Machine consciousness is relevant to ethics. How do you observe the Consciousness of a machine? If you csnt, then you care about something you can't observe. Which could be a counterargument to "rent was supposed to be paid in anticipated experience, not claims about possibility", if you are willing to treat it as unfalsifable.
Don’t illusions invite investigation? My impression was that the various Buddhist sects had explanations, although they don’t seem very satisfactory to an outsider.
OP claims that when adopting the particular Buddhist approach discussed, things which do not exist do not call for explanation. That is what I disagreed with. I did not intend to imply that only Buddhists might want to explain illusions. My point, rather, was implicitly that science would want to explain illusions, but that even if we adopt the Buddhist approach we might be curious regarding the origin of this illusion. And in fact, although I am not conversant enough with Buddhism to quote chapter and verse, I think that some sects of Buddhism do claim to know what causes the illusion of self.
If perspective, self, ego, are illusions, what is it that is deceived? Does an illusion in the very act of deceiving create its own illusory victim? I've long been puzzled by this, and am curious about answers.
I find it quite demoralizing to read yet another speculation about how the Easy Problem of consciousness might be solved, followed by a declaration that these ideas would solve the Hard Problem, as if the basic premise has not been internalized at all. In 100 years, I think people will begin to think seriously about this again, once traumatized thinkers shake off their fear that serious thinking will damn them to religion and woo.
People do think about this seriously, just not in these circles. Bertrand Russell might be a good entry-point for rationalists, given his existing prestige. His "The Analysis of Matter" and "The Analysis of Mind" are a pretty comprehensive survey of his views in the area.
I'm glad this was the first comment, as it made me feel less lonely upon reaching the end of this article.
Rarely have I read something better calculated to seem like an accidental full-on refutation of physicalism, or to stoke concerns that physicalists are p-zombies.
(Note: I don't actually believe they are. They're awe-inspiring human beings like all the others, and I'm so glad the writer is interested and giving this a shot. May truth be his reward.)
I didn't say this solves "the hard problem". In fact I think "the hard problem" is a racket. You can just continue to claim it is unsolved regardless of what happens. Unfortunately this creates negative externalities: It lets people use consciousness as a blanket excuse for them to believe whatever they want to believe, much like epigenetics is misused the same way.
I enjoyed this article - seems like a pretty compelling theory of how conscious thoughts might work mechanically. However, while the oscillating nature of bigger thoughts helps explain why only they are capable of self-reflection, it's less clear if or why lesser, non-oscillating thoughts (or eg computer programs) don't have subjective experience at all - as opposed to just basic/non-reflective experience.
Well, you may think the "hard problem" is a racket, but it's the problem people typically have in mind when they talk about qualia being hard or impossible to explain, or decide that physicalism is inadequate. So if you write a blog post that promises to explain qualia and show that physicalism is adequate, and the only attention you give to the hard problem is to say it's "a racket" and "a blanket excuse to believe whatever [people] want to believe," you can't be surprised when people are underwhelmed.
In fact, it seems to me that if you want to convince people to adopt your views, it's much more important that you focus on convincing them of your claims in this comment than convincing them of the answer to different questions, however interesting they may be (and they were interesting!)
There is a standard dynamics with complicated problems. As soon as some part of the problem is explained and therefore not confusing anymore, people immediately start claiming that it has never been the "real problem" and move the goal post further towards still unsolved parts of the problem, without noticing that fraction of solved to unsolved keeps steady increasing. At some point there is finally nowhere to pass the buck and then the "real problem" is revealed to be nothing more than a semantic confusion and therefore dissolves itself.
Knowing about this dynamics there are two strategies. You can talk about this particular meta-point in attempts to persuade some people who are still confused about it and facsinated about their own consfusion. Or you attempt to solve more and more of the "easy problems" with people who are already understand the meta-point, therefore increasing the ratio of known to unknown, bringing closer the moment where everything is finally resolved.
Well, that's a fair point. But it's hardly as if the "hard problem" is new, or has been in retreat behind a smaller and smaller curtain. In fact, the basic point is far older than Chalmers.
It's not as though people have been vague or unclear about what it is about qualia and consciousness makes physicalism (as standardly understood) implausible, or why qualia seem hard to explain by physics in the first place. It's perfectly fine to declare that you think their concern is a pseudoproblem and you're going to talk about something else that you think is real. But then don't start your article by saying you're going to address their concerns.
> Well, that's a fair point. But it's hardly as if the "hard problem" is new, or has been in retreat behind a smaller and smaller curtain. In fact, the basic point is far older than Chalmers.
Exactly! Chalmers just gave a name for the problem at the state he encountered, but the problem itself is ancient. Long before people were talking about souls and how they, despite their immaterial nature manage to control material bodies, how living organism seem to be behaving by completely different rules than non-living matter. How matter only deterministically follow the rules and therefore can't possibly make decisions under uncertanity. All these seemed to be hard parts of the problem until they weren't anymore and the goalpost shifted.
> It's not as though people have been vague or unclear about what it is about qualia and consciousness makes physicalism (as standardly understood) implausible, or why qualia seem hard to explain by physics in the first place. It's perfectly fine to declare that you think their concern is a pseudoproblem and you're going to talk about something else that you think is real. But then don't start your article by saying you're going to address their concerns.
Initially people who came up with the notion of qualia assumed that they are not physical by definition, while also pointing to some observable referent. There are three options:
1. Qualia indeed are non-physical and exist
2. Qualia do not exist, and the referent they were pointing to is somehow not real.
3. The referent is real but physical, therefore we need to adjust the term.
The post works with the third approach. It shows how all the desired properties of the qualia can be achieved and explained if qualia are physical, therefore addressing some of concerns of people who thought that things with such properties can't possibly be physical. Whether to call it to be part of the hard problem or the easy problem is just semantics.
>Exactly! Chalmers just gave a name for the problem at the state he encountered, but the problem itself is ancient. Long before people were talking about souls and how they, despite their immaterial nature manage to control material bodies, how living organism seem to be behaving by completely different rules than non-living matter. How matter only deterministically follow the rules and therefore can't possibly make decisions under uncertanity. All these seemed to be hard parts of the problem until they weren't anymore and the goalpost shifted.
Actually, for much of philosophical history, there was not much concern with these issues, because they appeared to be easily solved problems. At least in the past millennium, it was mostly post-Descartes and enlightenment that the puzzle began to arise, precisely because people began to have ontologies that were not rich enough to play host to the qualia that are the most obvious aspect of existence.
> The post works with the third approach. It shows how all the desired properties of the qualia can be achieved and explained if qualia are physical, therefore addressing some of concerns of people who thought that things with such properties can't possibly be physical. Whether to call it to be part of the hard problem or the easy problem is just semantics.
Well -- it shows how all of *its* desired properties of qualia can be achieved, I suppose. But it completely ignores what qualia actually are, and where the name comes from -- their qualitative aspect. This is the interesting thing about qualia that seem inexplicable on a physicalist ontology. So it does not even try to make headway on the problem that keeps people thinking that physicalism is inadequate to explain qualia. But it does *promise* to try. And I think it should not have done so.
It's always easy to explain anything with any theory, if you're allowed to redefine the terms you're trying to explain.
>Initially people who came up with the notion of qualia assumed that they are not physical by definition
That is false.
>There are recognizable qualitative characters of the given, which may be repeated in different experiences, and are thus a sort of universals; I call these "qualia." But although such qualia are universals, in the sense of being recognized from one to another experience, they must be distinguished from the properties of objects. Confusion of these two is characteristic of many historical conceptions, as well as of current essence-theories. The quale is directly intuited, given, and is not the subject of any possible error because it is purely subjective.[5]: 121
>. As soon as some part of the problem is explained and therefore not confusing anymore, people immediately start claiming that it has never been the "real problem" and move the goal post further
Or...they just point out that solving part of a problem is not solving the rest of it.
"Physicalist theories have trouble bridging what has been termed the 'explanatory gap'"
> Yes, the hard problem.
"An area that does not give physicalism such trouble is how the mind’s information processing (conscious or not) is based on neuronal activity"
> Yes, the easy problem.
"Philosophically, this explanation seems to obviate alternatives to physicalism, such as idealism and dualism."
> This is saying that solving the easy problem solves the hard problem. The article does have any bearing on idealism, dualism or physicalism.
"You can just continue to claim it is unsolved regardless of what happens."
> If I did that, it would make me annoying, but it would not make it less of a question.
If a fundamental question appears to be insoluble or impractical, one is free to ignore it.
"It lets people use consciousness as a blanket excuse for them to believe whatever they want to believe"
> And there it is. We cannot abide by these daunting, fundamental questions, no matter how glaring, because we cannot allow a loophole where someone can get away with believing in religion or woo.
Yes, exactly. OP's reply here seems (to those of us who find the hard problem weighty, or anyway to me) to be saying, "my belief system is so obviously true and fully adequate that anything it cannot easily answer must be relegated to being a non-question, probably asked in bad faith." There's nothing slightly new about the attitude, of course (including in the context of very different belief systems than his).
Well, it doesn't look like there's a way for currently available methods to even begin interfacing with the hard problem, and yet all non-physicalist paradigms are pretty much discredited these days, so a certain amount of tension is unavoidable. I don't expect much clarity on this until we're close to be able to build conscious minds from scratch.
As soon as those with profound mystical understanding would oblige to turn their paradigms to actually useful ends (but at least tangible will suffice at first), we benighted Westerners might just notice.
Agree. It always seemed like more of a pseudo-problem to me, more a function of philosopher's lack of imagination (or perhaps overactive imagination, I'm not sure). People used to think that figuring out what made life work would forever be mysterious, or that weird forces like elan vital would be needed to account for the miracle of life. Then we made more discoveries and a hard problem turned into an easy one.
For me, when Nagel asks “what is it like to be a bat” he’s collapsing millions of individual moments into a single word — like — and that obscures that each one of those moments can be interrogated and (eventually, I think) explained by hard science.
It also elevates novelty — say, the ability to use sonar after previously being unable to — but to me that’s no different than the novelty of the gap in your mouth when you lose a tooth. Eventually you stop noticing the gap and being “like” yourself is the same as it always was.
So ultimately I think feeling “like” a bat is probably not much different than any other conscious experience.
Not that I know of, but we're starting to make progress in linking qualia to neural activity, even if it's still in coarse resolution. There's no reason in principle that I can think of why we shouldn't be able to develop such a theory once we make more progress on the empirical side.
"Nobody knows for sure how subjective experiences relate to objective physics. That is the main reason why there are serious claims that not everything is physics. It has been called “the most important problem in the biological sciences”, “the last frontier of brain science”, and “as important as anything that can possibly exist” as well as “core to” all value and ethics. So, let’s solve that in a blog post."
How subjective experiences relate to objective physics is the hard problem. If you think the hard problem is a racket, just say so from the get go.
Isn't the whole reason why the Hard Problem of Consciousness is allegedly untractable via materialistic science that it's not about "*how* subjective experiences relate to objective physics" but "*why* subjective experiences relate to objective physics the way it relates"?
That's a decent enough one sentence summary, though depending on how you word it you can switch the "how" and "why" between the two.
The Easy problem of consciousness is figuring out what things happening in the brain are correlated with conscious experience. So you see X happening in the brain, and you know that's correlated with Y experience in the mind. Seems doable, we're getting better at seeing what is happening in the brain, and a lot of what we do know seems to correlate with experiences in the mind (say, for instance, a lesion on Broca's area leading to aphasia).
