The brain and the physical world in general are sufficient to explain consciousness, so therefore any assumptions beyond that get a complexity penalty.
And...that’s the only reason. All your “possibilities” are indeed possible… but improbable. I don’t reject the idea that consciousness could theoretically reside outside the brain, but it’s much more parsimonious to assume it does not.
I’ve read a fair amount on Less Wrong
If so, I take it you already understand about parsimony and its importance for hypothesizing, since there has been a good deal of discussion about that. Additionally, as a neuroscience major you have sufficient background to at least agree that consciousness could theoretically be an entirely physical phenomenon. You’ve got all the background information you need to make the required inferences.
So for my own curiosity, do let me know: Was reading above sufficient to cause you to alter your belief, and do you now know that consciousness probably (more probably than “50/50”) is an entirely physical phenomenon which centers around the brain?
Edit: Reading your other comments, you still don’t get parsimony.
Hypothesis 1: I have only two left toes.
Hypothesis 2: I have only two left toes and own a bunny rabbit hat.
1 is more parsimonious than 2. It’s not 50⁄50.
Hypothesis 1: The brain is enough to explain the evidence of consciousness
Hypothesis 2: The brain is enough to explain the evidence of consciousness, but there are additional things which are also conscious, only we can’t observe them.
You keep coming back to if consciousness could be not just the brain, if there could be something else.
Yes, it could. It’s just extremely un-parsimonious to postulate that it does.
If you agree that consciousness could be explained by the brain alone, then parsimony should lead you to agree that consciousness is probably explained by the brain alone. If you don’t agree that consciousness could be explained by the brain alone, then that’s a longer problem.
Every time you catch yourself saying “but what if X” or “but how do we know X for sure”, consider parsimony.
From what I understand, things that are parsimonious are less likely than the simpler option because of compounded probability. Ie. something parsimonious requires a1..a25 to be true, while a simpler option just requires a1..a5 to be true.
My point is that I don’t see reason to think that we have any information about the probabilities. How can we say that “a1..a500 needs to be true in order for consciousness to remain after brain destruction”? What observations have we made that would lead us to think that? My feeling is that we’ve never actually made an observation that says x ⇒ unconsciousness, because we’ve never actually been able to infer a state of unconsciousness.
Note: Sorry if I’m just not understanding the point about parsimony. I know that everyone seems to disagree with me and is making that point, so I’ve been trying to understand it and think about how it disproves my current belief, but for the reasons I explain above I don’t think the argument that “parsimony ⇒ it’s unlikely that consciousness remains” is valid. That argument requires information about what causes unconsciousness that I don’t think we have.
We have some data on what preconditions seem to produce consciousness (ie. neuronal firing). However, this is just data on the preconditions that seem to produce consciousness that can/do communicate/demonstrate its consciousness to us.
So you agree that brains are sufficient to explain consciousness. This consists of a1...a5.
Let Hypothesis 1 be that brains are conscious.
Then, as Hypothesis 2, you have the other conscious beings (a5...a25). Note that H2 also believes that brains are conscious. (a1...a5). So you have...
H1: a1...a5 “brains are conscious”
H2: a1...a25 “brains are conscious, and there’s other types of consciousnesses outside our perception
H3:a1...a5, b1...b5: “brains are conscious, and there’s a teapot on Andromeda.
H4: a1...a5, c1...c5: Brains are conscious, and there is no such thing as a consciousness outside our perception.
H5: a1...a5, d1...d5: Brains are conscious, and there are no teapots in Andromeda.
H6: a1...a5, e1...e5: Brains are conscious, and there is no water in Andromeda
H7: a1...a5, f1...f5: Brains are conscious, and there is water in Andromeda
In order of parsimony, it’s obvious that H1>H2,H3,H4, H5, H6, H7, right?
Right, but your real question is: is H2 more parsimonious than H4. And you’re right, practically there is no mathematically rigorous way to get the answer.
The same can be said of H3 vs H5, but you’ve got a strong intuition that H5 is more likely, right? Can you prove it rigorously? No, but you’ve got a pretty strong sense that their are no teapots to be found on Pluto.
Similarly, between H6 and H7, you’ve got a fair sense that there is probably water somewhere on Andromeda. In fact, it would be more complicated to describe some circumstance by which Andromeda didn’t have any water.
So is consciousness more like the teapot, or more like the water molecule? Well, your description of the universe gets simpler when you don’t need to explain the teapot, whereas it gets more complex when you have to explain why there is no water in Andromeda.
