do this" and one that says "what if the laws of nature were not as they are, and we could do this."

I've no problem with postulating a computer of potentially unlimited power - but it should be recognised that as far as we now understand the universe, it's impossible for any computational device to simulate the universe, even in principle. That limits the usefulness of the thought experiment considerably. The universe is not computational, as far as we are able to detect. There's no reason to suppose that it will turn out to be even if we learn more.

I don't understand, as a thought experiment, whether it can occur in the real world or not has any impact on this. Why does this limit the usefulness of the thought experiment?

Einstein couldn't actually travel along with a beam of light, even in theory, but did this limit the usefulness of his thought experiment?

 

[Myriad]

A "flying machine", by definition, flies. Unless it's broken. It doesn't simulate flight: it flies.

Fine. And a conscious machine, by definition, is conscious. Unless it's broken. It doesn't simulate consciousness; it experiences consciousness.

The claim being made by computationalists is that simulated flight is "real in the simulated world", but not real in "our world". Some of us see that as dualism. A machine that actually flies isn't a simulation.

No one is making that claim. Simulation of flight is a real thing in the real world. Flight simulators are made of matter. You can kick them and everything. Real pilots use them to really learn things. Simulation of flight is only flight in the sense of that being a description of the behavior of the state of the simulation (which is what we mean by "in the simulated world") but the existence of a state and the behavior of that state are also real phenomena in the real world.

I think conscious machines are possible. If you slowly replace neurons with transistors, I don't see why consciousness would disappear.

"I think flying machines are possible. If you slowly replace a bird's feathers, bones, muscles, and other parts with composite materials and miniature actuators, I don't see why flight ability would disappear."

The difference is, after replacing neurons with equivalent neuron-shaped transistorized devices, we would not have to stop there. Since electronic impulses along wires are faster than action potentials, we can do more replacements.

For instance, we can replace (not simulate, replace) the system of interconnections between neurons, with an internal list of its connections and a database in a sufficiently fast external computer. Each neuron (for now) stays where it is, but instead of having direct connections with the other neurons on its list, it now reports its state through a wire to the database and it receives the data for the state of each of its connections through another wire. So it receives and processes the same information as before.

Now we can replace (not simulate) the neurons' internal lists of connections, and instead, have the database keep track of that as well. The neuron still receives and processes the same information as before.

Also, since the neurons are no longer physically connected to one another, but instead to the database, we can rearrange the neurons spatially however we wish (as long as signal propagation times are taken into account and suitably adjusted as needed). Each neuron still receives and processes the same information as before.

But, if our database system can handle the load, we can replace (not simulate) much of the input processing the neuron performs on the information it receives, by having the database system pre-process the data it's communicating to the neuron to perform the same computation -- so that instead of communicating all the information about all the neuron's inputs, it only tells the neuron, for example, the weighted sum. The neuron now merely has to perform its output processing (e.g. comparing the weighted sum to its current threshold, and updating its history state that determines that threshold) and telling the database when it fires.

But, at this point it makes more sense to replace (not simulate) the remainder of the neuron's behavior, by having the database system take that over as well. So, instead of sending processed neural input information to the neuron and receiving the neuron's state information back, it adds the rest of the processing the neuron was doing, to its own processing task list instead. That adds more processing work, but reduces the communications work.

So now we have replaced -- NOT SIMULATED -- the transistorized neural brain with a computer system that does not use or need neurons at all, and that can be architected in whatever way we wish (or will work most efficiently). Does the computer have a vast array of small processors each doing the work that was formerly done by a single neuron, or is each neuron's state just represented by some data in some area of memory, with just one ultra-powerful CPU doing all the processing? It doesn't matter, except insofar as practical design issues are concerned.

But this is much different than the claims that simulated consciousness is actual consciousness.

Correct. That is a different kind of claim, that relates to a different line of argument.

However, only some concepts of conscious machines are reasonable to describe as simulated consciousness, and those concepts exist only as an existence argument. A conscious machine need not be a simulated brain, though that (as for the machine I just described) could be one path toward designing one.