The Hard problem is figuring out how it's possible for the movement of atoms and electrons, or the processing of information through neurons, to create subjective experience at all. As far as we can tell atoms and electrons don't experience anything, why would putting them in a network change that? Thoughts also have qualities that simply don't map to material objects. For example, a thought can be true or false, or rationally valid or invalid. But what does it mean for a material process to be "true" or "valid"? Certainly we can assign that kind of meaning to material processes; say we build a special sheep gate that flips a flag up and down whenever a sheep passes through it. We could then use that flag to know whether the sheep is in the pen, or has left the pen, and in that sense the flag would be telling us whether the idea "The sheep is in the pen" is true or false. However that assigned meaning is all in our heads. There's nothing in the flag itself, or the system of the gate, that contains the idea "The sheep is in the pen". If someone came along and found the gate they might be able to figure out how it works, but not what it meant to the person who made it because the meaning was in that person's mind, not in the atoms of the gate. Objects and processes don't have meanings, but thoughts do, so if our thoughts are just a very complicated system of material processes then where is the meaning coming from? You can't say that the meaning of the neural process is assigned by our minds, because the mind that assigns meanings is what we're trying to explain in the first place!
> That's a decent enough one sentence summary, though depending on how you word it you can switch the "how" and "why" between the two.
Well, yes, because answers to "how" and "why" are not some separate magisteria. This is why hard problem of consciousness is much less special than some people claim it to be.
> The Easy problem of consciousness is figuring out what things happening in the brain are correlated with conscious experience. So you see X happening in the brain, and you know that's correlated with Y experience in the mind. Seems doable, we're getting better at seeing what is happening in the brain, and a lot of what we do know seems to correlate with experiences in the mind (say, for instance, a lesion on Broca's area leading to aphasia).
Are we talking only about correlations or causations as well? You use the verb "lead" in your lesion example which implies the latter.
> The Hard problem is figuring out how it's possible for the movement of atoms and electrons, or the processing of information through neurons, to create subjective experience at all. As far as we can tell atoms and electrons don't experience anything, why would putting them in a network change that?
When you put it like that, huge part of the Hard Problem sounds just as not understanding some very basic things in general and has little to do with consciousness per se. For example a lot of things together have more degrees of freedom than one single thing. And so it shouldn't be surprising that they can have more properties than a signle thing. One can come up with as many examples from real world as necessary, but at its core its observation about logic, not physics.
> Thoughts also have qualities that simply don't map to material objects. For example, a thought can be true or false, or rationally valid or invalid. But what does it mean for a material process to be "true" or "valid"?
It's ironic because you are using the word "map", while questioning what could it mean for material objects to be "true" or "valid". This also sounds as a confusion about what "truth" means instead of problem related to consciousness. When we understand that it's about the relation of correspondance between things, the mystery dissolves.
> Certainly we can assign that kind of meaning to material processes; say we build a special sheep gate that flips a flag up and down whenever a sheep passes through it. We could then use that flag to know whether the sheep is in the pen, or has left the pen, and in that sense the flag would be telling us whether the idea "The sheep is in the pen" is true or false. However that assigned meaning is all in our heads.
A good example. An important point is that it's not just about which meaning we assign. The correspondance between the state of the flag and sheep being in or outside of the pen is something that other people, unaware of our intentions can notice by observing the state of the pen and the flag. Even if we didn't think to assign this kind of meaning and just so happened to design the system this way, another person may notice the correspondance and point it out. This person doesn't even have to be conscious, by the way.
> There's nothing in the flag itself, or the system of the gate, that contains the idea "The sheep is in the pen"
Not on their own. Only when they are part of the system which opperation preserves the correspondance between the state of the flag and the pen.
Again, none of these things are actually relevant to consciousness. Just a bunch of separate topics, interesting on their own. It initially feels that they has to do something with the mystery of consciousness and the hard problem. But as soon as we try to engage with them they turn out to be their own thing, which we actually already have answers for. And this dynamics creates expectations that other parts of the "hard problem" will follow. That as soon as we've properly factorised the problem and engaged with its parts, the only mystery left will be the "easy problem".
>This also sounds as a confusion about what "truth" means instead of problem related to consciousness. When we understand that it's about the relation of correspondance between things, the mystery dissolves.
This is where you're getting confused. The correspondence between things is part of the Hard problem, because correspondence does not exist in the material world but only in the world of the mind. I have a thought: "The sheep is in the pen." That thought either corresponds to material reality or it doesn't (the sheep is either actually in the pen or not). If it corresponds, it is a "true" thought and if it doesn't it's "false". Yet the correspondence itself only exists in the mind, not in material reality. Even if you solved the Easy problem, and could point to exactly the atoms in my brain that correlate with the thought "The sheep is in the pen" and then observe that atoms that make up the sheep and the pen, you still wouldn't find any "correspondence" in the atoms themselves, only in your own mind as you judge whether a correspondence exists.
Which brings us to the pen again:
>The correspondance between the state of the flag and sheep being in or outside of the pen is something that other people, unaware of our intentions can notice by observing the state of the pen and the flag.
Yes, other minds can decide to assign that correspondence to the sheep and the flag, and assign their own meaning to the gate device that matches the meaning the original inventor had; but they can also assign different meanings to it. Someone might look at the system and decide that it is a method of being notified when grass in the pen has become too short; when the grass is too short, the sheep leaves and the flag changes. This machine is a grassometer! Or they might decide that the device measures whether it is raining or not. You see there is a shed in the pen, and when it's raining the sheep goes into the pen to seek shelter, raising the flag as he does. It's a rainometer!
The reason a mind could come to many interpretations of the device's meaning is that the meaning, and correlations, and purpose of the device don't exist anywhere in the device itself but only in the mind that interprets the device. Nowhere in the device will you find correspondence, only in the minds that observe the device.
So to dismiss the Hard problem as a confusion because the truth value of a thought exists in it's correlation to other things is a bit like a man trapped in a hole twenty feet deep concluding "There's a 20 foot ladder at my house, that will get me out. Problem solved!" The whole problem is that he can't get out of the hole, which means solutions involving being out of the hole are not solutions at all.
I don't accept the claim that how subjective experience arises out of physics is "the" "hard" Problem of consciousness. Because that claim presupposes a distinction between subjective experience and things we can easily study scientifically, and emphasizes the differences between those things, and I'm not persuaded that's a fruitful approach to the matter, unless you want to establish some separate magisterium that the neuroscientists aren't allowed into, as a kind of jobs program for philosophers.
Again, if you want to say that the hard problem of consciousness is ill-founded, then you should say that: but don't say that you've solved the hard problem when you haven't, and especially don't say you've solved it when you think it's not a solvable problem at all.
Well now I'm confused: I thought you don't accept the claim of the hard problem; if you don't accept the hard problem as a valid problem, how can you claim to have solved it?
I think you're wrong to claim that the same neutron can't participate in multiple waves. Consider heterodyne and radio waves. If the wave were just on one sequence of neurons you would be correct, but if it's on multiple neurons, then reinforcing peaks could happen (and, of course, reinforcing troughs). But multiple parallel (within the path) neurons could average things out, and waves could pass right through each other.
I am inclined to be in sync with you on this,( to stick with cycles and oscillations).
I appreciated the hall of mirrors metaphor. It’s not difficult to get lost in one’s own mind. I think of it as a group of singers who can form a choir, or become a terrifying noise. I do think the idea of oscillation and wavelength in the brain is compelling. I think the most unique thing about humans is our ability to willfully direct our attention.
The great 19th-early 20th Century actress Eleonora Duse spoke of self- consciousness quite lucidly in her memoir.
“Duse achieved a unique power of conviction and verity on the stage through intense absorption in the character, "eliminating the self" as she put it, and letting the qualities emerge from within, not imposed through artifice.”
There is something about that that resonates with the tone of this article.
I tend to agree with your premise regardless, but (as I just commented to another reply) I think it's much more defensible to point out that the question of whether there is a Hard Problem at all is one that we can only ask properly when we've gotten much farther on the path of characterizing the physical and electrochemical behavior and nature of the structures that correspond to qualia. Otherwise we're getting into Hitchhiker's Guide Deep Thought territory levels of ill-defined questions.
I suspect that there will be Qualia Engineering degrees, or equivalent, long before there's anything close to consensus on the Hard Problem.
Don't let the hard problem people get you down. In all of these debates, it does seem like the internal awareness becomes the stumbling block. If you do some meditation, and have experienced some of these 'thoughts thinking themselves', then the hard problem doesn't seem that hard. And people that haven't will continue arguing that we just don't get the hard problem. It's back to Plato's cave, you can't argue with the people in the cave.
This was amazing article. Would like to have seen more mention of Bayesian Logic, like maybe the oscillations are somewhat 'updating' their priors based on merging oscillations, and our awareness of phenomena is Bayesian. This seems to be giving a bit of a theory to explain the concepts in 'being you' by Anil Seth.
Some questions, like "why are there beings and not nothing," I expect I will always claim are unsolved, but I the hard problem, at least as I understand it, seems solvable to me, and so I can conceive of, even if I struggle to imagine, hearing an account that purports to solve it and then thinking "this is probably it." So I'd like to understand better why you think that it isn't a real question, or why you think the real part of that question is potentially answered in your post (I'm not sure which one you believe).
As I understand your post, you're telling a story about how a type of physical process in the brain is plausibly the cause of or is conscious thought, and the core case for this is that properties of those physical processes themselves and properties of their relations to each other map onto characteristics generally attributed to qualia. Insofar as this is what the post is doing, I find it interesting, feel that it's quite a bit above my pay grade, and would like to return to it if at any point I become more familiar with modern neuroscience.
However, I don't think this addresses the hard problem, I think the hard problem is a real problem, and I don't think I'm moving any goalposts on you. I think of the hard problem in the context of your theory as "why are oscillations conscious?" or "why is there an experience of the oscillation?" or "why is the oscillation an experience?". "Why, if the oscillation causes or is an experience, is the experience ineffable," on the other hand, seems to be the sort of question this post takes up, and that question seems to take for granted that the oscillation is or corresponds to a conscious thought. The oscillation may very well be a conscious thought or be the correlate to a conscious thought, but is it not a reasonable question to ask how it is that the oscillation is conscious or causes a conscious thing?
When I have a physicalist hat on, I'm thinking of the world as made up of things that aren't conscious. Admittedly, I'm also thinking of a world that's made up of things that aren't individually wet. But when it comes to wetness, I can understand not only that x, y, and z are the conditions necessary and sufficient for this thing to be wet with respect to this other thing, but also *how* x, y, and z cause the phenomenon of wetness. Now, I don't understand that explanation very far down at all. In this case, all I understand is that the wetness of a liquid on a solid is a matter of the ration between the adhesion between the liquid and the solid and the internal cohesion of the liquid, and that the adhesion strength and liquid's cohesion are mostly caused by electromagnetic interaction between the molecules of the two substances. But this is a satisfying "how" explanation to me, even though I don't know the details, and even though I don't know how electromagnetic interaction exists in terms of something that isn't electromagnetically sensitive from the start, because the electromagnetic interaction in this case is a matter of force that can cause motion, and the phenomenon of wetness is easy to analyze into a phenomenon of the motion of the parts of a liquid with respect to a solid surface. So the explanation of wetness, circumscribed in this way, seems unproblematic to me because the phenomenon is clearly about a certain kind of group motion (the group being the molecules of the liquid) and the explanation refers to a cause of motion.
In the case of conscious thought, however, there doesn't seem to me to be this continuity of principle, for lack of a better phrase. I just don't understand how the oscillation is conscious, or where that property comes from. This is in no way an objection to the claim that the right sort of neuronal oscillation is *the* physical source or aspect of conscious thought, even that this is the sort of thing without which there is no conscious thought, and with which there is conscious thought. I don't contest that, and my question to you I don't think has that much bearing on that question. My question is that, assuming it indeed is the physical source or aspect of conscious thought, how is it that it becomes conscious? The parts don't seem to be conscious, and conscious thoughts don't seem to me to be analyzable into spatial objects that can move, so how is it that this system of moving things in space becomes conscious when it is a chain of rhythmically firing neurons?
