It looks like the article only references philosophers and psychologists, not people like, for example, Thomas Metzinger, Stanislas Dehaene and Joscha Bach.
A recent comment on HN led me to Metzinger's The Ego Tunnel. IMO that book alone would get a lot of people past neuroscience-uninformed Philosophy 101 takes on philosophy-of-mind.
I'm very impressed by some of the advances in neuroscience over the last 10 years or so. It seemed like the field was stalled for a long time since the 80s but some new technology and clever experimental techniques have got progress moving again. I wouldn't be surprised if we hit another sticky patch, I suspect it's going to be a long road to solving this problem, but it's exciting to have new ideas and demonstrations to think about.
The problem with solving the mystery of consciousness is that we don't even really have a definition of it that makes sense. Until then, the answer will inevitably be 42.
How are qualia necessary/sufficient for consciousness? Seems more related to the subjectivity of perception, which in itself doesn't require consciousness (or does it?)?
Qualia is the term for the sensations of color, sound, etc. And not just perception, but dreams, hallucinations, etc. That's what makes up consciousness/subjectivity.
Not everyone agrees with using the term qualia for sensation. There's various debates over the nature of perception. Some see sensation as just being directly aware of properties in the world, such as color. Doesn't work for all sensation (pain for example), and color realism has its own problems.
Can you expand a bit more? Roughly speaking, someone has to tell you about consciousness before you can experience it? And how does that relate to a challenge of defining it?
Are you a robot? If not you already know. The images you see, the feelings you feel, the smells etc, the fact that the something that gathers all of that into a neat packet to be experienced is the consciousness. Every human feels these things are special and thus experience consciousness but it has yet to be described by science how that works.
Are you contending that somebody who grows up in complete social isolation would not be conscious? If so, that doesn't make sense. And besides, we currently have no way of knowing that consciousness actually exists in any being other than ourselves, so there would be no way to measure it to test your hypothesis.
Until we have some way of doing that, I think it's more rational to assume all humans in a non catatonic state have some level of consciousness.
Could a person who grows up in complete social isolation learn (or invent) a language? If not, could such a person name things or have a way to think about maps and symbols (representations of things) in a way that could lead to consciousness?
Consciousness as a grouping category of many aspects of mental experience is a social construct, for sure. But the subjective phenomena themselves are not.
I think you confuse consciousness with conscience. Apart from that, the Bible contains almost no definitions (and this has caused many problems in the past, even people killing each other as they disagreed on what certain things mean). If one wants precise definitions, they can be found in literature such as Abhidharmakosha.
What mystery? This is the same kind of question as “how does aspirin find a toothache?” It’s based on a whole rat’s nest of false assumptions, and therefore cannot be answered.
Seems to me that the bigger problem with the headline is the the a priori assertion that these are the World's Greatest Minds. But substituting for "Philosophy's Biggest Egos" really detracts from the mystique I guess :p
It's not badly defined. Consciousness is our subjective experience of color, sound, taste, smell, touch, feelings in perception, imagination, hallucination, illusion, dreams and inner dialog. It's something we're all familiar with.
The modern philosophical term for this is qualia. Chalmers, Block, Nagel, etc. have discussed this at length. Even philosophers who disagree with them like Dennett and Keith Frankish do so by engaging with the definition above.
It's unanswerable to date because there is no consensus on what the nature of subjectivity is. How are some brain states conscious? Is it a cognitive illusion? Is it weakly emergent? Is consciousness a property of everything or perhaps rich information streams? Is it functional or biological? Are we cognitively closed to the answer?
> It's not badly defined. Consciousness is our subjective experience of color, sound, taste, smell, touch, feelings in perception, imagination, hallucination, illusion, dreams and inner dialog. It's something we're all familiar with.
I don’t know this doesn’t seem like a great definition to me. It lists a dozen distinct properties, most very complicated in their own right, and then kind of collapses into “we all know what it is anyway”
It is clear, consciousness is what allows you to experience things rather than being a mindless automaton. We can't detect this property nor can we explain it with the laws of physics we have today, but the definition of what it is is extremely clear.
What you mean might be that you'd prefer a definition that you can use to measure objects around the world to detect consciousness? Well, as I said we don't have any laws or explanation for it so you wont get that. But you and everyone else knows that consciousness is a thing and you can prod at your consciousness yourself to see how it responds to stimuli etc, but that is as far as human knowledge go with respect to consciousness so far.