Consider this, like all humans you have instinctive animist tendencies. You are emotionally biased to favor H2 as something that seems possible, because you emotionally think of consciousness as a simple, basic element of reality, like water. I say, consciousness is more like a teapot than it is like water.
Does Math make special exceptions for consciousnesses that it does not make for teapots? Think about it...you can imagine water forming on a star somewhere, it’s fairly simple. How do you envision these separate consciousnesses forming? All possible ways in which these extra-physical consciousnesses could form are really complex. Your description of the universe, once you add these extra consciousnesses in, is going to get larger, not smaller.
You don’t automatically find yourself scrambling to explain why there might not be afterlife...rather, you find yourself searching for an explanation for why there might be one. And that’s because afterlives make the description of the universe larger and more complex and therefore require you to generate a story.
I admit, this isn’t a proof, and you’re not going to get a proof. But it’s a really strong intuition.
...I wonder if it would help, if I came up with an unrelated idea that could only be rejected using intuition-parsimony and asked you to refute it. You’ll instinctively call on parsimony, and then you can apply the same methods to the afterlife hypothesis.
My point is that I don’t see reason to think that we have any information about the probabilities. How can we say that “a1..a500 needs to be true in order for consciousness to remain after brain destruction”? What observations have we made that would lead us to think that? My feeling is that we’ve never actually made an observation that says x ⇒ unconsciousness, because we’ve never actually been able to infer a state of unconsciousness.
Right, but your real question is: is H2 more parsimonious than H4.
So you say that it can’t be proven, but you have a “really strong intuition” that it is. Why? What observations have you made about what causes unconsciousness that would lead you to believe that it involves parsimony? And how have you been able to infer unconsciousness?
Because consciousness (regardless of whether it’s based in something physical and observable, like the brain) by its inherent nature involves complex information processing. Even if you separate consciousness from the brain and put it in the abstract-unobservable-spirit-place-thing, it’s still a mathematically defined structure with a lot of complexity. When the brain is right in front of you, you can point to it and say, there! that’s a complex information processing structure!”When the brain is no longer in front of you, you necessarily have to posit an un-observable* complex information processing structure in spirit land.
Just replace brain with any other object, and you″ll get the same intuition. How do you know that some sort of time-keeping doesn’t continue in an unobserved location, even after the physical clock is destroyed? How do you know that the teapot-ness doesn’t continue on somewhere after the actual teapot is destroyed? And what about your past selves? They are all now destroyed? Are they all continuing on somewhere?
by its inherent nature involves complex information processing
Interesting. There does seem to be evidence that you need a complex structure with complex information processing to provide a variety of conscious experiences. The evidence for this I think is just that outcomes have independent causes (on the smallest levels). You’d need a complex structure to take in all the different inputs and produce the corresponding conscious experience as the output. A simple structure can’t do that.
When the brain is no longer in front of you, you necessarily have to posit an *un-observable complex information processing structure in spirit land.
I wouldn’t say that. Right now, we have some data on what parts of the brain need to be active for consciousness, but we can’t measure things at a level of precision below single neurons. What if something is happening on a quantum level that underlies consciousness? What if the thing that underlies consciousness is present at a level below the quantum level, like something beyond our current understanding of physics? This is quite possible, and I don’t think it’s ridiculous to posit that this level may go undisturbed when the brain is destroyed.
So, as far as retaining “complex consciousness” (able to experience a variety of things) after the brain is destroyed, I see two possibilities:
The thing that causes the consciousness you experience when you’re alive is destroyed, but some new structure is created and provides you with complex consciousness.
The thing that is currently causing your complex consciousness remains intact, and thus continues to provide you with that complex consciousness.
I agree that 1) is unlikely for reasons of parsimony. It’s 2) that I’m questioning. Why is 2) more likely to be false than true?
Answering that question myself, I actually think that it is. If I knew more about physics I’d have a stronger opinion here, but I figure that when you destroy the brain on macro/microscopic level, it’s unlikely for the nano/quantum/small level that I’m saying consciousness might be on to go undisturbed.
So back to my original objection—“What have we observed that would tell us that x ⇒ unconsciousness”. Answer:
1) We’ve observed that the world is governed by cause and effect. Consider a given outcome. You can’t have two different physical states lead to that outcome. (I’m not explaining this well, but hopefully you know what I mean).