If to achieve a flying machine we had to mimic every characteristic of a bird, we would probably not have managed it yet. But, even before airplanes were invented, we could answer a philosophical argument that artificial flying machines are inherently impossible with a philosophical thought experiment, "suppose we built a machine that perfectly replicated the relevant characteristics of a bird; why would we not expect it to fly?" The answer "That would only be a simulated bird, so it couldn't really fly in the real world" would make no sense. Neither does the claim that a machine that perfectly replicated the functioning of a brain, such as the one I just described, would be only a mere simulation of a brain, so it couldn't really be conscious in the real world.

If a machine replicated a human brain entirely, it should be conscious. If it weren't I don't know if "magic" would enter the picture. It could be that consciousness is limited to biological creatures in some strange way we don't yet understand.

"If a machine replicated a bird entirely, it should be able to fly. If it weren't I don't know if 'magic' would enter the picture. It could be that flying is limited to biological creatures in some strange way we don't yet understand."

Yeah, maybe. I'm not ruling it out. It could have been true for birds and flying, too, but it wasn't. Could have been true for any number of other characteristics of biological creatures as well, but it wasn't. It turns out, plants create substance out of molecules from the air, water, and soil rather than from nothing, muscles move by forces applied by the electromagnetic interactions of matter rather than from the force of will, and diseases cause illness by mere chemical and physical interaction with living tissue, rather than by the spiritual effects of sin.

Why do you seem to expect a different kind of answer, for consciousness? I can see not ruling out the possibility in principle, but why do you seem to think it's the more likely possibility? That's not a rhetorical question. I'm really curious. Is it because conscious experience "feels" so different than other things in nature (if such a comparison could even be made)?

Respectfully,
Myriad

 

[westprog]

Yet, is a simulated person performing a calculation calculating or not? If we simulate someone adding two plus two, doesn't the action of adding actually occur?

No. No more than when two rocks fall down a cliff.

I don't understand, as a thought experiment, whether it can occur in the real world or not has any impact on this. Why does this limit the usefulness of the thought experiment?

Einstein couldn't actually travel along with a beam of light, even in theory, but did this limit the usefulness of his thought experiment?

Einstein's experiment was useful in determining what wasn't possible.

 

[westprog]

Fine. And a conscious machine, by definition, is conscious. Unless it's broken. It doesn't simulate consciousness; it experiences consciousness.

And a perpetual motion machine, by definition, carries on running forever. Which doesn't demonstrate that such a thing can exist.

The argument is not that a conscious device is inherently impossible. ( I think perhaps some people have doubts that it is, which is a different thing). What's been disputed is the nature of such a device.

 

[westprog]

For instance, we can replace (not simulate, replace) the system of interconnections between neurons, with an internal list of its connections and a database in a sufficiently fast external computer. Each neuron (for now) stays where it is, but instead of having direct connections with the other neurons on its list, it now reports its state through a wire to the database and it receives the data for the state of each of its connections through another wire. So it receives and processes the same information as before.

Now we can replace (not simulate) the neurons' internal lists of connections, and instead, have the database keep track of that as well. The neuron still receives and processes the same information as before.

Also, since the neurons are no longer physically connected to one another, but instead to the database, we can rearrange the neurons spatially however we wish (as long as signal propagation times are taken into account and suitably adjusted as needed). Each neuron still receives and processes the same information as before.

But, if our database system can handle the load, we can replace (not simulate) much of the input processing the neuron performs on the information it receives, by having the database system pre-process the data it's communicating to the neuron to perform the same computation -- so that instead of communicating all the information about all the neuron's inputs, it only tells the neuron, for example, the weighted sum. The neuron now merely has to perform its output processing (e.g. comparing the weighted sum to its current threshold, and updating its history state that determines that threshold) and telling the database when it fires.

But, at this point it makes more sense to replace (not simulate) the remainder of the neuron's behavior, by having the database system take that over as well. So, instead of sending processed neural input information to the neuron and receiving the neuron's state information back, it adds the rest of the processing the neuron was doing, to its own processing task list instead. That adds more processing work, but reduces the communications work.