> consciousness is what allows you to experience things rather than being a mindless automaton.
Well this is certainly a better definition IMO, however, what does a "mindless automaton" mean? If the automaton is intelligent then it's not mindless.
> What you mean might be that you'd prefer a definition that you can use to measure objects around the world to detect consciousness?
That's not what I mean. I actually think Turing created a decent method of detecting consciousness. What I mean is that someone on this thread said the problem with consciousness is that it is not well-defined, and I happen to agree in the sense that when people talking about consciousness they are often talking about different things. Just in this one thread 2 different definitions were proposed, and the article uses a 3rd: "the feeling of being inside your head"
> If the automaton is intelligent then it's not mindless.
Does the automation have conscious experiences? Take a simulation of a city populated with characters. Do those characters experience the city? Do they feel the warmth of the digital sun on their digital face?
> I actually think Turing created a decent method of detecting consciousness.
Would the turing machine feel anything during its conversation? Is there anything it's like to be code? All of these kinds of phrases are ways of specifying subjectivity.
Everyone who is conscious is 100% aware that they are conscious. That is the only definition you need. Of course we can't know if others are conscious, but that doesn't mean the definition isn't easy to make. The hard part is tying this definition to physical properties, not the definition itself.
I think part of the problem might be people too often try to come up with some grand unifying theory all at once, without standing on the shoulders of giants. Sort of like trying to write a gigantic codebase all in one sitting (no test suite, no syntax highlighting) and then sending it off for code review without even running it.
This preprint my co-authors and I are working on might offer something of a babystep, not toward consciousness itself but toward the simpler problem of self-reflection, and more specifically of measuring self-reflection in reinforcement learning agents: https://philpapers.org/archive/ALEEET-3.pdf
Not sure how to describe it but I see it like more of a approach and limitations or tech problem.
1- can't study it,in details, live with 0 risk of damage
2- can't test injecting data without risk of damage
3- the live data that we can extract/record (while monitoring input) with current tech is limited. I'm thinking that we're still at the stage comparable with using a multimeter to test a 8 layer pcb, with no schematics, while it's powered on.
I, personally, think science will get there at some point, but we'll need some serious tech advancements to do that.
That's one of the reasons it's so nice to use reinforcement learning agents as guinea pigs. I can "pip install" a bunch of industry-standard RL agents and put them through the most bizarre scenarios, scenarios it would never be possible to put a human through. No ethical issues involved, since these RL agents are nothing more than a bunch of matrix multiplications.
In theory, and this is a pretty airtight theory, it is computation. Because if it isn't, it's magic.
In practice, it is only implemented in one type of animal, or maybe a few types, depending on how it is defined. And we do not have a practical answer for how and why. Partial at best. And not in a way that would enable enough understanding about what kind of computation it is to implement similar-enough computation in a machine.
It feels like the scale of computation needed is now, or soon will be, available. The 100 billion neurons with 100 trillion connections seem daunting until you look at the connection and neuron (and sensory) data rates. It was clearly never going to happen in a Symbolics 3600. It is not clear it could not happen in a modern data center.
That's one answer. Another is that you would be within your rights to draw the inference that the system you are observing is conscious (which is slightly different, and deeply unsatisfying to many).
For Turing, the criteria for intelligence was a matter of observers (if it walks like a duck . . .); what is "actually happening" inside is irrelevant. Searle's objection was that what is happening "inside" is surely not irrelevant. So while some answered that in the Chinese Room argument the "whole system is conscious," others argued that Turing's point still stands. Someone observing the system would have to conclude that the system was intelligent, regardless of what was happening inside.
The two answers are roughly on the same side of the argument, but subtly different in important ways.
Ok thanks. Yeah I think that Turing came up with the best solution for detecting intelligence. If it passes that test then either the system is truly intelligent or a “zombie” which only appears intelligent. As mentioned in some other replies on this post, I think “zombies” are something created by philosophers but do not exist in real life. So i think the observation of intelligence is equivalent to actual intelligence.
The Chinese Room argument is tendentious because scaling a Chinese Room metaphor to a human brain's neurons and connections is hard to imagine. It is still hard-ish to imagine because, currently, it would take a large data center and 4-6 orders of magnitude the power consumption.
The other side of the coin is that we are asked to believe that what happens in an animal brain turns into magic when the animal is human. That is even harder to imagine.