2) We’ve observed that consciousness involves a large variety of different outcomes (seeing red, seeing blue, feeling hot, feeling cold...). From 1) we can infer that there must be a certain physical state that leads to each of these outcomes.
3) We’ve observed that the brain is a complex structure that is correlated with consciousness. I don’t think we know the cause. Maybe it’s on the neuronal level. I think it probably has to be on a smaller level. But regardless, it seems likely that when you destroy the brain, you’d destroy whatever it is that’s producing consciousness, regardless of what level it’s on.
4) We’ve observed parsimony. So it’s unlikely that whatever causes consciousness will be regenerated out of the blue. once it’s destroyed.
So to be explicit, I think it’s unlikely that consciousness remains after the brain is destroyed.
But one thing might remain. There might be a sort of basic/flat level of consciousness, as opposed to “nothingness”. And as opposed to the idea that consciousness has to involve our complex consciousness of experiencing all the various thing we experience. There may be a basic level where we only sort of experience one thing.
If this level exists, how do we know that destroying the brain interferes with it? What do you think?
That’s all I’ve got for now. I probably haven’t expressed these points too clearly, as I’m just coming up with a lot of them and haven’t had time analyze them enough. Please let me know what you think, and if you could sum it up and express it a little more clearly than I have. Thanks for the conversation!
There might be a sort of basic/flat level of consciousness
That position is called pan-psychism. I don’t think pan-psychism violates the rules of parsimony, but I also think once you find yourself asking how quantum vacuums or a baseball bats or water molecules subjectively feel, you need to back up a bit.
If this level exists, how do we know that destroying the brain interferes with it? What do you think?
Personally? I think that my qualia is that which separates reality from all the other hypothetical mathematical structures, and just leave it at that (so I evaluate consciousness from a strictly information-processing standpoint).. Well, that’s the short version, I’d probably need to write a bit more for that to make sense.
The brain and the physical world in general are sufficient to explain consciousness
The problem is that they aren’t, as Richard explains here.
1) In context, we’re talking about consciousness as in beings which are behaviorally aware, not about subjective-experience qualia, right?
I don’t understand the distinction you’re making there. As I use the words, consciousness is awareness, which is experience. These are just different ways of pointing at the same thing.
2) Richard says he doesn’t know, not that they aren’t.
The problem is worse than merely not knowing, in the sense in which we do not know a cure for cancer. We can imagine that eventually, we will discover enough about the mechanism of cancer to devise an intervention as effective as penicillin for bacterial infection. But we cannot see, in terms of what we understand of physics and the brain, how there could be any such thing as consciousness, despite people giving the matter a great deal of thought and experimental work. That’s a strong argument that they are not sufficient to explain it.
There’s a tautologous sense in which they are sufficient, by taking “the physical world” to include whatever the real explanation turns out to be, but in a discussion of this issue “the physical world” means the world as understood in terms of the sort of physical theories we currently have.
On the other hand, the very close observable connection between brain states and consciousness is a strong argument that the brain and the physical world are sufficient to explain consciousness.
In short we are faced with a gigantic problem:
We are conscious.
Consciousness is very closely connected with the brain.
We cannot see how there could be any such phenomenon.
Go to 1.
A lot of discussion on the subject consists of people writing their conclusion in different words and using it as an argument for their conclusion.
I don’t understand the distinction you’re making there
“Behaviorally aware” is a term I’m using to talk about consciousness without invoking the “hard problem of consciousness”.
The brain is a structure which takes various inputs, does a bunch of operations to them, and produces various outputs. We can see how that works, and to some extents we can make machines that do the same.
When someone’s “unconscious”, it means they are no longer responding to the environment (taking inputs, producing outputs) appropriately. A “conscious” being is responding appropriately to the environment. It’s various internal parts are interacting with each other in an organized way, and they are also interacting with the external world in an organized way. Behaviorally speaking, they are aware and reacting.
None of the above has yet involved qualia, subjective experience, or the hard problem of consciousness. We are using the word “conscious” to mean two separate things—“aware-in-the-information-processing-sense” and “subjectively-experiencing-qualia”.
So you can talk about whether or not there are conscious (information-processing) structures which continue on after we die without tackling subjective experience or qualia. And when you talk about these structures, you still have to use parsimony...just because the information-processing structure is no longer in the “observable-physical world” or whatever doesn’t mean it’s not still a complex, rule-following mathematical structure.