Why do you think this question doesn't make sense?
I think the way to bridge the apparent gap between physical configurations of atoms and experience is information processing. Nobody doubts that certain configurations of atoms can process information. I don't think anyone doubts that qualia are some specific type of information.
The question of how qualia arise from physics is absolutely a real question and I am trying to answer it. I just really dislike how proponents of this "hard question" outright define it as "hard" by emphasizing the differences between known physics and subjective experience, and by claiming that subjective experience is so different that it needs to be an entirely separate category of thing where it is not obvious if the natural sciences and known epistemology would even apply. They seem to me to be more busy insisting that there is a problem rather than working on a solution.
Your comparison with wetness is delicious! I am definitely stealing that one. Did you come up with that yourself or can you name someone who came up with it?
I don't think this is that far above your pay grade. You gave a more thoughtful comment than most here.
I wish I could like comments on here, but anyway, thanks for the post and for engaging in discussion. I find discussions of qualia to quickly descend into little more than smug navel gazing (a "racket" indeed!), so I appreciate your grounded approach.
Cui bono? is the biggest problem in all of this. Let’s say we discover the grand secret of consciousness and can showcase your every thought inside an fMRI machine or whatnot. What’s next? How does this benefit anyone other than philosophers?
Isn't that a fully general argument about all science other than applications? Let's say we find a consistent theory of quantum gravity. What's next? How does this benefit anyone other than physicists? Let's say we prove the Riemann hypothesis. What's next? How does this benefit anyone other than mathematicians? Let's say we decipher the Phaistos Disc. What's next? How does this benefit anyone other than palaeographers?
Is this meant seriously? No interest or value whatsoever in truth for its own sake? Or in anything other that what tangibly benifits one's own group?
Do you also wonder why anyone would ever study something unless it will make them lots of money? Or why ever help out a friend if you can't see how it benefits you personally?
The question to every proposal is solely "how does it benefit me?"
More in terms of: "this is why nobody funds it seriously". I'm personally highly interested in the question and care about it but also understand that my wishes are not what the market cares about .
I find it quite demoralizing to read yet another comment trying to force all discussion of consciousness onto the procrustean bed of "easy vs hard problems," as if the basic assumption of the dichotomy is above questioning. In 100 years, I think people will have on average continued the trend of shaking off mysticism and woo.
I'm not forcing anyone to talk about the hard problem. All one has to do is not put an article in my inbox that posits "physicalism" as an alternative to "idealism" because they came up with some speculations about the mechanics of the brain--which rather suggests a lack of apprehension of the topic at hand--and I won't leave a comment. Just say that you're thinking about the mechanics of how conscious behaviour arises in the brain, and ignore the hard problem. Say that it's an impractical problem, or pointless to think about, much like asking where the universe (or the laws that gave rise to it) came from.
But I think a lot of people don't actually believe that. At some level, they are vaguely aware that there's a huge question mark hanging in the air. And if they don't say something about it, then souls, ghosts, magic and Bibles will start pouring through the crack in the door. So they cap off the article with a glib comment about how detailing the physical processes that produce conscious behaviour removes the need to believe in a soul. Sorry, I mean "is an alternative to idealism and dualism."
I agree. There will be less mysticism and woo, and people will be able to think more clearly about this, because they will not be so afraid of stirring up belief in mysticism and woo by doing so.
Halfway through the article I had a feeling I was wasting my time and and scrolled down to the comments, thanks for saving me from reading the rest of it! People in this community too often conflate finding physical correlates of qualia with giving a mechanistic explanation of how the physical can give rise to the mental.
What amazes me is how these people seem to tie themselves into knots to deceive themselves into believing that they solved the problem of consciousness!
I agree that the post does this, but also, the Hard Problem is (often and in part, but not always and only) an assertion of its own unsolvability. Historically, these kinds of assertions get resolved by noticing how and why the idea that seems to result in unsolvability is a confusion. I'm not sure quite what that looks like, of course, but I think any proposed serious attempt at resolution seems likely to have certain properties:
1) Ability to read qualia out of existing minds.
2) Insertion of arbitrary qualia into existing minds.
3) Manipulation of existing qualia within existing minds.
4) Transmission of qualia between existing minds.
5) Determination of which of the claimed necessary properties of a quale can be separated out, or altered.
6) Ability to repeat 1-5 within and between increasingly dissimilar minds (animal or artificial).
The type of approach outline in this post - improved ability to identify and interact with the electrochemical structures in the brain that do, in fact, correspond to qualia in the mind - seems like an essential step towards most of those capabilities I listed, which will let us better map out and characterize the nature and behavior of qualia and discuss what physical systems can generate or support what qualia. Maybe, when we walk that path and get to that point, we'll be able to really say whether there is still a Hard Problem to be resolved at all, but my guess is there won't be. Not everyone will find the solution satisfying, but that fact is not itself problematic.
You're not ignoring the Hard Problem if you say: We're not ready to answer this, we don't have the right tools, but we're building some better ones, and these are the kinds of tools we'll need to get ready to ask the actual right questions."
Random aside: I don't remember when, but I once saw a post somewhere asking what would be the most surprising fact someone from the future could tell you they'd discovered and confirmed. My answer (for a while) has been: "Actually, everyone has the same favorite color-quale, but we differ in which frequency combinations (or patterns of retinal stimulation) our brains map to that quale."
I don't see how my theory would suggest a way to insert arbitrary qualia. It does, however, conceivably lead into a way to disrupt existing qualia, through something like well-targeted transcranial magnetic stimulation, if it is timed correctly to disrupt the rhythm of the oscillation that contains that contains the qualia in question. Interesting angle, I had not thought of that. Thank you.
I agree that I don't have any idea what tools or methods could insert arbitrary qualia, but I think it should be possible-in-principle with sufficiently precise tools to induce patterns of neural firing.
You're right that disruptions seems much more achievable than manipulation or insertion
Recently I had brain surgery and in preparation for that, the surgeons determined exactly where inside the tumor my left-side motor neurons (which the surgeons needed, and later managed, to avoid cutting) were located. They did transcranial magnetic stimulation to induce activity in various neurons. The quale of them hitting the right motor neurons that was an impulse to move my fingers, the strength of which I had to verbally report. I could literally feel in my awareness how they (and the model of my brain they were building) homed in on the right spot.
They said they even found of the synaptic weights of the individual axons involved. They can do that, with a lot of time and money. Good news for Uploading.
Right? Light sees light. You cannot scalpel an existence that sings to us. It must be received and that is the greatest evidence. Knowledge can be a powerful tool, here we see the flip side of that power. An uncountable number of humans have had meaningful answers to these questions for millennea, yet so many great thinkers have not built up their strength and tolerance for the brightness.
The "hard problem" arises from the assumption that the difference between qualia and other information being processed in the brain is an ontological one; that qualia are an entirely separate category of thing. I do not share this assumption; I believe the differences (which I enumerated) are explainable by them being processed on their own separate information channel that is internal to an oscillation. So I don't posit an ontological difference on need of an explanation, which would indeed be "hard" if that ontological maneuver was the correct way to go about things - but it is not.
Which basic premise is not internalized? You're making it sound like the premise is that the "hard" problem is just supposed to be not solvable by looking at brains.
Back in the second wave of AI in 1980's, most AI research (at MIT, Stanford, and elsewhere) was done using Lisp programs. It was a standard exercise for advanced undergraduates to write Lisp programs that would inspect and perform operations on their own code and running state, including both internal state and data that symbolically represented external states of the world. These programs straightforwardly fulfilled the description of being "recursively reflective."
I don't for one instant believe that these programs were conscious (unless one adopts a panpsychism according to which lots of things are conscious that we don't normally consider to be, which would contradict the spirit of OP's proposal).
I agree with the OP's basic idea, and I think that we do indeed have to bite that bullet. Those LISP programs were not smart and had only a very limited consciousness, but I do think we will ultimately have to say that they had consciousness.
It's thinking like this that will in the end bring everyone around to my view that consciousness is a bit of a red herring, and actually what makes our human experience distinctive is desire and intention. Consciousness detached from desire is a weird experience, as any Buddhist will tell you. Consciousness (that can't be turned off), coupled with the fact that we want stuff, is the everyday experience we all have.
This is an important implication of OP's proposal to expose: that some Lisp programs commonly written by undergraduates in the 1980's were indeed conscious, to however limited a degree.
Many of the early fans of IIT were turned off by the demonstration by Scott Aaronson that IIT would declare many systems conscious that we overwhelmingly do not normally believe are conscious. In that sense, the theory doesn't seem to fit the phenomenon it aims to describe. My observation is that "recursive reflection" is subject to the same criticism.
Yep, I completely agree with that. But I think the problem is that the word "consciousness" has been a bit mythologised by the likes of Thomas Nagel and David Chalmers. Honestly, half the time I see the word "consciousness" these days, I think it's just standing in for "soul". LISP programs aren't human-like, and they definitely don't have soul. Ergo they can't be conscious.
What the OP is doing, rightly, I think, is attempting to define a more constrained notion of what consciousness is. This should lead to more clarity over exactly what things are conscious and what things aren't, and then on to a better demarcation of what exactly distinguishes us from everything else.
Of course, the OP's definition could be wrong or bad in some way. But I think the project is definitely worthwhile.
The move you're describing simply contributes to the confusion. Whatever problems there may be around consciousness, one doesn't helpfully address them by trying to correlate consciousness to some physical specification that does not in fact match the phenomenon.
I'm all for empirical study of the neural correlates of consciousness. But theoretical claims that this or that physical process (whether IIT or recursive reflection) "just are" consciousness are not merely mistaken but bogus.
I was going to argue, but... I don't think we disagree on anything important. If you're all for the empirical study, and I'm all for the empirical study, then let's let those who can go and do those empirical studies. I also agree with you that there will be lots of philosophical work to be done after the empirical work has been done.
I didn't say the internal information of some conscious thoughts "just is" phenomenal consciousness. I demonstrated it has all (to my knowledge) the properties of phenomenal consciousness that phenomenology hast established.
Yes it does get used as some secularized stand-in for the soul a lot. That contributes to the confusion. That's why I warn in the beginning not to identify with it.
Unlike IIT, Recursive Reflections (RR once it is worth its own abbreviation?) does not claim a thermostat is conscious. It does claim that LORETA could conclusively establish whether (some thoughts of) cows or seaslugs are.
> I don't for one instant believe that these programs were conscious (unless one adopts a panpsychism according to which lots of things are conscious that we don't normally consider to be, which would contradict the spirit of your claim).
You do not need to adopt panpsychism to believe that. You can adopt much weaker claim: execution-of-some-algoritms-psychism.
"Execution-of-some-algorithms-psychism" is already a variety of panpsychism. Every serious version of panpsychism proposes that there are gradations of consciousness depending on the particular properties of the system in question. "Some algorithms" is just a particular suggestion regarding those properties.
No, it's not. What you are describing is panpsychism-light: everything has some non-zero amount of consciousness, and some things based on some principle have "more consciousness" than others.
Execution-of-some-algorithms-psychism, on the other hand, is a weaker claim than that. According to it, only things that execute an algorithm from a particular class of algorithms have non-zero amount of consciousness.