> The other side of the coin is that we are asked to believe that what happens in an animal brain turns into magic when the animal is human. That is even harder to imagine.
Nobody is saying that animals aren't conscious. Or maybe a few are, and maybe human brains have some unique properties, we can't tell since we don't know what we are looking for. But animals display many of the same things as humans, and we know for a fact that humans are conscious so it isn't a stretch to imagine animals being as well.
> magic
Yes, we have no formula to describe consciousness, so as far as we know today it is magic. But if we manage to properly map that out in the future then consciousness will be science.
I think you only regard that theory as "airtight" because you accept the premises of ontological naturalism. What you're basically saying is that it must be this way, because there is no other way to for it to be.
But it remains the case that there is exactly no evidence at all for the belief that consciousness is an emergent property of computation. Not some evidence -- no evidence.
We know factually that there is computation, because we can observe it happening using wires inserted at the inputs and outputs. The burden is on people who insist there is something else besides computation. Thus far, no evidence means, exactly, no phenomenon, and nothing to be explained.
You may be right about where the burden lies, but your explanation renders the thermostat conscious.
Edit: I should put that differently. Your sense of the criteria by which the problem may be considered solved does not preclude thermostat consciousness. And if that is allowed, then yes: the problem is solved. But I sense, again, the assumption of physicalism (or ontological naturalism), and I don't know why that premise admits of no dissent or argument.
When a term is undefined, any application of it is undecidable. Are you conscious? Is the thermostat? Y/Y, Y/N, N/Y, N/N are all equally valid conclusions.
If physicalism is true, then there is nothing here to explain. If physicalism is not true, then there is nothing here to explain.
Biologists used to talk about a "vital force". Alive things had it, dead things didn't. Now biology is mature enough to recognize that there is no such force. People chasing consciousness are in the position of the biologists who insisted, in the face of biochemistry, that there was still a vital force.
There is a physical phenomenon: we observe people insisting there is such a thing as consciousness. Why do they do this? That is something an explanation could address. Not everybody insists. Are people who don't not conscious? What is common to everybody who insists, and also everybody they insist it about? They all observe themselves experiencing, and act on those observations. So, if there is anything to be defined, it is probably defined by that.
We could make a machine that perceives, and perceives itself perceiving. If that machine would not be described by insisters as conscious, I would expect a reason why not.
We know there is computations done in the brain. We don't know whether this is sufficient to create a consciousness. Consciousness doesn't exist as a phenomena in physics, hence there is no scientific explanation how or when it would appear, saying it is an inherent property of computation is therefore nonsense. Humans just know it exists, that is all we can say, everything else is just pseudoscientific bullshit until we can properly tie consciousness to some physics phenomena.
Or in another way, if you say "it is an inherent property of computation", then I'd counter "define computation in terms of laws of physics". You can't. Computation is as ill defined as consciousness from the laws of physics perspective.
I understand what you mean when you say "pseudoscientific bullshit," but it might be better to simply say "non-scientific" in the sense of "beyond the ambit of what science is designed to explain."
The problem of "being" (why there is something rather than nothing) is similar. It's not just that we don't have a physics explanation for why anything exists in the first place; it's that physics may not be able to answer that kind of question at all.
None of that is necessarily "pseudoscience" or necessarily bullshit (though it's easy enough to find examples of both).
I am skeptical of the circularity of the argument, though. There must be a "physics explanation" because physics explains everything. And we know that, because if physics can't explain it, it doesn't exist. And we know that, because physics explains everything.
(Not accusing you in particular of making that argument . . .)
Physics is the field of explaining everything in the universe. Therefore, if physics can't explain a phenomena we don't understand it, as physics is the ultimate explanation we have. I didn't say that if physics can't explain it then it doesn't exist, as you seem to think, just that if physics can't explain it then our understanding of the topic doesn't exist.
No, no. I did not think you are saying that (that if physics doesn't explain something it doesn't exist). That's why I put in that last line.
But now I'm actually not so sure. You seem to not admit of a category of things that physics can't explain. There are things it does not currently explain, things it has not yet explained, things it has explained wrongly, and things it may never explain as an historical matter. But not things it can't explain ("can't" in the sense of "is incapable of").
If physics can't explain it then humans can't explain it. That is my stance.
If there is a separate world outside of this world etc, then that would get incorporated into physics. I don't see how any scientific explanation for consciousness wouldn't become a part of physics.