Which is why I say, in the context of this conversation, there is no need to invoke The Hard Problem
A lot of discussion on the subject consists of people writing their conclusion in different words and using it as an argument for their conclusion.
I think this is because this is primarily a matter of definition. The “answer” to the “hard problem” is decidedly not empirical and purely philosophical. All non-empirical statements are tautological in nature. Arguments aiming to dissolve the question rely upon altering definitions and are necessarily tautological.
“Behaviorally aware” is a term I’m using to talk about consciousness without invoking the “hard problem of consciousness”.
The brain is a structure which takes various inputs, does a bunch of operations to them, and produces various outputs. We can see how that works, and to some extents we can make machines that do the same.
Why call this “consciousness”? Pretty much any machine that we make takes various inputs, does a bunch of operations to them, and produces various outputs. Is my computer (my real computer, not an imaginary one programmed with an imaginary AI) “behaviourally aware”? It even runs tests on itself and reports the results.
I don’t think it useful to broaden the definition of “conscious” to include things that are clearly not “conscious” (in the meaning it normally has). This doesn’t let you talk about consciousness without invoking the “hard problem of consciousness”. It lets you talk about something completely different that you are calling by the same name, without invoking the “hard problem of consciousness”.
A lot of discussion on the subject consists of people writing their conclusion in different words and using it as an argument for their conclusion.
I think this is because this is primarily a matter of definition. The “answer” to the “hard problem” is decidedly not empirical and purely philosophical.
The problem is clearly an empirical one. We are aware, seek an explanation, but have not found one.
Is my computer (my real computer, not an imaginary one programmed with an imaginary AI) “behaviourally aware”? It even runs tests on itself and reports the results.
Yes, actually? To the extent that a worm is aware.
We don’t normally use the word “aware” to describe it, but what it’s doing seems very, very close to the things we do describe with the word awareness.
The problem is clearly an empirical one.
Then I’ve misunderstood your claim. The Hard Problem of Consciousness as popularly understood is that even if we understand all the mechanisms of thought to the point that we can construct brains ourselves, it won’t explain the subjective experience we have. We can understand the universe with mathematical precision down to the last photon and it still wouldn’t explain it. Seems like a non-empirical question to me. That’s why they call it subjective experience.
Is my computer (my real computer, not an imaginary one programmed with an imaginary AI) “behaviourally aware”? It even runs tests on itself and reports the results.
Yes, actually? To the extent that a worm is aware.
Is a worm aware? I don’t know. Is my computer aware? I see no reason to think so, not in the sense of “aware” that we’re discussing. Is a thermostat aware? That too has input and output. Is a rock aware? If the answer to all of these is “yes”, then that is not a useful sense of “aware”. It’s just another route for the mercury blob of thought to escape the finger of logic.
In other contexts, I have no problem with talking about a robot (a real robot really existing in the real world right now, such as Google’s driverless cars) as being “aware” of something, or for that matter my computer running self-tests, but I would also know that I was not imputing consciousness to the devices. If we’re going to talk about consciousness, that is what we must talk about, instead of broadening the word beyond what we are talking about and using the same word to talk about some other thing instead.
The Hard Problem of Consciousness as popularly understood is that even if we understand all the mechanisms of thought to the point that we can construct brains ourselves, it won’t explain the subjective experience we have.
I would say that’s one particular position, or class of positions, on the Hard Problem. The other class of positions are those that hold that if we understood all etc. etc. then it would explain the subjective experience we have.
The Hard Problem, to me, is that both of these positions are both ineluctable and untenable.
That’s why they call it subjective experience.
That we have subjective experience is an objective fact.
Is there no middle ground between “aware” and “not aware” then? This is like asking “Is a boulder a chair?”, “is a tree stump a chair?” “Is a stool a chair?” Words are fuzzy like that.
That we have subjective experience is an objective fact.
Rather, that you have it is an objective fact to you. The empirical questions involved here are applied to other minds, not your own.
Is there no middle ground between “aware” and “not aware” then? This is like asking “Is a boulder a chair?”, “is a tree stump a chair?” “Is a stool a chair?” Words are fuzzy like that.
Yes, there’s a whole range. Maybe a worm has a microconsciousness, or a nanoconsciousness, or maybe it has none at all, relative to a human. Or maybe it’s like asking about the temperature of a cluster of a few atoms. The concept is indeed fuzzy.
That we have subjective experience is an objective fact.