So now we have replaced -- NOT SIMULATED -- the transistorized neural brain with a computer system that does not use or need neurons at all, and that can be architected in whatever way we wish (or will work most efficiently). Does the computer have a vast array of small processors each doing the work that was formerly done by a single neuron, or is each neuron's state just represented by some data in some area of memory, with just one ultra-powerful CPU doing all the processing? It doesn't matter, except insofar as practical design issues are concerned.

Except that we don't, in fact, understand the functionality of even a single neuron to the extent that we can replace it with an artificial substitute. We can't repair nerve damage.

What it comes down to is that if we understood the function of neurons totally, and we could produce an artificial substitute, then we could gradually replace the entire brain with such an artificial system and it would work just the same. That's only a powerful concept if we actually know precisely what the essential functionality is. We certainly don't know enough to be able to describe the precise nature of such an artificial system. If we did, we'd be able to make people with spinal injuries walk again.

 

[westprog]

Why do you seem to expect a different kind of answer, for consciousness? I can see not ruling out the possibility in principle, but why do you seem to think it's the more likely possibility? That's not a rhetorical question. I'm really curious. Is it because conscious experience "feels" so different than other things in nature (if such a comparison could even be made)?

Respectfully,
Myriad

Something like "flight" is relatively easy to define, if not to achieve. Even if we'd never seen anything fly, we could easily describe what would need to happen for it to take place. Consciousness is something which we can experience, but not describe. Hence even knowing whether it's happening is problematic. We cannot measure it, and the only way we have of guessing that it exists is that other people look and behave like us.

 

[Malerin]

A "flying machine", by definition, flies. Unless it's broken. It doesn't simulate flight: it flies.

Fine. And a conscious machine, by definition, is conscious. Unless it's broken. It doesn't simulate consciousness; it experiences consciousness.

Right, and such a machine doesn't exist and we have no idea if it will ever exist. Anyway, such a machine would not be a simulation.

The claim being made by computationalists is that simulated flight is "real in the simulated world", but not real in "our world". Some of us see that as dualism. A machine that actually flies isn't a simulation.

No one is making that claim.

That's exactly what they're saying:

Water in a simulation is real water in the simulation. It isn't the same as water 'out here'.

Simulation of flight is a real thing in the real world. Flight simulators are made of matter.

Flight simulators are real. "Simulation of flight" is not a real thing.

You can kick them and everything. Real pilots use them to really learn things. Simulation of flight is only flight in the sense of that being a description of the behavior of the state of the simulation (which is what we mean by "in the simulated world") but the existence of a state and the behavior of that state are also real phenomena in the real world.

Exceptt when you simulate it. Then it becomes "real in the simulated world", and we're back to dualism.

I think conscious machines are possible. If you slowly replace neurons with transistors, I don't see why consciousness would disappear.

"I think flying machines are possible. If you slowly replace a bird's feathers, bones, muscles, and other parts with composite materials and miniature actuators, I don't see why flight ability would disappear."

Note the difference: flying machine (i.e. something that flies) and flight simulator (i.e., something that doesn't fly).

The difference is, after replacing neurons water molecules with equivalent neuron-shaped water molecule-shaped transistorized devices, we would not have to stop there.

For instance, we can replace (not simulate, replace) the system of interconnections covalent bonds between neurons hydrogen and oxygen atoms....

With enough replacements, you would have a simulated glass of water. You would not be able to drink it.

But this is much different than the claims that simulated consciousness is actual consciousness.

Correct. That is a different kind of claim, that relates to a different line of argument.

However, only some concepts of conscious machines are reasonable to describe as simulated consciousness, and those concepts exist only as an existence argument. A conscious machine need not be a simulated brain, though that (as for the machine I just described) could be one path toward designing one.