Basically, physics describes all the rules of this world. Consciousness exists and is a part of the world, so it is a part of the rules of the world, so it is a part of physics.
If you think computation is undefined, you have a much bigger problem than consciousness. We have a physics of information. Computation is operations performed on information. We build machines to do it, on purpose. But a bacterium moving toward a rising gradient of sugar concentration is seen to be computing and acting on the result. The physics involved is in no way mysterious.
No, computation is objectively observed. But not all computation is equally interesting. What is subjective is just how interested you or somebody is in a particular instance.
If what you imagine to be consciousness is a recognizably interesting computation, the details that make it recognizable are objective, but your interest in it as a topic to talk about is purely subjective.
If you increased the size of the brain 1,000 times it would be about the size of Madison Square Garden.[0] Now take any cubic meter of this mega-brain. There will be about 40,000 neurons in each. Good luck figuring out how these work in tandem by themselves. Now what about in tandem with the rest of our Madison Square Garden brain? Now shrink it back down and fit it in our heads again ...
Mapping the synapses of a brain is useless without mapping the behaviour of the connections, they have already mapped flea brains etc but that didn't help them create a flea AI.
Each neuron is unique though, you can't just replace them. So they are more than just switches and memory, they are a part of the brains "software" as you call it.
That sounds like false reasoning, take a computer for example. You can have a computer emulate itself perfectly (a finger touching itself by that analogy).
That's a subprocess and not the thing we started with (that's analogous to me imagining myself... its not the same self that is doing the imagining), the assumption here is consciousness is single threaded.
I like the theory that consciousness is a 5th energy force after gravity, electromagnetism, strong/weak nuclear force that informs inanimate collection of atoms to become living and thoughts, ideas and feeling are packets of information on the conscious field. Our brains are like radio receivers tuning into this collective energy where common thoughts ideas and feelings exist (the collective unconscious Jung talks about.) Each of us has a unique collection of those thoughts and feelings existing as a conscious energy pattern that constitutes our personality.
It seems to me that the way to combine our very real experience of consciousness and monist philosophy is through a kind of cybernetic panpsychism, which is to say that "consciousness" is a range on a sliding scale that includes all feedback-driven processes. That's my angle anyway.
Consciousness (as in: a subjective EXPERIENCE) is IMHO in the same category as "why there is something rather than nothing?".
That is, it falls back to the anthropic principle. If we didn't exist, we wouldn't even be able to consider this question. Or in other words: it is hard to formulate the opposite.
I can be wrong, of course. For long decades scientists considered the hidden variable theory of quantum mechanics ("God does not play dice" - as summarised by Einstein) a philosophical challenge. John Bell was not able to publish his research as it was considered "not even physics". It was surprising that a seemingly philosophical challenge turned into a testable one.
The anthropic principle might help explain the reason, but not the mechanics. You could use the anthropic principle to help explain why we live on a planet in the Goldilocks zone, but that doesn't mean there aren't laws of physics that explain why that's important for life. Likewise, we still might be able to explain the mechanics of consciousness.
Since everybody is sharing their pet theory, mine is that consciousness is a bug, a side effect of having a prefrontal cortex, long term memory and being able to "predict" the future. These tools have important needs for our survival, and come with the unintended effect of asking ourselves "who am I and what is my purpose?" Now it's too late for natural selection to select it out if not beneficial, for we have overridden and grown more powerful, at least in hubris, than Nature.
There is no purpose for humanity. No grand design. We are just replicating lifeforms that have gone self aware. Accept it, it's particularly freeing.
That's what I personally feel too -- some few necessary components combine to give the illusion of self. I would add to this, though, the ability to perceive input from our senses and becoming an "observer", thus creating the separation of oneself from the universe: the self is who observes, everything else is the universe.
This distinction of self from the universe is also talked about in Buddhism as being a massive flaw of human understanding. In reality, all of us are one with the universe and arise from it.
You'd still need a new law of physics to explain that though. Ultimately every phenomena in the universe is explained by physics, so until consciousness is a part of physics we don't understand it.
I'd accept it, but it's illogical. There are just too many accidents needed to make this happen.
Paradoxically, at the same time our minds are so limited they can accept only two possibilities: (1) everything was created by a god, (2) everything happened by accident. My pet peeve is that there is a third option, but our minds are unable to grasp it.