Rather, that you have it is an objective fact to you. The empirical questions involved here are applied to other minds, not your own.
Other people seem to be the same sorts of thing as me, and they report awareness of things. That’s good enough for me to believe them to have consciousness. When robots get good enough to not sound like spam when they pretend to be people, then that criterion would have to be reexamined.
As Scott Aaronson points out in his discussion of IIT, experiences of oneself and intuitions about other creatures based on their behaviour are all we have to go on at present. If an explanation of consciousness doesn’t more or less match up to those intuitions, it’s a problem for the explanation, not the intuitions.
The Hard Problem of Consciousness as popularly understood is that even if we understand all the mechanisms of thought to the point that we can construct brains ourselves, it won’t explain the subjective experience we have. We can understand the universe with mathematical precision down to the last photon and it still wouldn’t explain it. Seems like a non-empirical question to me.
The common meaning of “empirical” is something based on experience, so it seems that the Hard Problem of Consciousness fits that definition.
No? There is no subjective experience I can have that can distinguish you from a P-zombie (under the (wrong) assumption that the hard-problem even makes sense and that there is a meaningful distinction to be made there)
It is a simple case of parsimony.
The brain and the physical world in general are sufficient to explain consciousness, so therefore any assumptions beyond that get a complexity penalty.
And...that’s the only reason. All your “possibilities” are indeed possible… but improbable. I don’t reject the idea that consciousness could theoretically reside outside the brain, but it’s much more parsimonious to assume it does not.
If so, I take it you already understand about parsimony and its importance for hypothesizing, since there has been a good deal of discussion about that. Additionally, as a neuroscience major you have sufficient background to at least agree that consciousness could theoretically be an entirely physical phenomenon. You’ve got all the background information you need to make the required inferences.
So for my own curiosity, do let me know: Was reading above sufficient to cause you to alter your belief, and do you now know that consciousness probably (more probably than “50/50”) is an entirely physical phenomenon which centers around the brain?
Edit: Reading your other comments, you still don’t get parsimony.
Hypothesis 1: I have only two left toes.
Hypothesis 2: I have only two left toes and own a bunny rabbit hat.
1 is more parsimonious than 2. It’s not 50⁄50.
Hypothesis 1: The brain is enough to explain the evidence of consciousness
Hypothesis 2: The brain is enough to explain the evidence of consciousness, but there are additional things which are also conscious, only we can’t observe them.
1 is more parsimonious than 2. It’s not 50⁄50.
This is my position—http://lesswrong.com/lw/k8a/what_do_rationalists_think_about_the_afterlife/awx0.
that’s just a link to this thread.
Oh sorry, I thought it was a permalink. This link works -
http://lesswrong.com/lw/k8a/what_do_rationalists_think_about_the_afterlife/awx0
*When I put a period after the link, it included the period in the url, that’s why it wasn’t working.
You keep coming back to if consciousness could be not just the brain, if there could be something else.
Yes, it could. It’s just extremely un-parsimonious to postulate that it does.
If you agree that consciousness could be explained by the brain alone, then parsimony should lead you to agree that consciousness is probably explained by the brain alone. If you don’t agree that consciousness could be explained by the brain alone, then that’s a longer problem.
Every time you catch yourself saying “but what if X” or “but how do we know X for sure”, consider parsimony.
From what I understand, things that are parsimonious are less likely than the simpler option because of compounded probability. Ie. something parsimonious requires a1..a25 to be true, while a simpler option just requires a1..a5 to be true.
My point is that I don’t see reason to think that we have any information about the probabilities. How can we say that “a1..a500 needs to be true in order for consciousness to remain after brain destruction”? What observations have we made that would lead us to think that? My feeling is that we’ve never actually made an observation that says x ⇒ unconsciousness, because we’ve never actually been able to infer a state of unconsciousness.
Note: Sorry if I’m just not understanding the point about parsimony. I know that everyone seems to disagree with me and is making that point, so I’ve been trying to understand it and think about how it disproves my current belief, but for the reasons I explain above I don’t think the argument that “parsimony ⇒ it’s unlikely that consciousness remains” is valid. That argument requires information about what causes unconsciousness that I don’t think we have.
So you agree that brains are sufficient to explain consciousness. This consists of a1...a5.
Let Hypothesis 1 be that brains are conscious.
Then, as Hypothesis 2, you have the other conscious beings (a5...a25). Note that H2 also believes that brains are conscious. (a1...a5). So you have...