If to achieve a flying machine we had to mimic every characteristic of a bird, we would probably not have managed it yet. But, even before airplanes were invented, we could answer a philosophical argument that artificial flying machines are inherently impossible with a philosophical thought experiment, "suppose we built a machine that perfectly replicated the relevant characteristics of a bird; why would we not expect it to fly?" The answer "That would only be a simulated bird, so it couldn't really fly in the real world" would make no sense. Neither does the claim that a machine that perfectly replicated the functioning of a brain, such as the one I just described, would be only a mere simulation of a brain, so it couldn't really be conscious in the real world.

It wouldn't be a simulated bird because the machine could actually fly. Very important distinction. A simulation of a bird can't fly. It just sits there, simulating flight.

If a machine replicated a human brain entirely, it should be conscious. If it weren't I don't know if "magic" would enter the picture. It could be that consciousness is limited to biological creatures in some strange way we don't yet understand.

"If a machine replicated a bird entirely, it should be able to fly. If it weren't I don't know if 'magic' would enter the picture. It could be that flying is limited to biological creatures in some strange way we don't yet understand."

Yeah, maybe. I'm not ruling it out. It could have been true for birds and flying, too, but it wasn't. Could have been true for any number of other characteristics of biological creatures as well, but it wasn't. It turns out, plants create substance out of molecules from the air, water, and soil rather than from nothing, muscles move by forces applied by the electromagnetic interactions of matter rather than from the force of will, and diseases cause illness by mere chemical and physical interaction with living tissue, rather than by the spiritual effects of sin.

Why do you seem to expect a different kind of answer, for consciousness? I can see not ruling out the possibility in principle, but why do you seem to think it's the more likely possibility? That's not a rhetorical question. I'm really curious. Is it because conscious experience "feels" so different than other things in nature (if such a comparison could even be made)?

Conscious experience is unlike anything else in the physical universe.

 

[Myriad]

Except that we don't, in fact, understand the functionality of even a single neuron to the extent that we can replace it with an artificial substitute. We can't repair nerve damage.

The concept of gradually replacing neurons in the brain with artificial substitutes was introduced on the thought-experiment level by Malerin to whom I was responding. I accepted that as a premise for discussion, not as a practical plan.

Of course we can't manufacture artificial neurons, nor surgically implant them if we could. You know what else we can't do? We can't manufacture feathers from scratch. Not a single one. So, does that mean we have no hope of understanding the principles of aerodynamics and structural engineering sufficient to permit artificial flying machines? After all, maybe there's something special about the incredibly intricate protein structures and morphologies of feathers (other than the obvious things like ratio of weight to rigidity that could be mimicked by artificial materials or worked around if forced to use an inferior substitute) that is crucial to making it possible for birds to fly. In 1900, could anyone have known for sure?

Neurons are, of course, almost unimaginably complex. Most of that complexity appears to be in the machinery that lets them do things all body cells (which are also comparably complex) do: metabolize, reproduce and differentiate during development, grow into complex shapes and configurations under the influence of internal and environmental signals, heal damage. By comparison, the computational behavior of individual neurons within a developed brain is far simpler and well understood. Output is a function of a weighted summing of inputs. That seems to be the basic mechanism, and that alone is now known to be sufficient to result in powerful computation once arrays of neurons are interconnected into larger functional networks. In addition, the presence of some substances in the cell, in the local bloodstream, or diffusing in the intracellular spaces alters the weightings, thresholds, and coefficients involved, and some inputs trigger production and/or release of those substances.

So yeah, there could conceivably be some hidden secret to neurons that is the true answer to how brains manage to think, besides their already well known, obvious, computationally rich and robust switching behavior. (And maybe there is some hidden secret to feathers that is the true answer to how birds manage to fly, besides being lightweight, water resistant, rigid, and resilient.) But I don't see any evidence for that and again I must ask why you think it is at all likely.

Respectfully,
Myriad

 

[Frank Newgent]

Describe in detail how, specifically, Mozart's Requiem results from underlying processes. Not Henryk Gorecki's Symphony No. 3. Not "Earth Died Screaming" by Tom Waits.