I see this as a quite secondary distinction, almost hair-splitting. You are pointing to the difference between "literally everything" and "lots of things, picked out my criterion, including many things that we don't normally consider conscious." I consider the important bit here to be "including many things that we don't normally consider conscious." If you don't want to call this "panpsychism" because it's not literally "pan," fine; I will concede the terminological point. But my substantive point remains.
Yup - If Materialism is True, Microsoft is conscious, the United States is conscious, so the United Nations is probably conscious, and I think we can assume the biosphere more broadly is too. Of course, Earth cannot be the only physical matter exhibiting recursive reflexivity, so it's likely that the Solar System, the local group of stars, and the Milky Way is conscious too. From there, it's save to assume the galactic supercluster is conscious, nestled within the consciousness of the Universe as a whole.
The Universal Consciousness. Hmm - anyone got a name for what to call this?
I think there's plenty of scope for argument there. Firstly, I do think it would be reasonable to say Microsoft experiences qualia, in the form of sales and R&D outcomes.
But I doubt that MS is conscious, because of a speed issue. In order to be conscious, an entity has to be able to have multiple subsystems focusing on (recursions of) a qualium at the same time. The oscillation frequencies for MS are measured in days, as its neurons (people) coordinate to receive information (hold meetings), and their ability to get in sync is quite limited.
The same goes for the USA and other large groups of people.
(That said, I put this argument to Schwitzgebel, and he didn't accept it, so I could be wrong!)
Right - but I don't think it's an impasse at all. It only looks like an impasse because a load of mystical nonsense has been loaded onto the word 'qualia,' so it's now not just the way in which one material system responds to a material stimulus, it's the SuBjEcTiVe eXpErIeNcE of that material stimulus. When that move has been made, any time anyone says, "Oh, look, we can now track what happens in the brain when you see red," philosophical heavyweights can jump out and say, "But of course that's not the qualium of red, the qualium of red is quite different..." They've introduced the ineffable into the argument by definition, hidden inside clever sounding words like "subjective".
But we don't have to accept that framework. You can say, look, clearly Microsoft experiences things (like takeovers and sales). And clearly MS is not like us. So what are the differences, then? And as I say, I think there are still loads.
One is that the kind of qualia MS experiences are very different to what we experience. A company is fundamentally a financial and legal entity, so I think its qualia are limited to the financial and legal things that happen to it. The people inside MS can see colours and feel emotions (er, I think), but MS itself doesn't have the apparatus to do that. So it's living in a very different perceptual world.
Another is that it's absurdly simple. MS has tens of thousands of employees, who I reckon are roughly equivalent to its neurons; our brains have billions. MS is a nematode at best. (Plus those neurons are much slower than ours - you can sent an email about sales at the rate of what? One every five minutes? - our neurons work on the millisecond level.)
And a third is that speed/organisational issue I mentioned: MS isn't organised in such a way that its neurons can reflect on its own perceptions at high levels of efficiency.
If you don't accept that there has to be a spooky element to the definition of qualia, then you can look with much clearer eyes at everything downstream of them. Which is what I think the OP is trying to do.
Writing "subjective experience" with alternating capitalization is not going to make me think that a financial statement written in a book is in any sense the same "thing" as seeing the color red. The neuron that fires when you say red *really isn't* the same thing as the experience of seeing red. If the financial statement is "qualia" then "qualia" is meaningless.
Or... I can "know" that an apple is red separate from "experiencing" redness. A company clearly "knows" what it's profits are if any employees know or if it's written down. How does the company "experience" profit in a way different from "knowing" it? How could we tell?
Sure. By the same token, putting asterisks around really isn't won't make me think that experience and physical events aren't the same. My typographical emphasis - and yours - does not constitute an argument. My argument was in the the four paragraphs that follow.
Look, I don't believe in ghosts or souls or any immaterial stuff. More importantly - much more importantly - I'm methodologically committed to believing that everything is in the material stuff. I think you have to make that commitment in order to do science properly, don't you? Therefore, assuming from the beginning that a neuron firing cannot be the experience is a non-starter for me. I genuinely don't know whether experience consists in neurons firing, or patterns of neurons firing, or the secretion of some chemical, or what. But I am committed to assuming that it's one of those things (or some other physical event in the brain), for the purposes of doing the science (other people doing it, I'm not qualified).
As an analogy, think of the discovery of DNA. They thought it was crazy that all of human heredity could be encoded onto such a simple molecule. But it would have been a mistake to dismiss the discovery based on the an argument like, "a long string of four bases *really isn't* the same thing as a human being".
The word qualia isn't meaningless, it's just not supernatural. Qualia consist in the specific events in the brain that are associated with specific external stimuli.
Not to be too dismissive - it's a really cool elaboration on and testable prediction for the so-called easy problem. Maybe we're going to get to the point where we call them the "Scientific problems of consciousness" and the "Philosophical problems of consciousness". Maybe we'll get to a point where we believe Physicalism is non-falsifiable and let them duke it out with the Neutral Monists, Dualists, Panpsychists, and others over which substance philosophy wins while neuroscientists solve all the useful problems.
I am confused by this easy / hard thing. It seems to have been invented by people begging the question? 'Look, consciousness will definitely never be explained by physicalist theories. So all physicalist explanations are easy, but the real problem is hard.'
If you're ignorant of consciousness, then you're not allowed to rule out physicalism. Especially since physicalism is the most promising philosophy as OP points out.
If you disagree, can you taboo the hard / easy distinction and write specific criticisms of the post?
"hard / easy" is "explain intrinsic properties / explain structural properties". To have charge, for instance, is to stand in a particular relation to states of the electromagnetic field, and to be an electromagnetic field in some state is to stand in a particular relation to charged objects, and so on, all the way down to raw correlations between sensor readouts. None of this tells us what an electron *actually is*. Structuralism about physics gives us an out here. An electron *actually is*n't anything other than a marker for a particular part of the structure of physics as a whole.
Qualia, on the other hand, have nonrelational properties - redness is not just not-green-or-blue-ness: after all, suffering isn't greenness or blueness either. If you believe in other things with intrinsic properties (the folk-physics notion of "substances" that you can "really touch", for instance) then explaining how those come about is also a "hard problem", it's just that almost no one does anymore.
Can you give an example of any analogous "hard" problem from any field that has ever been solved to your satisfaction, and/or a description of what sort-of-thing would count as a valid solution?
I don't think the hard problem is fully solvable in the way that physical problems are, for exactly the same reason that "why does physics exist" is not solvable. They might *have solutions* - Tegmark's mathematical universe hypothesis could conceivably be right, for instance - we might even find them and all agree that they're probably right, but we'll never really know for sure.
I think this reflects a lack of clarity on what "knowledge" means in this context. What would it even mean to know something "for sure" as opposed to thinking something is "probably right"?
Sure, I'll be more precise. Whereas we can hopefully keep accumulating evidence (in the bayesian sense) in favor of our best physical theories forever, the amount of evidence supporting any particular solution to the hard problem that we will ever be able to access is bounded, and I suspect that the bound is tight enough to matter in practice. We will never reach a point where we're all comfortable rounding it off to "definitely right" in the same way that we (modulo crackpots) are with some scientific theories.
I don't think the divide is as stark as @Amicus does. I think that physics experiments really do tell us something about what an electron really is. So do experiments about quasiparticles and virtual particles, which are like real particles in some ways but not others. Given that, I have quite a few examples I see as analogous.
Elan vital. Resolved by molecular biology.
Phlogiston. Resolved by statistical mechanics.
Zeno's paradoxes. Resolved by transfinite set theory and formal underpinnings of the real numbers.
Aristotle's actual and potential infinities. Resolved in part by calculus and set theory and other higher math, in part by admitting there's no actual paradoxical outcomes.
Aether (as medium of light transmission). Resolved by special relativity and the discovery of photons.
For me personally, free will is in this category. Resolved by asking, "What is it I want from the idea of free will? Which of the possible forms of solution (natural deterministic, supernatural deterministic, random) are capable in principle of providing that? Does the world enable such mechanisms to exist? Do I instantiate them?"
I'm unclear what your listed solutions did for your listed problems that this article isn't doing (or at least working towards doing) for consciousness.
First of all, I gather from their listed solutions that they find the Hard Problem to be a distraction and believe that as our understanding of physical laws improves, we’ll realize that consciousness is perfectly explainable.
Second of all, this post is different from the examples above in that it provides no explanation for why the physical processes it describes would cause us to have some first-person subjective experience. While it predicts things that correlate with subjective experience, there would still be a question of why the experiencer exists. With elan vital, there’s nothing else to explain once you understand biological mechanics.
Sorry, I didn't specify in my comment, I'm one of the readers that *does* think this post lays out a plausible path towards a solution for the Hard Problem.
Exactly. I do claim that demonstrating why (something with all the properties of) subjective experience arises out of physics does constitute a solution.
> None of this tells us what an electron *actually is*.
You seem to reject physicalism altogether. Seems you are looking for for the Ding an Sich behind phenomena. On the other hand, I believe that's all an electron actually is, to the extent that you cut Reality that way.
We may perceive qualia as non relational; that doesn't mean they are 'nonrelational entities'. It simply means we must explain this perception and how it emerges from simpler mechanisms.
Similarly, you perceive yourself as One, even though you are an aggregate of cells and neural patterns etc. That doesn't mean that you are intrinsically one and indivisible despite physics. How the brain/body perceives itself as one in time and space is something to explain.
Yes and no, depending on how narrow the sense of physicalism you're using is. I think neutral monism is more plausible than dualism, while eliminativist materialism is somewhere in "there are true contradictions" territory. Whether there's any real distinction between neutral monism and physicalism depends on what you're willing to call physical - if you're a scientific realist in the conventional sense then they're arguably the same thing, if you're an ontic structuralist then maybe not.
> Seems you are looking for for the Ding an Sich behind phenomena. On the other hand, I believe that's all an electron actually is, to the extent that you cut Reality that way.
The opposite, really - qualia are things in themselves, and I want to know how they relate to the phenomenal world of positions and times and masses and charges. I agree that there's probably no electron-essence out there in the universe.
> It simply means we must explain this perception
I agree, although I'm sure we're using perception in different senses - the thing to be explained is not the self-report, it's the correspondence between the self-report and the actual thing being reported. The moon, not the finger pointing at it.
> Similarly, you perceive yourself as One, even though you are an aggregate of cells and neural patterns etc.
Well, no, I don't perceive myself to be anything in the relevant sense. I intuitively believe it, like everyone else, but the thing being perceived isn't "I am a coherent entity", it's "what-it's-like-to-believe-I-am-a-coherent-entity".
Is there such a thing as a nonrelational property? I am doubtful. To enter into by way of your claim on qualia, as nonrelationality being something other than a simple negation of other states, "redness is not just not-green-or-blueness: after all, suffering isn't green or blueness either" .. we would find that rather the degree is complexity, from a simple relation (red, not other inputs of same category) to a complex relation (suffering as related to red via routing through a virtual representation of a self-world model.) In other words, all perceiving, including all potential knowledge, exist within an contextual process. 'The electron in itself' may be a concept that is reifying a proposed decontextualization that is, per the medium of experience, not possible. There's also the experiment comparing minimal perceptions between red #24 and #25. Next to each other, each quale (shade) is perceivable, but without relation, the quale itself disappears, and no longer exists. This would mean quale is itself functionally a process of relation. Or, take the Ganzfeld Effect -- the lack of context dissolves the quale.
How sure are you of that? We only know that any given thing exists (or does not exist) by observing its interactions with other things in the same state of being, or extrapolating from patterns of such interactions previously observed. If something were irreversibly causally isolated, https://orionsarm.com/eg-article/5062fd4d7f5cb banished from our future light cone, what distinguishes from it having ceased to exist - or from the banished entity's perspective, what distinguishes that state from the remainder of the formerly-observable universe having ceased to exist?