Imagine if we could find a scientific explanation for consciousness, and that would lead to a ton of new discoveries. The world might become more sci-fi than we could ever dream of today.
So it is sad to see so many view this question as impossible to solve, or just give out hand wavy explanations like "computation" etc, rather than try to think about what would actually be required to understand what consciousness is.
> long term memory and being able to "predict" the future
If this were the case, wouldn't drugs that temporarily inhibit the access of long term memory and the ability to predict the future also inhibit consciousness? This doesn't appear to be the case, because high doses of psychedelics and dissociatives do this, but consciousness definitely continues, just in a very odd, altered state.
Certain dolphins have larger prefrontal cortices with more neurons, are highly intelligent with the ability to predict and long term memory yet seem to lack what we think is a unique human capacity for first person introspective meta-cognition. Which is essentially what most and this article mean when talking about the hard problem or qualia.
We can also look at chimpanzees. They have a prefrontal cortex, future prediction and long term memory. While their prefrontal cortex is smaller, it is also the most human like. Yet, they too seemingly lack capacity for that human style of conscious being. There is alas, something more your definition has failed to touch upon.
It looks like everyone on HN either has their own answer, or just ignore that there is a debate and decide that there is no mystery, or that consciousness cannot understand its own mechanism. So yeah, we definitely haven't cracked it...
I think we cannot know anything at all, in the sense of completely understanding and knowing the "essence" of something. All our knowledge is instrumental. We have pragmatic theories and that's all we'll ever have. The existence of a (physical) reality? A (very useful) theory. Existence of "things"? It's the pragmatic way our thinking works.
Consciousness comes first, theories about "reality" come after. Expecting of "solving" the "mystery of consciousness" is futile. Existence is mystical and unknowable.
Still, it's very useful to strive for understanding and explanations, we just have to accept our limitations.
> “Look, I’m not a zombie, and I pray that you’re not a zombie,” Chalmers said, one Sunday before Christmas, “but the point is that evolution could have produced zombies instead of conscious creatures – and it didn’t!”
This statement surprised me. If evolution could have created zombies, and equally possibly could have not created zombies, then consciousness would be a weird and unexplained chance occurrence. Despite having evolution to explain how we came to exist as complex information-processing organism, we would need an additional and entirely different explanation for why we happen to be conscious.
We get a more elegant situation if we suppose that consciousness is a necessary side effect of creatures like us. Which feels no less strange and magical, even if it is simpler, and took me straight to panpsychism, and the question of whether the machines we create might be conscious, and whether we are creating them to have pleasant, unpleasant, or neutral experiences. Equally, since we can see how machines are made of smaller machines and so on, we can wonder if parts of us have consciousness and whether it is the same consciousness as our own, or if a group of people has a consciousness that experiences our world in some way but is unable to make its presence known because it exists at a level unconnected from the consciousnesses that direct human speech and action. Or maybe it has an illusion of agency just like we do.
> We get a more elegant situation if we suppose that consciousness is a necessary side effect of creatures like us
That's the thesis in Blindsight (well, more 'interesting evolutionary incident' than 'necessary') . There might be other intelligent creatures that are unlike us and that are not conscious.
The question there is what exactly, and very specifically do you mean by 'thinking'. We perform very complex intellectual tasks completely subconsciously all the time, without any concious awareness that we're doing them. Driving a car, while listening to the radio for example. Playing football or any complex sport at the highest level is done very largely on autopilot. Grandmasters playing chess, or Go, don't consciously think through every possible permutation of move, yet they can out play extremely sophisticated algorithms.
Huge swathes of the cognitive load on these very sophisticated tasks is subconscious, so clearly those tasks must in principle be doable by entities not capable of consciousness. We don't yet know where the boundaries are between tasks that do and don't involve consciousness, if there are any.
>We get a more elegant situation if we suppose that consciousness is a necessary side effect of creatures like us.
We are concious, so creatures like us must be concious, otherwise they're not 'like' us. This is a conceptual pit trap.
I'm sorry but I don't see this as at all elegant, it's the usual hierarchical inversion error somewhat disguised. A is a kind of B, therefore B is a kind of A. Phrased like that it's obviously false, but we see people doing this all the time. In this case the most common manifestation is conscious systems are complex systems (A is a kind of B where A=conciousness, B=complex), therefore complex systems are conscious (B is a kind of A). Therefore panpsychism.