H1: a1...a5 “brains are conscious”
H2: a1...a25 “brains are conscious, and there’s other types of consciousnesses outside our perception
H3:a1...a5, b1...b5: “brains are conscious, and there’s a teapot on Andromeda.
H4: a1...a5, c1...c5: Brains are conscious, and there is no such thing as a consciousness outside our perception.
H5: a1...a5, d1...d5: Brains are conscious, and there are no teapots in Andromeda.
H6: a1...a5, e1...e5: Brains are conscious, and there is no water in Andromeda
H7: a1...a5, f1...f5: Brains are conscious, and there is water in Andromeda
In order of parsimony, it’s obvious that H1>H2,H3,H4, H5, H6, H7, right?
Right, but your real question is: is H2 more parsimonious than H4. And you’re right, practically there is no mathematically rigorous way to get the answer.
The same can be said of H3 vs H5, but you’ve got a strong intuition that H5 is more likely, right? Can you prove it rigorously? No, but you’ve got a pretty strong sense that their are no teapots to be found on Pluto.
Similarly, between H6 and H7, you’ve got a fair sense that there is probably water somewhere on Andromeda. In fact, it would be more complicated to describe some circumstance by which Andromeda didn’t have any water.
So is consciousness more like the teapot, or more like the water molecule? Well, your description of the universe gets simpler when you don’t need to explain the teapot, whereas it gets more complex when you have to explain why there is no water in Andromeda.
Consider this, like all humans you have instinctive animist tendencies. You are emotionally biased to favor H2 as something that seems possible, because you emotionally think of consciousness as a simple, basic element of reality, like water. I say, consciousness is more like a teapot than it is like water.
Does Math make special exceptions for consciousnesses that it does not make for teapots? Think about it...you can imagine water forming on a star somewhere, it’s fairly simple. How do you envision these separate consciousnesses forming? All possible ways in which these extra-physical consciousnesses could form are really complex. Your description of the universe, once you add these extra consciousnesses in, is going to get larger, not smaller.
You don’t automatically find yourself scrambling to explain why there might not be afterlife...rather, you find yourself searching for an explanation for why there might be one. And that’s because afterlives make the description of the universe larger and more complex and therefore require you to generate a story.
I admit, this isn’t a proof, and you’re not going to get a proof. But it’s a really strong intuition.
...I wonder if it would help, if I came up with an unrelated idea that could only be rejected using intuition-parsimony and asked you to refute it. You’ll instinctively call on parsimony, and then you can apply the same methods to the afterlife hypothesis.
So you say that it can’t be proven, but you have a “really strong intuition” that it is. Why? What observations have you made about what causes unconsciousness that would lead you to believe that it involves parsimony? And how have you been able to infer unconsciousness?
Because consciousness (regardless of whether it’s based in something physical and observable, like the brain) by its inherent nature involves complex information processing. Even if you separate consciousness from the brain and put it in the abstract-unobservable-spirit-place-thing, it’s still a mathematically defined structure with a lot of complexity. When the brain is right in front of you, you can point to it and say, there! that’s a complex information processing structure!”When the brain is no longer in front of you, you necessarily have to posit an un-observable* complex information processing structure in spirit land.
Just replace brain with any other object, and you″ll get the same intuition. How do you know that some sort of time-keeping doesn’t continue in an unobserved location, even after the physical clock is destroyed? How do you know that the teapot-ness doesn’t continue on somewhere after the actual teapot is destroyed? And what about your past selves? They are all now destroyed? Are they all continuing on somewhere?
Interesting. There does seem to be evidence that you need a complex structure with complex information processing to provide a variety of conscious experiences. The evidence for this I think is just that outcomes have independent causes (on the smallest levels). You’d need a complex structure to take in all the different inputs and produce the corresponding conscious experience as the output. A simple structure can’t do that.
I wouldn’t say that. Right now, we have some data on what parts of the brain need to be active for consciousness, but we can’t measure things at a level of precision below single neurons. What if something is happening on a quantum level that underlies consciousness? What if the thing that underlies consciousness is present at a level below the quantum level, like something beyond our current understanding of physics? This is quite possible, and I don’t think it’s ridiculous to posit that this level may go undisturbed when the brain is destroyed.