Abstract things have being, and they interact with non-abstract things. Being is not about "existing" in the concrete sense, but more about a ground-level something that all other properties rest on. Also I am unsure what the particle horizon has to do anything. This isn't about "human perspective", it's about metaphysics. Things have being irrespective of human (or any other) observer. I would posit a counterargument of you looking at an object, then gouging out your eyes. Does this mean that the object is now not able to be seen in general? Has its properties been altered in some fundamental way? Particle horizons arent magic and don't physically affect the objects.
> We only know that any given thing exists (or does not exist) by observing its interactions with other things in the same state of being, or extrapolating from patterns of such interactions previously observed.
Is this the same point as in Erogamer quote below?
> "No, wait," Charles said, once his mind had caught up. "Wait, I understand why you can't ask how fast time is going without being outside of time, but how is my existing or not existing like that? That's an absolute if anything is an absolute! Cogito ergo sum, I think therefore I am!"
> "Yeah?" said the preacher, turning his head again. "You, whoever you are, mister…"
> "Charles," said Charles.
> "Mister Charles, suppose I told you that you, right now, didn't exist. Say, you're only an animation in a Disney movie. How could you tell?"
> "I… what?" said Charles. "If I know literally anything, I know that's false! I have to exist in order to ask whether or not I exist!"
> "Aha!" said the grey-bearded preacher, holding up a finger. "But maybe you don't exist and therefore you're not asking whether you exist. How can you tell the difference? If you were an animated drawing, you'd be saying just the same thing."
> "That… honestly sounds to me like nonsense," said Charles.
> "Exactly, mister gentleman Charles sir! It is nonsense just as you say! Maybe somebody outside of existence thinks that we don't exist, but from the inside of reality we've got no way to know that and no reason to care. It's the same way with Disney princesses. You can say all you want that Snow White doesn't exist, but she can't hear you, so as far as she knows, she exists. She's got just as much proof of her own existence as you have."
---------------
> “You claimed yourself that Snow White exists as much as we do,” said Sonia. “Then why don’t we run into her at the corner shop? Why do I find chocolate biscuits there, instead of Snow White? There is a story in which the two of us meet, and the people within that story have no way of knowing themselves to be unreal. And yet I find myself discovering chocolate biscuits instead. Clearly, there is some factor that the possibility containing myself and chocolate biscuits possesses in greater quantity, compared to the less real possibility containing myself and Snow White. Even if we reply to the great question by answering that nothing exists, some zeroes are greater than other zeroes and quantitative ratios may be established between them. Like any other self-observing structure, Snow White finds herself to be exactly as real as herself, a ratio of one to one, and in this sense her existence is locally an absolute. But to say this does not say whether Snow White is more or less encounterable than other things. There is some quantitative degree to which our universe is looking more towards the chocolate biscuits. Some essence of how much something is observed or experienced, which the chocolate biscuits have more of than Snow White. We could call it quintessence, or propensity, or mana, or the blood of God.”
> “Or bullshit,” Charles said.
> “I suppose that term is as good as any other,” Sonia said. “By whatever name, it is the single, sole, and only truly universal form of money. And we can extend the same logic further. Having postulated the notion of bullshit, it would follow that things are more real only to the extent that they are, in some sense, more bullshit. Or rather, by definition, anything that makes a possibility more encounterable is exactly what we are calling bullshit. Possibilities are experienced by conscious beings in exact proportion to the total bullshit breathing fire into those possibilities. Then to whatever extent a mathematical model is not describing bullshit, it is mere math. Only to the extent that a mathematical structure is about bullshit does that structure form an encounterable part of its universe. It follows that every sapient species, if they investigate the physics of their universe far enough, will eventually find a level at which physics seems solely to describe the arrangement of pure bullshit—some physical quantity whose apparent meaning is making possibilities more encounterable. By your own argument, the presence or absence of bullshit must be falsifiable for anyone inside the universe to notice a difference. Then the eventually-discovered laws will show that variations in bullshit are experimentally observable and cause other events to proceed differently. Indeed, variations in bullshit will be the only causally potent factor. Across every universe with conscious life, any inquisition into physical law, sufficiently advanced, will render everything into bullshit, which is and must be the sole constituent of experienceable causality.”
> “Can you simplify that?” said Charles.
> “No, but I’ll do it anyways,” said Sonia. “Whatever it is that makes one possibility more encounterable than another, we are calling that bullshit by definition. Then whatever bullshit is, if something isn’t made of it, we won’t encounter it. And if we couldn’t detect something experimentally, we’d have no way of encountering it. So bullshit has to appear to us as a quantity in our equations—in fact, the only thing structured by our equations, because we can’t encounter anything else. ‘By convention there is sweetness, by convention bitterness, by convention color, in reality only atoms and the void.’ And underneath atoms and the void, it’s all just bullshit.”
5 is the crux, though I would phrase it as "only partially analyzable" rather than "unanalyzable" - qualia have both intrinsic and relational properties.
Much less importantly, I think many of the others are wrong, irrelevant, or overgeneralizing from typical human mental states:
1. probably false, we don't have perfect introspection.
2. I'm not sure it's even meaningful to talk about a quale's duration in an objective sense. Clearly some *seem* to take longer than others, but that's just another quale!
3. again, who knows?
4. Depends on what you mean by "communicated". Obviously we can refer to them, but we can't offer an exact characterization in completely non-phenomenal terms.
6. I think this is probably tautological, the lack of joint qualia grounds our belief that persons are separate to begin with.
7. Definitely false, we experience things without introspective awareness all the time.
8. More a fact about human cognition than a fact about qualia
9. ¯\_(ツ)_/¯
10. ¯¯¯\__(ツ)__/¯¯¯
11. Unclear what this means. Obviously you can't choose not to experience the thing you're already experiencing, but directing your attention to change what you experience next is a totally mundane process we all do all the time.
12. Category error, qualia aren't the sorts of things that can be "used"
13. False
14. Unclear. Epistemologically convenient if true, but nature never promised us that.
Im pretty sure all qualia can be distinguished in purely relational terms. This should be obvious from the fact that we can give them different names, just follow the causal chains back from there. For your example, red, blue, and green occur with a position in the visual field, while suffering doesnt.
Well of course all the qualia we can assign different names to can be distinguished in relational terms: naming is a physical process. But finding a difference between two things is not the same as offering a full characterization of either.
We need to distinguish here between "electrons" as roles in a physical theory and "electrons" as the real objects, if any, that play those roles. Let's call these structure-electrons and substance-electrons. Or, since we don't want to get sidetracked by the question of whether *ontic* (as opposed to epistemic) structuralism about physics is correct, let's move to a domain where it's definitely not, and talk about phonons.
A structure-phonon is a noninteracting boson, with creation operators given by, up to constants, a_k = sqrt(w_k) * (q_k + ip_{-k}/w_{k}), where q and p are positions and momenta (by which I mean only that they satisfy the appropriate canonical commutation relation).
A substance-phonon is a quantized sound wave: the collective motion of atoms in a crystal lattice. And their positions and momenta above are not the positions and momenta of atoms - they're fourier transforms of the position/momentum vectors of the whole lattice.
The properties of phonons can all be defined in terms of each other (giving you a structure-phonon) but what they really are are *relations* between more fundamental physical properties like the positions of atoms. Now step one level down - fundamental physical properties can be defined in terms of each other (giving you the structure of a physical theory), but what are the relata? Where's the "categorical ground" of these behaviors? Physics is necessarily silent on this question.
The epistemic ground, the thing we posit these structures to explain in the first place, are our observations - not things like "the meter read 5 volts" that require a background theory that posits the existence of things like "objects" (not necessarily an explicit one: with the possible exception of newborns, we're all doing tons of interpretational work all the time) but the raw uninterpreted perception underlying them. This exist, clearly, or there would be no need to posit the existence of anything at all. And this is what we're referring to by "qualia".
"Natural nunbers can be studied by counting rocks. But there are negative numbers, there is an intrinsic property of negativity, and you can't study that by counting rocks. That Hard Problem of Negativity you have to leave for us philosophers, which is why we will always have a job, regardless of how good you get at rock-counting."
> "Natural numbers can be studied by counting rocks. But there are negative numbers, there is an intrinsic property of negativity, and you can't study that by counting rocks.
Well, no, "negativity" isn't an intrinsic property, it's relational in the most literal possible sense: to be negative is to be less than zero. All mathematical properties are clearly relational - otherwise something would be lost in translation between formalists and platonists and what have you, and nothing ever is except by human error. But clearly something is lost when we equate the world to the solution to a set of differential equations on a chalkboard - one, to start with, is a physical thing, and the other is not.
Nonetheless this is a caricature of an argument that actually does succeed: you can't study anything beyond the most elementary of elementary number theory by counting rocks. No matter how good our numeric techniques get, they'll only ever be an aid for real mathematics, never a substitute.
There is a sense in which if you believe in the hard problem, you already think physicalism is an incoherent concept. You could call that "begging the question," but it's no more circular than saying I exist because I experience that I exist. I'm not really interested in the argument that I could be mistaken about existing.
In my opinion "physicalism" shouldn't even be a topic. When most people say this word, I think they are thinking of "science vs. miracles and spirits," or vaguely imagining claims that would be similar to interactionism.
This sounds about right to me. Conversely, if you believe physicalism is obviously true, as I do, the assertion that there is a “Hard Problem” kind of just seems like an assertion that you believe the experience of sensations/perceptions is metaphysically separate from the rest of reality (I would be tempted to say, "like some kind of miracle"), and no possible explanation will ever be sufficient to convince you otherwise.
“I think they are thinking of "science vs. miracles and spirits,"” Pretty much. I’m happy to concede that the question why it is like anything to have experiences is puzzling… But it seems clear that that’s just what brains do. I assume it’s a necessary consequence of their design. I can't wrap my head around quantum physics either, but that's not a problem that causes me to think it must be wrong or to doubt the ability of science to make sense of reality.
What I am more certain than I am of anything is that the concept of a disembodied mind is incoherent. Hence no ghosts, Gods, spirits, afterlife, etc. A mind is the activity of a brain; a brain is an information processing device designed by evolution to cause muscles to contract in the right ways at the right times to facilitate survival and reproduction. We have a reasonably clear picture of how, starting from the big bang, the interaction of matter according to the laws of physics gave rise to creatures with brain/minds. Potentially some of these creatures will create artificial minds with technology. There is no other plausible way we know of for minds to come to exist.
Insistence on the hardness of the hard problem does seem to correlate with a skepticism of that understanding of reality, even though technically it may not necessarily be incompatible with it. But it seems to me that unless you’re going to use it as a God of the Gaps argument through which to smuggle, if not God and/or souls, then…what exactly? If “physicalism” could be “not true” in a way that is still consistent with the above picture, then…I don’t really know what you’re talking about or why it would matter.
Some of this stuff does kind of seem like philosophers tying their heads into knots, arguing over how many angels can dance on the head of a quale.
> an assertion that you believe the experience of sensations/perceptions is metaphysically separate from the rest of reality (I would be tempted to say, "like some kind of miracle"), and no possible explanation will ever be sufficient to convince you otherwise.
This is a conclusion, not the premise. The underlying assertion is that experiences are not *definitionally* identical to physical entities. It might be the case that they're different names for the same entities, but if so then this is true in virtue of some contingent fact about those entities, not just something that can be assumed.
> If “physicalism” could be “not true” in a way that is still consistent with the above picture, then…I don’t really know what you’re talking about or why it would matter.