Mozart's Requiem.

If you can't it must have been produced by magic.

I'm not sure what you are asking since there are many things that we do not know.

Myriad reported the logical possibilities. There is either one substance with its underlying physical laws (and a sufficiently sophisticated machine could replicate those laws and therefore simulate that reality) or there is not. If there is not one substance, then there are two or more. If there are two substances, then we have magic, since that word describes the interaction between completely different substances.

Do you have a problem with the ability of a machine to theoretically simulate physical laws/reality or the question over what dualism really means? I assume you are not trying to point out that we are ignorant, since that has no bearing on the logical possibilities.

The idea that either scientists know exactly what is going on or it must be supernatural is pretty much the definition of a false dilemma.

Concerning "substances" and logic... that there can be multiple logically coherent possibilities within that one substance means that that one substance is not the determining factor... or to put it another way note how the logic that allows a theory of one substance to be formed is not accounted for by that theory of one substance.

Not necessarily trying to defend "dualism" just pointing out you've hardly explained it away.

 

[Myriad]

That's exactly what they're saying:

No, it is not. Please read again more carefully. "Not the same as water" is not equivalent to "not real." There are a great many things that are very real, that are not the same as water.

Flight simulators are real. "Simulation of flight" is not a real thing.

Of course it is, unless your definition of "real thing" excludes all events and processes. In which case, consciousness is also not a "real thing" so your entire objection becomes meaningless.

Just curious, is a sunrise real? Is it a real thing?

Exceptt when you simulate it. Then it becomes "real in the simulated world", and we're back to dualism.

I'm sorry, I can't make any sense of that assertion. Except when I simulate what? The state of a simulation? Why hypothesize that I would do so?

Everything that is real in a simulation is also real within the substrate of the simulation, which means it's real in the real world provided the substrate exists in the real world (e.g. a computer).

With enough replacements, you would have a simulated glass of water. You would not be able to drink it.

You proposed transistors as a reasonable replacement for a neuron. (I, acknowledging the greater complexity of neurons, substituted a miniature transistorized device in the thought experiment.) So now you are saying your own argument was rubbish?

Miniature transistorized devices the size of neurons are plausible in theory, though not actually achievable at present. Miniature transistorized devices the size of atoms are not possible even in theory. But, if you could create such devices, that would really interact with their surroundings in the same way as the component atoms of water molecules, with the same electromagnetic forces and quantum interactions, then yes I would be able to drink a glass of them. What would prevent me? How would I even know it wasn't normal water? (On a more practical level, you actually could replace the hydrogen atoms with deuterium atoms, which do indeed have similar electromagnetic and quantum behavior. And I really could drink it.)

It wouldn't be a simulated bird because the machine could actually fly. Very important distinction. A simulation of a bird can't fly. It just sits there, simulating flight.

It would fly like a bird. It would not actually be a bird (e.g. not containing bird DNA, not hatched from an egg, etc.). Therefore it would be a simulated bird. So at least some types of simulation of a bird can fly. Your very important distinction is therefore proven ill-defined.

Conscious experience is unlike anything else in the physical universe.

How do you know?

Respectfully,
Myriad

 

[Meed]

It would fly like a bird. It would not actually be a bird (e.g. not containing bird DNA, not hatched from an egg, etc.). Therefore it would be a simulated bird. So at least some types of simulation of a bird can fly. Your very important distinction is therefore proven ill-defined.

It can fly. It doesn't simulate flying. That's the distinction.

BTW, what you are calling a simulation here is not the sort of simulation we've been discussing in the thread.

 

[Myriad]

It can fly. It doesn't simulate flying. That's the distinction.

BTW, what you are calling a simulation here is not the sort of simulation we've been discussing in the thread.

To a large extent it is.

The difference is this:

Something like "flight" is relatively easy to define, if not to achieve. Even if we'd never seen anything fly, we could easily describe what would need to happen for it to take place. Consciousness is something which we can experience, but not describe. Hence even knowing whether it's happening is problematic. We cannot measure it, and the only way we have of guessing that it exists is that other people look and behave like us.