I suppose the argument is that consciousness might be a generalised attribute of complexity, or slightly more plausibly of information processing systems, but I see no evidence for this, and frankly no reason to believe it might be true. I am very wary of these kinds of hierarchical inversion though, they must be approached an examined extremely carefully because it's a very seductive intellectual failure mode.
111 comments
[ 2.9 ms ] story [ 179 ms ] threadNot everyone agrees with using the term qualia for sensation. There's various debates over the nature of perception. Some see sensation as just being directly aware of properties in the world, such as color. Doesn't work for all sensation (pain for example), and color realism has its own problems.
What does it mean to experience consciousness?
Until we have some way of doing that, I think it's more rational to assume all humans in a non catatonic state have some level of consciousness.
I read GP's comment that in this case that someone wouldn't have a concept of consciousness in his consciousness.
I think it's a massive, massive stretch to make the claim that any of them weren't conscious.
The modern philosophical term for this is qualia. Chalmers, Block, Nagel, etc. have discussed this at length. Even philosophers who disagree with them like Dennett and Keith Frankish do so by engaging with the definition above.
It's unanswerable to date because there is no consensus on what the nature of subjectivity is. How are some brain states conscious? Is it a cognitive illusion? Is it weakly emergent? Is consciousness a property of everything or perhaps rich information streams? Is it functional or biological? Are we cognitively closed to the answer?
I don’t know this doesn’t seem like a great definition to me. It lists a dozen distinct properties, most very complicated in their own right, and then kind of collapses into “we all know what it is anyway”
What you mean might be that you'd prefer a definition that you can use to measure objects around the world to detect consciousness? Well, as I said we don't have any laws or explanation for it so you wont get that. But you and everyone else knows that consciousness is a thing and you can prod at your consciousness yourself to see how it responds to stimuli etc, but that is as far as human knowledge go with respect to consciousness so far.
Well this is certainly a better definition IMO, however, what does a "mindless automaton" mean? If the automaton is intelligent then it's not mindless.
> What you mean might be that you'd prefer a definition that you can use to measure objects around the world to detect consciousness?
That's not what I mean. I actually think Turing created a decent method of detecting consciousness. What I mean is that someone on this thread said the problem with consciousness is that it is not well-defined, and I happen to agree in the sense that when people talking about consciousness they are often talking about different things. Just in this one thread 2 different definitions were proposed, and the article uses a 3rd: "the feeling of being inside your head"
Does the automation have conscious experiences? Take a simulation of a city populated with characters. Do those characters experience the city? Do they feel the warmth of the digital sun on their digital face?
> I actually think Turing created a decent method of detecting consciousness.
Would the turing machine feel anything during its conversation? Is there anything it's like to be code? All of these kinds of phrases are ways of specifying subjectivity.
yes. Consciousness does not need to be a binary phenomenon, there are certainly different levels of consciousness.
> Take a simulation of a city populated with characters. Do those characters experience the city?
yes.
>Do they feel the warmth of the digital sun on their digital face?
No probably not, unless the simulation includes that level of perception for the characters.
> Would the turing machine feel anything during its conversation? Is there anything it's like to be code?
Consistent with my answers above: yes and yes.
Being the subject of experience is not enough. A camera experience color, but almost everybody would agree it is not conscious.
Another question The Guardian might ask: why can't the world greatest minds realize that Communism is the solution to all problems?
This preprint my co-authors and I are working on might offer something of a babystep, not toward consciousness itself but toward the simpler problem of self-reflection, and more specifically of measuring self-reflection in reinforcement learning agents: https://philpapers.org/archive/ALEEET-3.pdf
1- can't study it,in details, live with 0 risk of damage
2- can't test injecting data without risk of damage
3- the live data that we can extract/record (while monitoring input) with current tech is limited. I'm thinking that we're still at the stage comparable with using a multimeter to test a 8 layer pcb, with no schematics, while it's powered on.
I, personally, think science will get there at some point, but we'll need some serious tech advancements to do that.
In practice, it is only implemented in one type of animal, or maybe a few types, depending on how it is defined. And we do not have a practical answer for how and why. Partial at best. And not in a way that would enable enough understanding about what kind of computation it is to implement similar-enough computation in a machine.
It feels like the scale of computation needed is now, or soon will be, available. The 100 billion neurons with 100 trillion connections seem daunting until you look at the connection and neuron (and sensory) data rates. It was clearly never going to happen in a Symbolics 3600. It is not clear it could not happen in a modern data center.