No one is claiming your picture is is false, just that it's incomplete. In the same way that we can ask what "breathes fire into the equations", what distinguishes real physical stuff from differential equations on a whiteboard (which none of our physical theories seem capable of answering, even in principle) we can ask why being is like anything at all, what distinguishes our world from one where things are following all the same trajectories but the lights are out and all the self-reports of experiencing things are wrong.
It seems to me that it must be some sort of emergent property. You can't find "wetness" in H20 molecules, but that doesn't mean either that wetness isn't real, or that physics/chemistry can't account for wetness.
But fair enough. Maybe that's not right. I'm fine with saying it's a weird philosophical conundrum that we don't have a satisfying answer for, and it's conceivable (but by no means certain) that we never will.
But if you go on to postulate that therefore there must be "something else"...then I would want to know: what is that something else? Where did it come from? What is the evidence that it exists? Why does it exist? How does it better account for the observed phenomena?
I certainly don't think we have a complete understanding of reality...but there's what we can know with a fair degree of scientific certainty (which I think is quite a lot actually) and then there's speculation.
So you're an Emergent Materialist - which I think is probably a much better paradigm for examining this particular issue than Eliminativism or Reductionism. If you have two bits, and they are both 1s instead of 0s, the "threeness" permeates the arrangement even though the individual bits couldn't be said to have "threeness". This particular thing should be fairly obvious. Answering "How does subjective experience arise from apparently objective matter" may have a similar answer but it would substantially less obvious.
I think I'm personally going cross-eyed and getting to the point where Panpsychism and Physicalism seem like the same thing, just with weird quasi-religious attitudes about their use.
Physicalism can mean a 'range of things beyond "no ghosts or gods", and none of them is logically necessary. There is no apriori reason why everything should be objective as opposed to subjective, quantitative as opposed to qualitative, relational as opposed intrinsic, etc.
Would you refute then that the mind is dependent on what stimulates it? Because it could be argued that the “activity of the brain” is as dependent on things outside the brain as the brain itself.
I’m not contradicting you, I’m very ignorant about this, but I see the mind as the brain’s potential interaction with everything it can be made aware of, and in that sense it is a far bigger and complex space than what’s contained in your skull. Experience can reasonably be seen as crucial a component of the mind as the physical brain itself.
To be honest, I think physicalism is backed up by vibes rather than anything else. It feels kind of scienc-y and futuristic, so people who like to think of themselves as scientific and progressive are drawn to it for that reason.
I think that most such people are scientifically minded and they think philosophy is unproductive, so they gloss over this. They hear: "do you have a little soul inside your brain pulling the levers, or can material interactions explain the divine the wonders of human mind?" And then they write up a reply that says "no, there is no little soul," and move on to the next intellectually stimulating topic.
I'm not even sure what to say to a professional thinker who wants to deny that consciousness presents a problem. Fine, deny it. Rewrite the dictionary so that it says my conscious experience is defined as primary qualities of matter in the Lockean sense. Have yourself a ball.
If you want to argue with a dilettante about this (which I do not), I find that using morals and ethics vis-a-vis the realism of other minds is a better entry point than trying to pin down a definition of "qualia." Why are you mad that I'm about to torture this cat? Because it will set off a chain of physical patterns that are roughly isomorphic to ones that your brain has evolved to be upset about? Surely you should be able to overcome such.... atavisms.
Whether I stop caring about the issue before or after I make you let the cat go seems like a question of economics, tactics, and strategy, or maybe veterinary professional ethics, rather than philosophy or neurobiology.
You only have to regard physicalism as being not necessarily true, ie. being potentially falsifiable. It's odd how many of the pro-science crowd treat physicalism as an unfalsifiable dogma , and therefore an unscientific claim.
Interesting Point. What would falsification of physicalism look like? I'm so committed to empiricism that all falsification I can imagine would have to be empirical, i.e. physical, which could be construed as just another element of (some refined new version of) physicalism.
Do you have a better suggestion, where a replacement of physicalism would not just lead into another physicalism? Something like contradiction between different consequences that physicalism entails, some type of reductio ad absurdum?
The hard/easy distinction is entirely valid. "Hard" is used to to indicate a problem which the current paradigm cannot adequately cover without substantially shifting.
A problem will stop being "hard" when we have a framework in which we can posit one or multiple candidate models that can plausibly account for all observations, and all that's left is to hash out their details and figure out the experiments to determine which of them correspond to reality.
A problem will continue to be "hard" for as long as we operate under a framework where we can't even conceive of a superficially plausible answer that still fits into the rest of the framework.
It seems to me the field of physics is in a similar spot facing hard problems without being subject to the same accusations of mysticism as some are making here.
I don't really know what I'm talking about with respect to physics (I'm a psychologist) but have been reading Matt Strassler's "Waves in an Impossible Sea" and I'm struck by how much we don't know while continuing doggedly to try to know more with the threads and shards that we do have.
Gravity and electromagnetism and Higgs "particles" and mass itself are emergent properties of media (media like how water is a medium for waves) that we do not begin to understand or see. Strassler says it also looks impossible HOW we could ever understand or see what these things are emergent properties OF in physics terms given how relativity works.
Physics has been staring at a collection of hard problems that way for about 150 years without blinking as far as I can tell, without pushing the hard questions aside, without giving up on answering them, and without devolving into mysticism.
For all we know, consciousness is an emergent property of some other kind of wave-based field we can't see but that in animals involves neurons. Maybe how humans can only see a tiny part of the electromagnetic spectrum, we only engage with this field in a limited way, less complex-brained animals even less so. I'm not saying the hard problem of consciousness is literally a physics problem; I'm just saying we know how to look at impossibly hard problems without blinking.
In the frame that the hard problem of consciousness is a racket, then is string theory etc also a racket? Or is the fact that math can be constructed around some of these theories at least partially take it out of racket-land?
Racket seems like an epistemic closure kind of way to approach an unexplained phenomenon that universally exists at least among humans. The epistemic closure aspect of this response makes it seem unscientific.
I don't think string theory is a racket in the same way that the "hard problem" is. But you're right, if "racket" isn't going to be just a sui generis category, I should be able to give comparable examples of similar rackets. Excellent point.
I would nominate (as another example of a similar racket) the insistence that we need to "teach the controversy" about intelligent design. Because that also posits that there has to be a problem, and any disagreement, any claim that this isn't really an unsolved problem, is treated as further evidence that there is this controversy that needs to be respected.
One can observe that less progress has been made on the HP stuff in practice. The easy problems are related to functioning, behaviour and information processing. We have already made progress in understanding these, including being able to reproduce them to an extent in AI. We expect to continue making progress using much the same techniques.
By contrast the hard problem is one we have not even started on. It's not
that we can build AIs with simple qualia, but we don't know how to build or code qualia into a device at all -- its that we cannot code seeRed(). If we had a reductive explanation of qualia, we would be able to use it as a guide to building qualia.
One can also note that the HP is unique in science, because it is an attempt to explain that which is entirely subjective
In addition qualia cannot be detected -- there are no qualiometers. If qualia are phenomena that are closed to objective empirical investigation, they do not fit within the conventional scientific paradigm.
And lastly, it's hard to see how a theory of qualia could be predictive. It's difficult to see how a linguistic (including mathematical) expression could convey a quale.
I observe only the HP people denying that any progress has been made.
The above is a reductive theory of qualia but it doesn't help you build AIs with qualia. (Except in the trivial case of a particle-level emulation of an entire human brain.)
Subjects are not outside of science, so their experiences aren't either. Unless you're going to define "subjective" in terms of qualia, a ridiculous circular concept that is nevertheless out there.
Predictiveness doesn't require conveying qualia. For predictiveness, it would be enough if the press of a button on e.g. transcranial magnetic stimulation machine merely disrupts qualia in a predictable way.
The existence of subjective experience seems difficult to even begin to explain with our current understanding of the physical world - if a physical explanation exists, it’s unclear what it would look like. As an analogy, the ancient Greeks lacked the understanding to attribute lightning to anything other than the Gods, and you could only explain how lightning worked by substantially shifting their worldview.
While we might be able to describe the physical correlates of conscious experience (just as the Greeks could tell you that lightning typically comes when it’s cloudy) and predict when it would come about [easy problem], the explanation of how the thing itself exists might still elude us [hard problem].
I think the "hard problem" is something like "why would subjective experience, which doesn't in itself cause anything else, necessarily accompany (some) thoughts that do cause things"? So I think a solution would look like an explanation why subjective experience cannot not arise from the workings of (some) thoughts.
That's what I'm trying to do here, by explaining why something with the enumerated properties of qualia does have to arise out of the way human thoughts handle information that is internal to them.
This is fascinating! If true, it gives a way to measure whether things are or are not conscious--awesome!
It does not, alas, explain why the synchronized, oscillating firings of neurons feels like anything, meaning that it offers ~zero evidence for physicalism as opposed to something like dualism.
I'm still not quite convinced that the theory (even assuming it is correct about human consciousness) provides a universal test for consciousness. Without a deeper explanation of "why [it] feels like anything" then I'm not sure, e.g., whether a computer program in a different substrate -- with synchronized, oscillating firings of neurons -- is conscious. Or another human, for that matter. In either case, the being in question saying "I'm totally conscious right now!" isn't quite convincing (though I admit that in the human case, I'll generally just operate from the assumption that they are).
I think it lets you establish with reasonable certainty which things aren't conscious, which is better than we have now. Currently, the model is something like:
Brain does physical thing X -> ????? -> consciousness
Computer does physical thing Y -> ????? ? -> Consciousness?
"?????" is the answer to the Hard Problem. We know Y because we can see everything the computer does. But we don't really know what X is.
Suppose X is the brainwave thing this post makes the case for it being. Then we can compare X and Y. If they're structurally very similar, then we've learned something, and we can go on worrying about substrate dependence. But if they're dissimilar to the point where the brain's version of Y wouldn't cause consciousness, then we can be pretty sure the computer doesn't have humanlike consciousness.
Heck, you could even run the same comparison with insects, and finally settle (or throw gasoline on) that EA debate.
There's always a chance we're missing other types/generators of experience until we solve the Hard Problem. But this seems like an improvement.
I don't claim this explains whether computer programs on every possible different substrate would be conscious. But I do think a full atom-level simulation of a human brain would exhibit computations functionally indistinguishable from what we experience.
Why does it look like anything from the inside? Why aren't there just brainwaves mechanically oscillating and computing without anyone to notice? The hard problem asks why there is an inside at all in the first place.
If you say "anything that affects neurons firing in the brain is of course going to affect experience. We have found a process happening in neurons that explains why we feel what and how we feel!" then you have indeed made a great discovery! But the Hard Problem is swept away unexplained by the words "of course".
The reason there is an inside is precisely because thoughts handle internal information processing differently from the handling of information received out of sync. This is not unexplained anymore.
The software I work on at my job handles internal information processing differently from information that it receives out of sync with its cycle. Perhaps I'm mistsken, but I don't think you would say this is sufficient to conclude that my software has awareness or experience.
I think your awareness processes information. And when it does, it does it in a workspace - not a Global Workspace as in GWT, but a local one inside an oscillating thought.
That is not the only way of information processing (unconscious thoughts can do process information) so there's no need to assume anything else is conscious... at least until someone builds a variant of LORETA that can find oscillations in cows. Some of the animal welfare EAs should be interested in that.
I think you captured pretty much all of what it feels like, and that's why I'm very excited about this work! The piece of the explanation I claim is missing is *that* it feels like anything.
Here's a question for you: Do you think everything in the world is conscious? Ie, that it feels like something to be a rock, a washing machine, a tree, a submarine, a storm, or an office building? None of these things? Only some of them?