So, let's pretend that "flying" were not so easy to define or recognize. Then some (not all) of the arguments on this thread are the equivalent of saying that if we build a mechanical bird, it might do something that seems in some ways like flying, but it would not actually fly. We know that it would only be simulating flight because it is only a simulated bird.

Yes, that argument does seem silly when applied to something as clearly definable and recognizable as flying. That's my whole point. It makes no more sense when applied to hypothetical conscious machines, but that's less obvious.

Confounding that, in the discussion, are two other uses of the concept of simulation: simulation in the general sense of computer modeling, as in the discussion of flight simulation and the positing of a plank-level simulation of the universe as part of various philosophical thought experiments; and simulation as a concept in computing theory, by which one type of computing system can be proven to be equivalent to another in the sense of being able to perform any of the same computations, by showing that one system can exactly simulate the other.

This confusion (and consequent opportunities for equivocation, either accidental or deliberate) is unfortunate. But, I didn't start that fire. It's been burnin' burnin' since the thread's been turnin'.

Respectfully,
Myriad

 

[Malerin]

It can fly. It doesn't simulate flying. That's the distinction.

BTW, what you are calling a simulation here is not the sort of simulation we've been discussing in the thread.

To a large extent it is.

The difference is this:

Something like "flight" is relatively easy to define, if not to achieve. Even if we'd never seen anything fly, we could easily describe what would need to happen for it to take place. Consciousness is something which we can experience, but not describe. Hence even knowing whether it's happening is problematic. We cannot measure it, and the only way we have of guessing that it exists is that other people look and behave like us.

That's not the difference. What you're calling "simulation" would be more accurately be described as "imitation", "model", or "reproduction" (maybe there's an exact word that escapes me): A mechanical bird is functionally identical to an organic bird. Both fly. A simulated bird, on the other hand, is not functionally identical to an organic bird: one flies, the other doesn't.

The computationalist response to this has been dualistic: it really flies within the simulation. Water is wet within the simulation. So we end up with two categories of real /I]: real in 'our world'; and real 'in the simulated world'. Read Ichneumonwasp's posts, if you don't believe me.

I don't think anyone has any problem with a machine that can model or imitate consciousness. As you said, there's no function unique to neurons (so we think) that can't be reproduced by something mechanical. So it's possible to build a conscious machine. Maybe.

But a conscious machine would be functionally equivalent to us, just as the mechanical bird is functionally equivalent to an organic bird. It's not at all clear that simulated consciousness is real consciousness any more than simulated water is real water. That's where the dispute lies.

Maybe that clears things up.

 

[Meed]

^ Yep, that's what I meant.

 

[!Kaggen]

So, let's pretend that "flying" were not so easy to define or recognize. Then some (not all) of the arguments on this thread are the equivalent of saying that if we build a mechanical bird, it might do something that seems in some ways like flying, but it would not actually fly. We know that it would only be simulating flight because it is only a simulated bird.

Yes, that argument does seem silly when applied to something as clearly definable and recognizable as flying. That's my whole point. It makes no more sense when applied to hypothetical conscious machines, but that's less obvious.

The thing is with the property "flight" is that we need to actually see it to judge it is fight. There is no such thing as unseen flight.

Take the property "solidity". We need to actually touch something to judge its solidity. There is no such a thing as un-felt solidity.

Now take the property "consciousness". We need to communicate using natural language to judge whether something is conscious. There is no such thing as non-communicative consciousness.

In other words the properties of matter which we can describe with words always have a context in which there is a human judge.

Flight always has a context, it is not an abstract notion which moves from one substance (bird) to the next (aeroplane). When we match the percepts given to us with concepts we invent they make up the context of the flight we are currently witnessing. When the pioneers of mechanical flying conceived their machines they did not use the concepts of bird flight, they developed new concepts otherwise, like you said, we would still not have flying machines. The fact that there are similarities between the concepts employed for bird flight and mechanical flight does not make them the same concepts and therefore also not the same context of flight.