So, as far as retaining “complex consciousness” (able to experience a variety of things) after the brain is destroyed, I see two possibilities:
The thing that causes the consciousness you experience when you’re alive is destroyed, but some new structure is created and provides you with complex consciousness.
The thing that is currently causing your complex consciousness remains intact, and thus continues to provide you with that complex consciousness.
I agree that 1) is unlikely for reasons of parsimony. It’s 2) that I’m questioning. Why is 2) more likely to be false than true?
Answering that question myself, I actually think that it is. If I knew more about physics I’d have a stronger opinion here, but I figure that when you destroy the brain on macro/microscopic level, it’s unlikely for the nano/quantum/small level that I’m saying consciousness might be on to go undisturbed.
So back to my original objection—“What have we observed that would tell us that x ⇒ unconsciousness”. Answer:
1) We’ve observed that the world is governed by cause and effect. Consider a given outcome. You can’t have two different physical states lead to that outcome. (I’m not explaining this well, but hopefully you know what I mean).
2) We’ve observed that consciousness involves a large variety of different outcomes (seeing red, seeing blue, feeling hot, feeling cold...). From 1) we can infer that there must be a certain physical state that leads to each of these outcomes.
3) We’ve observed that the brain is a complex structure that is correlated with consciousness. I don’t think we know the cause. Maybe it’s on the neuronal level. I think it probably has to be on a smaller level. But regardless, it seems likely that when you destroy the brain, you’d destroy whatever it is that’s producing consciousness, regardless of what level it’s on.
4) We’ve observed parsimony. So it’s unlikely that whatever causes consciousness will be regenerated out of the blue. once it’s destroyed.
So to be explicit, I think it’s unlikely that consciousness remains after the brain is destroyed.
But one thing might remain. There might be a sort of basic/flat level of consciousness, as opposed to “nothingness”. And as opposed to the idea that consciousness has to involve our complex consciousness of experiencing all the various thing we experience. There may be a basic level where we only sort of experience one thing.
If this level exists, how do we know that destroying the brain interferes with it? What do you think?
That’s all I’ve got for now. I probably haven’t expressed these points too clearly, as I’m just coming up with a lot of them and haven’t had time analyze them enough. Please let me know what you think, and if you could sum it up and express it a little more clearly than I have. Thanks for the conversation!
That position is called pan-psychism. I don’t think pan-psychism violates the rules of parsimony, but I also think once you find yourself asking how quantum vacuums or a baseball bats or water molecules subjectively feel, you need to back up a bit.
Personally? I think that my qualia is that which separates reality from all the other hypothetical mathematical structures, and just leave it at that (so I evaluate consciousness from a strictly information-processing standpoint).. Well, that’s the short version, I’d probably need to write a bit more for that to make sense.
The problem is that they aren’t, as Richard explains here.
1) In context, we’re talking about consciousness as in beings which are behaviorally aware, not about subjective-experience qualia, right?
2) Richard says he doesn’t know, not that they aren’t.
I don’t understand the distinction you’re making there. As I use the words, consciousness is awareness, which is experience. These are just different ways of pointing at the same thing.
The problem is worse than merely not knowing, in the sense in which we do not know a cure for cancer. We can imagine that eventually, we will discover enough about the mechanism of cancer to devise an intervention as effective as penicillin for bacterial infection. But we cannot see, in terms of what we understand of physics and the brain, how there could be any such thing as consciousness, despite people giving the matter a great deal of thought and experimental work. That’s a strong argument that they are not sufficient to explain it.
There’s a tautologous sense in which they are sufficient, by taking “the physical world” to include whatever the real explanation turns out to be, but in a discussion of this issue “the physical world” means the world as understood in terms of the sort of physical theories we currently have.
On the other hand, the very close observable connection between brain states and consciousness is a strong argument that the brain and the physical world are sufficient to explain consciousness.
In short we are faced with a gigantic problem:
We are conscious.
Consciousness is very closely connected with the brain.
We cannot see how there could be any such phenomenon.
Go to 1.
A lot of discussion on the subject consists of people writing their conclusion in different words and using it as an argument for their conclusion.
“Behaviorally aware” is a term I’m using to talk about consciousness without invoking the “hard problem of consciousness”.
The brain is a structure which takes various inputs, does a bunch of operations to them, and produces various outputs. We can see how that works, and to some extents we can make machines that do the same.