Your theory accounts for the properties of human thought. I think your theory could observe that none of these things have similar properties to the brain, and so they have different properties. Eg, suppose the sailors on the submarine move in a way that processes the information of the orders they were given. Because submarines are cramped, they have to move through certain hatches one man at a time. From this, we can conclude that submarine qualia (if they have any) cannot be simultaneous. Submarines experience only one thing at a time.
Others might say: "I deny that submarines have experience. It's ridiculous to claim that people moving inside a submarine cause qualia, even a very exotic sort of qualia. Sure, molecules and EM waves moving inside the brain must somehow cause qualia, but brains and submarines seem very different and I suspect there is something about brains that submarines lack, such that it feels like something when a brain's particles move, but it doesn't when a submarine's sailors move."
I think your answer to this person must take one of 3 forms:
1. Submarines are indeed conscious because everything is conscious. Our theory explains which physical processes are associated with your experience; they may or may not have any applicability to a submarine.
2. Humans are conscious and submarines are not. We know this because we understand exactly how conscious experience results from the movements of molecules and fields. Motion normally happens without any experience, but our theory says that when some system has property X, consciousness arises by mechanism Y.
3. Neither submarines nor humans are conscious; consciousness is an illusion.
Some people legitimately claim 3, but it is wrong; I am conscious. 1, some form of panpsychism, is a very plausible answer that side steps the hard problem. I would accept 1, but you have not said anything that makes me believe this would be your answer. 2 is a complete answer only if you have both X (the property that causes consciousness) and Y (the mechanism by which motion generates experience). You have offered X, but not Y.
It may be tempting for a reductionist to say that Y is somehow just a macroscopic property od systems satisfying X; IE, that consciousness is emergent, and reduces to the motions of molecules. I think this is not a complete explanation. Reductionism has been very successful at explaining momenta and forces of macroscopic objects in terms of the motions and forces of smaller objects. It has shown convincingly that tenperature, a macro property that seems totally unlike motion, nevertheless reduces to the motions of tiny particles.
Consciousness--feelings--seem totally unlike motions and forces. But maybe it's like temperature and actually reduces to them anyway. If it does, then an actual reductive explanation should be able to answer questions like: What are the dimensions and units of feeling? Are they derived units, or fundamental? If I know X property of a system, is that enough to know anything about feelings? If it is, what calculation can I run on the particles of this system to learn about the feeling? What is the theoretical basis for believing that to be the correct calculation? Is there a toy model of a minimal conscious system that we can understand?
Unless you can answer questions like these, you haven't actually reduced consciousness to motions and forces. And so it remains an open question whether it in fact reduces or not! Maybe instead of being emergent from a system, there is a consciousness-particle that interacts through the electric field and a fifth fundamental force, the force of feeling. Or maybe there's a universal experience-field that brains just interact with especially well. Etc.
Showing that properties in the brain align with properties of qualia is genuinely interesting and great work. It shows us which processes in the brain are associated with consciousness. And in particular, I think it convincingly shows how self-reflection works! Anything lacking a brain thay can sync in this way is not going to be self-aware in the sense of being able to reflect on its own thoughts--that's awesome! It's an incredible breakthrough and I'm very excited about it!
But we have experiences that don't require self-reflection, like, say, the sensation of pain. The Hard Problem of Consciousness is about explaining these primitive, non-self-reflective sensations. Having a solution to the Hard Problem means being able to answer questions like "do ants feel pain? Do rocks or trees feel anything at all?" I would love to be proved wring, but I don't think your explanation can answer this type of question.
My answer is 2, and I contend I have given 2 but it is about the movement of information, which is implemented in the movement of electricity and ions, but also a few of the molecules you seem so keen on.
Qualia reduce to information.
Pain doesn't require self-reflection, but it does require neuronal information processing: that's why general anaesthesia works.
Sorry, but that can't possibly work. If you take qualia seriously as not reducible to mere emergent physical properties than Chalmer's zombie argument shows why it's in principle impossible to experimentally test a theory of qualia.
Yes yes there are all sorts of caveats to that statement. You can certainly test a unified theory of physics and experience and maybe that gives you some extra information but they can't be tested the way you suggest.
> If you take qualia seriously as not reducible to mere emergent physical properties than Chalmer's zombie argument shows why it's in principle impossible to experimentally test a theory of qualia.
This is true, but also not as unusual as it sounds. All "experimentally testable" claims, with the somewhat ironic exception of claims about raw sense data, require supporting theoretical infrastructure - you don't test "the mass of an electron is X", you test "my experimental apparatus will produce a readout of X when I do Y, which if [background theory] is correct means the electron mass is X".
Any such background theory for qualia is going to have to assume that self-report is reliable under at least some circumstances, at which point we're back to a problem which is merely very very very very hard, not impossible.
The nature of the hard problem is to identify what additional laws of nature are necessary to explain why we actually have qualia not merely say we do.
Few people doubt that we can explain at a microphysical level what causes us to say that we are conciouss the problem is to explain why we actually are. The whole point is to explain what laws connect up physical with the mental and if you assume the physical world is causally closed - no soul comes in to alter physics in the brain - that's empirically untestable.
And yes, one might assume that we are in fact having the experiences we say we do but because we are no longer asking for what physical state is necessary to cause (or at least predict) any physically observable outcome you can't falsify such a theory with empirical predictions.
What one can do is try and offer a theoretically attractive fundamental theory which unifies physical predictions and experiencal ones (taking our self-reports as correct) and show it agrees with physical predictions but it's still not falsifiable by physical observation.
---
And yes the apparent fact that qualia don't -- in some sense -- make a difference to physical outcomes raises questions about why self-report is correct. Hence why most people attempting a partial solution have wanted to suggest that certain computational states necessarily give rise to certain experiences -- so systems which report they have qualia X really tend to have it but it's a bit of a paradox.
I'm familiar with the problem, my point is that the "easiness" of physical problems is not a given. If the philosophical background radiation of our culture endorsed subjective idealism instead (and there can't be any direct empirical evidence for or against it vs. physicalism, for the reasons you note), for instance, then the physicalists would be the ones bashing their heads against the wall trying to explain that when they talk about "objects" they're not just referring to structural relations between quales. (Arguably you actually see a little of this in debates over scientific realism.) The counterpart of the hard problem in such a world would be something like "why do experiences have causal structure?", and the equivalent of naive illusionism would be something like " 'huh, it seems like my experiences have causal structure' is just what it feels like to be in a particular sort of mental state"
I am ultimately am sympathetic to the idea that it's a mistake to take physical claims not the qualia as fundamental. I mean at some level I'd argue they are epistemically prior. But that doesn't really help in this case.
You are correct that if someone offered a single theory that made predictions about qualia and unsolved questions of fundamental physics thst would absolutely be important evidence for the unified theory. I tried to indicate that but may have been unclear.
But however you view it (as physical regularities or regularities in qualia representing experimental outcomes) we have theories (the standard model) that are damn fucking good at predicting stuff and that we are damn sure will predict what happens at the brain at the level considered here. So the proposal here can only be wrong or isn't given further support by those tests.
Logically speaking predicting what happens in the brain isn't a hole in our fundamental theories just an issue of computational power/descriptive detail.
Unless you are claiming it's merely a question of consistency and it's not truly an empirical test but a question of mathematical compatibility (is this theory compatible with what we know about physics)? But the parts of the article before this make me doubt that this is the claim.
> Unless you are claiming it's merely a question of consistency and it's not truly an empirical test but a question of mathematical compatibility (is this theory compatible with what we know about physics)?
Pretty much, yes. All a physical theory not equipped with a choice of interpretation gives us is correlations between bits of sense-data, and formalism alone doesn't even give us that - you still need to decide which parts of the formal structure are "measurements".
Not sure what article you're referring to, I'm not the author of the piece we're commenting on.
I didn't mean to suggest you were. I was saying that my criticism of the original claim of the author's to be offering an empirically checkable theory was justified. I was assuming my remarks were contextually restricted to be about that and may have been a bit sloppy as a result.
I completely agree with the point you just said here. And it's one of the big mistakes that advocates of no collapse QM interpretations make. They assume that it's a simpler theory because they can dispense with any need for an extra law to handle collapse. But -- while no collapse may or may not be true -- once considered in the full light of the need to specify what parts of the theory correspond to experiences you pay that price no matter what.
However, I just don't think it's relevant to the article above since the kind of empirical confirmation they mention doesn't correspond to any plausible uncertainty in how to identify measurement with experience. If they'd offered a claim about the role of collapse in QM which stemmed from a theory about how to identify parts of the formalism with experience sure. But that just doesn't come up at the level of description their experiments and theories occur at and I felt my response was already too long w/o me including those caveats ;-).
> All "experimentally testable" claims, with the somewhat ironic exception of claims about raw sense data, require supporting theoretical infrastructure - you don't test "the mass of an electron is X", you test "my experimental apparatus will produce a readout of X when I do Y, which if [background theory] is correct means the electron mass is X".
That's the Duhem-Quine thesis (except for the bit singling out sense data as different).
By naive verificationism I mean the sort of contemporary internet "positivism" found in rationalist circles, which is neither remotely standard nor particularly similar to verificationism in the sense of the early logical positivists.
> If you take qualia seriously as not reducible to mere emergent physical properties
If you assume that qualia are by definition non-physical, then, unsurprisingly, you going to end up thinking that there can't possibly be a physical explanation of qualia.
This, of course, is a great reason not to make such question-begging assumptions in the first place.
Yes but the usual reason to talk about them as a theoretical entity is because you think you need that additional level of description -- it's at best extremely misleading to use the term differently without clarity. Or, to put the point differently, if qualia logically supervene on the physical then there isn't any extra fact to specify over and above the physical theory so why complicated things by introducing unnecessary terminology? As such and based on the rest of the piece I think that's an appropriate way to understand what they took themselves to be describing.
And half the assertions in the piece about the nature of qualia only make sense if they understood it as not being logically supervenient on thr physical. They are only private if you mean by qualia something different than "a physical system satisfying such and such configuration." They clearly had in mind the hard question of consciousness not some easy sense and so it's fair to judge their theory on those terms
If they'd instead wanted to argue for a physical reduction Dennet style they could have done that but that makes even less sense of a claim of empirical falsifiability.
> if qualia logically supervene on the physical then there isn't any extra fact to specify over and above the physical theory so why complicated things by introducing unnecessary terminology?
Trees logically supervene on the quarks and yet we use the term "tree" regardless. Clearly there is a reason to use separate words for things which are "merely physical".
> And half the assertions in the piece about the nature of qualia only make sense if they understood it as not being logically supervenient on thr physical. They are only private if you mean by qualia something different than "a physical system satisfying such and such configuration."
So you might have though. But apparently it's not the case. Here the author of the post shows how qualia can be private while still being physical. You are free to engage with his arguments of course, but you can't dismiss them just because you smuggled the notion of qualia's non-physical nature in the definition.
> They clearly had in mind the hard question of consciousness not some easy sense and so it's fair to judge their theory on those terms
If you beileve that explanation of privacy of qualia is part of the hard problem of consciousness, then the author has just solved some part of the hard problem of consciousness, even though previously it appeared absolutely impossible from your perspective. Feel free to make the obvious conclusions what it means for the rest of the hard problem. Or at least notice how you mind shifts the goalpost further.
This isn't the kind of thing you get to just prefer one answer to another because no one is trying to say they aren't similar — just that you can't derive facts about experience purely as a may of logic alone from physical facts with no extra assumptions. And for that task it's enough to show that there is any way in any conceivable situation they don't perfectly track each other*.
Besides, if you actually want to avoid having to deal with the hard problem the burden is really on you to give a positive argument showing that the existence of an experience follows purely by logic from a physical description. The zombie argument tries to show that no such argument is even possible but if you want to avoid the hard problem the burden is still really on you to make the positive argument.