OK, but what is "computation"?
If you perform the computation with a lot of pencils and a lot of paper, is it conscious?
The two answers are roughly on the same side of the argument, but subtly different in important ways.
The other side of the coin is that we are asked to believe that what happens in an animal brain turns into magic when the animal is human. That is even harder to imagine.
Nobody is saying that animals aren't conscious. Or maybe a few are, and maybe human brains have some unique properties, we can't tell since we don't know what we are looking for. But animals display many of the same things as humans, and we know for a fact that humans are conscious so it isn't a stretch to imagine animals being as well.
> magic
Yes, we have no formula to describe consciousness, so as far as we know today it is magic. But if we manage to properly map that out in the future then consciousness will be science.
But it remains the case that there is exactly no evidence at all for the belief that consciousness is an emergent property of computation. Not some evidence -- no evidence.
Edit: I should put that differently. Your sense of the criteria by which the problem may be considered solved does not preclude thermostat consciousness. And if that is allowed, then yes: the problem is solved. But I sense, again, the assumption of physicalism (or ontological naturalism), and I don't know why that premise admits of no dissent or argument.
I'm not setting a trap. I'm just trying to clarify your position.
Biologists used to talk about a "vital force". Alive things had it, dead things didn't. Now biology is mature enough to recognize that there is no such force. People chasing consciousness are in the position of the biologists who insisted, in the face of biochemistry, that there was still a vital force.
There is a physical phenomenon: we observe people insisting there is such a thing as consciousness. Why do they do this? That is something an explanation could address. Not everybody insists. Are people who don't not conscious? What is common to everybody who insists, and also everybody they insist it about? They all observe themselves experiencing, and act on those observations. So, if there is anything to be defined, it is probably defined by that.
We could make a machine that perceives, and perceives itself perceiving. If that machine would not be described by insisters as conscious, I would expect a reason why not.
Or in another way, if you say "it is an inherent property of computation", then I'd counter "define computation in terms of laws of physics". You can't. Computation is as ill defined as consciousness from the laws of physics perspective.
The problem of "being" (why there is something rather than nothing) is similar. It's not just that we don't have a physics explanation for why anything exists in the first place; it's that physics may not be able to answer that kind of question at all.
None of that is necessarily "pseudoscience" or necessarily bullshit (though it's easy enough to find examples of both).
I am skeptical of the circularity of the argument, though. There must be a "physics explanation" because physics explains everything. And we know that, because if physics can't explain it, it doesn't exist. And we know that, because physics explains everything.
(Not accusing you in particular of making that argument . . .)
But now I'm actually not so sure. You seem to not admit of a category of things that physics can't explain. There are things it does not currently explain, things it has not yet explained, things it has explained wrongly, and things it may never explain as an historical matter. But not things it can't explain ("can't" in the sense of "is incapable of").
Do I have you right on that?
If there is a separate world outside of this world etc, then that would get incorporated into physics. I don't see how any scientific explanation for consciousness wouldn't become a part of physics.
Basically, physics describes all the rules of this world. Consciousness exists and is a part of the world, so it is a part of the rules of the world, so it is a part of physics.
How would you define an operation that doesn't just makes every process in the entire world a computation?
Nature doesn't distinguish. But we do, for practical reasons.
If what you imagine to be consciousness is a recognizably interesting computation, the details that make it recognizable are objective, but your interest in it as a topic to talk about is purely subjective.
[0] https://waitbutwhy.com/2017/04/neuralink.html
If you will, the wiring between cells, not so much the cells themselves.
while the cells are just switches with memory, i.e. the hardware
In the words of Alan Watts: You can't touch the tip of your finger with the tip of the same finger, you can't bite your own teeth.
It's an irreducible problem. Lest there is some social way we can achieve it...
Dealt with exactly this.
[0] https://en.wikipedia.org/wiki/A_Glorious_Accident
That is, it falls back to the anthropic principle. If we didn't exist, we wouldn't even be able to consider this question. Or in other words: it is hard to formulate the opposite.
I can be wrong, of course. For long decades scientists considered the hidden variable theory of quantum mechanics ("God does not play dice" - as summarised by Einstein) a philosophical challenge. John Bell was not able to publish his research as it was considered "not even physics". It was surprising that a seemingly philosophical challenge turned into a testable one.