When someone’s “unconscious”, it means they are no longer responding to the environment (taking inputs, producing outputs) appropriately. A “conscious” being is responding appropriately to the environment. It’s various internal parts are interacting with each other in an organized way, and they are also interacting with the external world in an organized way. Behaviorally speaking, they are aware and reacting.
None of the above has yet involved qualia, subjective experience, or the hard problem of consciousness. We are using the word “conscious” to mean two separate things—“aware-in-the-information-processing-sense” and “subjectively-experiencing-qualia”.
So you can talk about whether or not there are conscious (information-processing) structures which continue on after we die without tackling subjective experience or qualia. And when you talk about these structures, you still have to use parsimony...just because the information-processing structure is no longer in the “observable-physical world” or whatever doesn’t mean it’s not still a complex, rule-following mathematical structure.
Which is why I say, in the context of this conversation, there is no need to invoke The Hard Problem
I think this is because this is primarily a matter of definition. The “answer” to the “hard problem” is decidedly not empirical and purely philosophical. All non-empirical statements are tautological in nature. Arguments aiming to dissolve the question rely upon altering definitions and are necessarily tautological.
Why call this “consciousness”? Pretty much any machine that we make takes various inputs, does a bunch of operations to them, and produces various outputs. Is my computer (my real computer, not an imaginary one programmed with an imaginary AI) “behaviourally aware”? It even runs tests on itself and reports the results.
I don’t think it useful to broaden the definition of “conscious” to include things that are clearly not “conscious” (in the meaning it normally has). This doesn’t let you talk about consciousness without invoking the “hard problem of consciousness”. It lets you talk about something completely different that you are calling by the same name, without invoking the “hard problem of consciousness”.
The problem is clearly an empirical one. We are aware, seek an explanation, but have not found one.
Yes, actually? To the extent that a worm is aware.
We don’t normally use the word “aware” to describe it, but what it’s doing seems very, very close to the things we do describe with the word awareness.
Then I’ve misunderstood your claim. The Hard Problem of Consciousness as popularly understood is that even if we understand all the mechanisms of thought to the point that we can construct brains ourselves, it won’t explain the subjective experience we have. We can understand the universe with mathematical precision down to the last photon and it still wouldn’t explain it. Seems like a non-empirical question to me. That’s why they call it subjective experience.
Is a worm aware? I don’t know. Is my computer aware? I see no reason to think so, not in the sense of “aware” that we’re discussing. Is a thermostat aware? That too has input and output. Is a rock aware? If the answer to all of these is “yes”, then that is not a useful sense of “aware”. It’s just another route for the mercury blob of thought to escape the finger of logic.
In other contexts, I have no problem with talking about a robot (a real robot really existing in the real world right now, such as Google’s driverless cars) as being “aware” of something, or for that matter my computer running self-tests, but I would also know that I was not imputing consciousness to the devices. If we’re going to talk about consciousness, that is what we must talk about, instead of broadening the word beyond what we are talking about and using the same word to talk about some other thing instead.
I would say that’s one particular position, or class of positions, on the Hard Problem. The other class of positions are those that hold that if we understood all etc. etc. then it would explain the subjective experience we have.
The Hard Problem, to me, is that both of these positions are both ineluctable and untenable.
That we have subjective experience is an objective fact.
Is there no middle ground between “aware” and “not aware” then? This is like asking “Is a boulder a chair?”, “is a tree stump a chair?” “Is a stool a chair?” Words are fuzzy like that.
Rather, that you have it is an objective fact to you. The empirical questions involved here are applied to other minds, not your own.
Yes, there’s a whole range. Maybe a worm has a microconsciousness, or a nanoconsciousness, or maybe it has none at all, relative to a human. Or maybe it’s like asking about the temperature of a cluster of a few atoms. The concept is indeed fuzzy.
Other people seem to be the same sorts of thing as me, and they report awareness of things. That’s good enough for me to believe them to have consciousness. When robots get good enough to not sound like spam when they pretend to be people, then that criterion would have to be reexamined.
As Scott Aaronson points out in his discussion of IIT, experiences of oneself and intuitions about other creatures based on their behaviour are all we have to go on at present. If an explanation of consciousness doesn’t more or less match up to those intuitions, it’s a problem for the explanation, not the intuitions.
The common meaning of “empirical” is something based on experience, so it seems that the Hard Problem of Consciousness fits that definition.
No? There is no subjective experience I can have that can distinguish you from a P-zombie (under the (wrong) assumption that the hard-problem even makes sense and that there is a meaningful distinction to be made there)