Abstracting the property flight out of context and then moving it around willy-nilly is dishonest. It is taking the human ability to form abstractions and then claiming it as a property of matter outside of the human. A hypnotizing ability for sure and a good candidate as to the origin of religion.

Now when it comes to consciousness it also becomes just as important to bring context into the discussion. The whole reality/simulation confusion is a direct result of not talking about consciousness in context.
The human consciousness as a property of the brain, we are all aware of, is what most of us use to reference consciousness as a property of matter. It is certainly difficult to imagine other types of consciousness. Inventing a concept of a consciousness (SRIP) which would result from computer parts is a great idea even though possibly unworkable. The dishonesty comes in when we are told we have to judge this new consciousness (even in a thought experiment) in the same context in which it was defined. This is the circular reasoning I keep referring too. That is like asking a bird to drop bird flight and pilot an aeroplane because after all, the aeroplane can fly. The joke of it all is that if the bird turned around and said no-thanks, we would have to accuse it of believing in magic.

 

[westprog]

How do you know?

If consciousness isn't unique, what is it like?

 

[Ichneumonwasp]

No. No more than when two rocks fall down a cliff.

How is it that someone must define that addition is even taking place after the process occurs (what happens when rocks fall down a cliff) with addition within a simulation?

With rocks falling there are physical substances that end up together and someone can view that as addition. With addition in a simulation concepts of numbers are manipulated, just as we manipulate concepts in our heads when we add numbers.

The constraints that lead these concepts of numbers to add together are put into the machine at the outset, so no one needs to interpret it for the process to occur. The process of addition is already defined in the simulation, so addition occurs when it occurs.

That is not the case with falling rocks.

Einstein's experiment was useful in determining what wasn't possible.

What difference does it make what use it was put to after the fact? It's primary usefulness was in understanding something fundamental about the universe; so the charge that something is not possible limiting the usefulness of a thought experiment is just silly.

We are not discussing what we can physically do right here and now with this thought experiment; the entire thrust of it is to expose assumptions brought to the table.

No one is trying to 'determine if the simulation contains consciousness' -- you decide if it does or does not and that decision carries consequences.

 

[Ichneumonwasp]

The idea that either scientists know exactly what is going on or it must be supernatural is pretty much the definition of a false dilemma.

Concerning "substances" and logic... that there can be multiple logically coherent possibilities within that one substance means that that one substance is not the determining factor... or to put it another way note how the logic that allows a theory of one substance to be formed is not accounted for by that theory of one substance.

Not necessarily trying to defend "dualism" just pointing out you've hardly explained it away.

Scientists knowing exactly what is going on is not what is being proposed. That we could theoretically model all of the forces of nature provided there is a single substance is.

No one is saying that magic is not logically possible; and, yes, if it makes sense to speak of the origin of a single substance, then that origin would fit under 'magic'. That is why many argue against the very idea of origin; eternal presence makes more sense logically.

There is no way to explain away dualism. It is logically possible. It is simply intellectually unsatisfying/an intellectual dead end.

 

[westprog]

Confounding that, in the discussion, are two other uses of the concept of simulation: simulation in the general sense of computer modeling, as in the discussion of flight simulation and the positing of a plank-level simulation of the universe as part of various philosophical thought experiments; and simulation as a concept in computing theory, by which one type of computing system can be proven to be equivalent to another in the sense of being able to perform any of the same computations, by showing that one system can exactly simulate the other.

This confusion (and consequent opportunities for equivocation, either accidental or deliberate) is unfortunate. But, I didn't start that fire. It's been burnin' burnin' since the thread's been turnin'.

Respectfully,
Myriad

It's something well worth defining on a regular basis. Part of the problem has been the insistence that it's so obvious that the brain is a computational device and that consciousness is a product of computation, that it's merely necessary to point out the equivalence principle and the whole thing is tied up.

What's also crept in is the idea that a simulation is a "world" in itself, just as real as this one. I find this a very odd concept.