There is no purpose for humanity. No grand design. We are just replicating lifeforms that have gone self aware. Accept it, it's particularly freeing.
This distinction of self from the universe is also talked about in Buddhism as being a massive flaw of human understanding. In reality, all of us are one with the universe and arise from it.
Paradoxically, at the same time our minds are so limited they can accept only two possibilities: (1) everything was created by a god, (2) everything happened by accident. My pet peeve is that there is a third option, but our minds are unable to grasp it.
So it is sad to see so many view this question as impossible to solve, or just give out hand wavy explanations like "computation" etc, rather than try to think about what would actually be required to understand what consciousness is.
If this were the case, wouldn't drugs that temporarily inhibit the access of long term memory and the ability to predict the future also inhibit consciousness? This doesn't appear to be the case, because high doses of psychedelics and dissociatives do this, but consciousness definitely continues, just in a very odd, altered state.
We can also look at chimpanzees. They have a prefrontal cortex, future prediction and long term memory. While their prefrontal cortex is smaller, it is also the most human like. Yet, they too seemingly lack capacity for that human style of conscious being. There is alas, something more your definition has failed to touch upon.
Consciousness comes first, theories about "reality" come after. Expecting of "solving" the "mystery of consciousness" is futile. Existence is mystical and unknowable.
Still, it's very useful to strive for understanding and explanations, we just have to accept our limitations.
This statement surprised me. If evolution could have created zombies, and equally possibly could have not created zombies, then consciousness would be a weird and unexplained chance occurrence. Despite having evolution to explain how we came to exist as complex information-processing organism, we would need an additional and entirely different explanation for why we happen to be conscious.
We get a more elegant situation if we suppose that consciousness is a necessary side effect of creatures like us. Which feels no less strange and magical, even if it is simpler, and took me straight to panpsychism, and the question of whether the machines we create might be conscious, and whether we are creating them to have pleasant, unpleasant, or neutral experiences. Equally, since we can see how machines are made of smaller machines and so on, we can wonder if parts of us have consciousness and whether it is the same consciousness as our own, or if a group of people has a consciousness that experiences our world in some way but is unable to make its presence known because it exists at a level unconnected from the consciousnesses that direct human speech and action. Or maybe it has an illusion of agency just like we do.
That's the thesis in Blindsight (well, more 'interesting evolutionary incident' than 'necessary') . There might be other intelligent creatures that are unlike us and that are not conscious.
Agreed. I never understood the zombie distinction. It seems ridiculous that some thinking machines are “zombies” and some are not.
Huge swathes of the cognitive load on these very sophisticated tasks is subconscious, so clearly those tasks must in principle be doable by entities not capable of consciousness. We don't yet know where the boundaries are between tasks that do and don't involve consciousness, if there are any.
I think the problem with this line of reasoning is that it implies consciousness is binary. There are various levels of awareness and experience.
We are concious, so creatures like us must be concious, otherwise they're not 'like' us. This is a conceptual pit trap.
I'm sorry but I don't see this as at all elegant, it's the usual hierarchical inversion error somewhat disguised. A is a kind of B, therefore B is a kind of A. Phrased like that it's obviously false, but we see people doing this all the time. In this case the most common manifestation is conscious systems are complex systems (A is a kind of B where A=conciousness, B=complex), therefore complex systems are conscious (B is a kind of A). Therefore panpsychism.
I suppose the argument is that consciousness might be a generalised attribute of complexity, or slightly more plausibly of information processing systems, but I see no evidence for this, and frankly no reason to believe it might be true. I am very wary of these kinds of hierarchical inversion though, they must be approached an examined extremely carefully because it's a very seductive intellectual failure mode.
https://news.ycombinator.com/item?id=8923008 [2015-01-21] (89 comments)
https://www.reddit.com/r/cogsci/comments/2t8ctb [2015-01-22] (182 comments)
https://www.reddit.com/r/philosophy/comments/2t6mjz [2015-01-21] (663 comments)
All discussions: https://discussions.xojoc.pw/q/https%3A//www.theguardian.com...
Discussions with similar title: https://discussions.xojoc.pw/q/Why%20can%E2%80%99t%20the%20w...
Archive: https://archive.md/newest/https://www.theguardian.com/scienc...
There is a similarity to using the Turing test to detect intelligence. At the present, this is just another such useless rabbit hole.