—
*: It's analagous to arguing that — even though in our universe gravitational and inertial mass are always the same — that the two notions aren't merely different logically equivalent descriptions of the same thing. For that it's enough to say: but you could imagine a world in which how hard it is to accelerate something could vary independently of how much it attracted/was attracted by other objects.
Doesn't make Einstein's analysis that sees acceleration and gravitation as two sides of the same coin any less correct or influential but it tells us that this analysis is a substantive law of nature and evidence for it should be evaluated the same way we evaluate other laws of nature.
To meet that burden it should be possible to write down a physical description and as a matter of pure logic (only math no assumptions) tell me what experience it represents. I see no argument that is possible.
> If you assume that qualia are by definition non-physical, then, unsurprisingly, you going to end up thinking that there can't possibly be a physical explanation of qualia f
If. Not al! arguments work.that way., EG. chalmers.
Feel free to present this particular argument, so we could engage with it, instead of vaguely pointing towards the author, whose most famous argument - p-zombies - begs the question in exactly this way.
P-zombies argument appeals to non-physicalist intuition as its premise, which is justified only if physicalism is false which is the whole question being investigated:
Physicalism is false -> Zombie World where everything is exactly the same but humans are not conscious is logically possible ->Physicalism is false.
You can go read the book of you want but Yudkowsky is alot less smart than he thinks he is -- there is a reason that even the philosophers of consciousness who disagree with it take it quite seriously. Here Yudkowsky is the idiot going "see look I disproved quantum physics."
If you want the first point Yudkowsky goes off the rails it's with his claims about causation. Chalmers assents to the claim that there is a natural law relationship *in our universe* between your qualia and your physical states. I don't remember if he calls it a causal relationship but I sure will and that's perfectly coherent with the zombie argument.
The fact that you can imagine a universe with different natural laws which breaks the law like relationship doesn't show there was no causal relationship in your world so doesn't establish that yo didn't know it.
No, I'm rejecting the analogy between quantum physics and Chalmers' theory. That doesn't make me a "troll". If you're arguing against Yudkowsky's critique of the zombie argument, you can't just assume your conclusion about that argument and then use that to dismiss anyone else who doesn't share your take.
That's fair..I deleted the comment because I realized you likely understood the analogy differently than I meant it which wasn't about the degree of confidence in the theory.
I'm sorry, I shouldn't have presumed you were understanding it the same way I was.
No, it might very well be wrong. And I don't begrudge Yudkowsky for arguing against it (even if he made some mistakes there). I do think he failed to constructively respond to the many good counterpoints raised but that's just what being wrong sometimes looks like.
My intent wasn't to suggest what he was doing was unreasonable (hell I don't think the guy who thinks they've disproved QM is necessarily unreasonable just not an authority for others) only that it is unreasonable for someone else to presume Yudkowsky's argument is correct absent having throughly studied the material themselves.
Relative to you he has the position of the lone doubter with the bad track record challenging the established view. What doesn't make sense is citing Yudkowsky as if he was an authority.
"If you want the first point Yudkowsky goes off the rails it's with his claims about causation. Chalmers assents to the claim that there is a natural law relationship *in our universe* between your qualia and your physical states. I don't remember if he calls it a causal relationship but I sure will and that's perfectly coherent with the zombie argument."
I don't really follow what claims about causation you mean or why they're a problem.
"The fact that you can imagine a universe with different natural laws which breaks the law like relationship doesn't show there was no causal relationship in your world so doesn't establish that yo didn't know it."
Having a rough time parsing this as well - didn't know what?
Maybe it's fair if I extract the main point of Yudkowsky's argument here instead of just linking to a long blogpost. AFAIU the main point is simply that the zombie argument can be summarized as follows, and that this shows the argument to be incoherent:
"
- Matter has additional consciousness-properties which are not yet understood. These properties are epiphenomenal with respect to ordinarily observable physics—they make no difference to the motion of particles.
- Separately, there exists a not-yet-understood reason within normal physics why philosophers talk about consciousness and invent theories of dual properties.
- Miraculously, when philosophers talk about consciousness, the bridging laws of our world are exactly right to make this talk about consciousness correct, even though it arises from a malfunction (drawing of logically unwarranted conclusions) in the causally closed cognitive system that types philosophy papers.
The relevance of the causal talk is that Yudkowsky argues that we can't know that qualia exist because they don't have any causal impact. It's roughly the same as your claim that the qualia are epiphenomenal.
And my point is that there is a subtlety that's being missed here. What Chalmers claims is that it's logically possible that **if the laws of nature were different** all the physical stuff behaves exactly the same and this isn't the same as the claim that qualia have no causal power and even if you stipulate that it doesn't call into question the knowledge claims about qualia like you suggest.
--
For instance, one attractive principle is that we can claim to know X is true if there is a reliable connection between what we observed and the truth of X. For instance, suppose that physical law guarantees that whenever I see the color red a blurg is nearby. Prima facia that's enough to justify my claim that I know a blurg is there when I see red.
And qualia are connected via exceptionless natural law to the physical states of our brain so seemingly the same justification works there.
Now maybe you want to say -- but how can we know there is the right kind of natural law? But this demands too much because there are infinitely many theories consistent with our observations and at some point we just have to appeal to our priors (attempts to declare some universal prior have been disastrous failures and we have no clue how we would rule out some priors but not others much less this one.
--
More generally. I think it's wrong to even conclude qualia are causally impotent from the fact that every physical effect has a physical cause.
To illustrate what I mean, consider the question of whether "gravitational mass is causally impotent". Now it's certainly true that because our laws of physics say inertial and gravitational mass are always equal I can rewrite the laws of physics in an equally predictive form that only mentions inertial mass. Of course, the new way I've written the laws of physics would fail to hold in a world where those two values aren't equal but it doesn't show what we might intuitively think of as the claim "inertial mass is causally impotent". After all, I could have played the game the other way.
Same deal with qualia. If there is a natural law in our universe that says such and such qualia is realized iff such and such physical state is realized I can redescribe the laws of physics in an equally empirically valid way for our universe replacing the role of those physical states by the qualia. Or, to put the point another way, if physical law says X and Y always co-occur then just because you can write the laws without mentioning X doesn't entail X has no causal power.
>Now maybe you want to say -- but how can we know there is the right kind of natural law?
I dont think Yud is commited to any account of "knowledge", just belief, truth, and subjective confidence. Am I right to translate the following paragraph as "If someone says they have qualia I belief they have just those qualia purely on priors."?
>To illustrate what I mean, consider the question of whether "gravitational mass is causally impotent".
This seems like a Motte-and-Bailey. Noone cares whether you draw a distinction between the two or call it all just mass. Just like noone cares about the distinction between "yin-charge" (influencing the strength of electrostatic force a particle experiences) and "yang-charge" (influencing the strength of force a particle exerts on others) that I just made up. The only reason it would matter for qualia is that we *feel like* the qualia would have some kind of additional effect that the physical states alone dont - but that would make the argument inapplicable.
Detailed response below but basically I can take the same form of argument to show we can't possibly know about an object's gravitational mass.
1) I can rewrite all the laws of physics in an empirically equivalent form by just using inertial mass everywhere we use gravitational mass because it's a natural law they are always equal.
2) In principle it's coherent to imagine a world where the gravitational constant was 0 but there was another kind of force that works like gravity called smavity. Now objects can have all sorts of smavity charges but it's possible to have a universe where the gravitational constant is 0 but objects all just happen to have a smavity charge equal to their inertial mass.
So it's possible to have a universe where gravitational mass is different than inertial mass but is empirically identical (because in that universe smavity happens to play exactly the right role.
3) Therefore gravitational mass could be different without there being any way for us to tell so we must not have any justified belief or knowledge in gravitational mass.
But of course that argument is wrong. We can know about gravitational mass because in our world it absolutely does have causal power and it's irrelevant what we can imagine about a world with different laws.
Same problem here. In our universe the natural laws do connect up qualia with physical effects. Since you can't have brain state B without quala Q as far as natural law is concerned everything caused by B is equally caused by Q. Doesn't matter what I can imagine about another universe with different laws that connection lets me have knowledge of qualia.
--
Call it what you want. He used the word "know" so I followed but same argument.
And I wouldn't say just based on priors, I'd also say because you have the experience directly because I'm not willing to identify 'you' with a brain but we can bracket this, but also because it's kinda misleading.
I mean ultimately, how do we know there are physical objects or an external world at all? We could be Boltzmann brains or idealism could be true. Ultimately, we just apply our priors and if those priors actually get the correct answer in a reliable fashion (ie, if there is a natural law which in fact guarantees every time you see A that B is true and your priors say that you are justified in believing B when you say A and you do see A and believe B we call it justified). Nothing different here.
---
All the rest is just a means to point out you can't treat qualia somehow worse because they aren't physical by appealing to the causal closure of the physical. Yudkowsky can't argue that because we think the physical world is causally closed that therefore qualia don't cause your mental states. Natural law can still say, you have mental state X iff you had qualia Y a second ago (and some list of other facts).
Besides it's silly anyway because we claim to know things like "the universe isn't filled with cheese past the cosmic horizon" even though it's causally isolated from us (in a way qualia are not).
--
Importantly, it's not necessary this argument convince you that qualia exist. Yudkowsky is arguing that it is incoherent to claim to know you have qualia while believing they don't supervene on the physical and he hasn't shown anything like that because it's both false that we can't have justified belief in things that don't causally affect us *and* it's false to say that the zombie argument entails qualia are causally isolated.
"The relevance of the causal talk is that Yudkowsky argues that we can't know that qualia exist because they don't have any causal impact. It's roughly the same as your claim that the qualia are epiphenomenal."
If you see Yudkowsky arguing qualia don't have causal impact it is surely in the context of Chalmers' arguments and with the goal of showing Chalmers is wrong. I'm not sure if that's what you meant.
"And my point is that there is a subtlety that's being missed here. What Chalmers claims is that it's logically possible that **if the laws of nature were different** all the physical stuff behaves exactly the same and this isn't the same as the claim that qualia have no causal power and even if you stipulate that it doesn't call into question the knowledge claims about qualia like you suggest."
Imagine the laws of nature changing (in the zombie universe) to somehow remove qualia, yet all physical stuff staying the same, including people like us arguing about qualia online. In the zombie universe, we are arguing about qualia for no reason at all, yet somehow it follows from the qualia-less physics that we do so anyway. *If* indeed we imagine that our zombie twins talk about qualia without having qualia for a reason within standard physics, then that reason would be sufficient to explain us talking about qualia in this universe as well, for our universe's standard physics. Yet at the same time proponents of this argument say that there *are* nevertheless qualia in this universe, *and* that our talking about them is somehow related to that fact. Both of these are plain absurdities.
"More generally. I think it's wrong to even conclude qualia are causally impotent from the fact that every physical effect has a physical cause."
To reiterate, we (Yudkowsky and I, if I may presume to speak for him for a bit) don't claim qualia are causally impotent. We claim that *on Chalmers' view* they are causally impotent, and therefore we reject that view.
"Or, to put the point another way, if physical law says X and Y always co-occur then just because you can write the laws without mentioning X doesn't entail X has no causal power."
I don't think this is quite right, when we're talking about physical laws. If you rewrite the laws so that you don't mention X, yet what you write is still a complete and accurate description of the laws of physics, then you are actually thereby still mentioning X (and the causal power implied by it), just under a different name, in a different mathematical formalism, or whatever.
Look, I agree that there is a deep weirdness that Chalmer's view entails in that it implies that there are all sorts of ways that qualia could have been related to the physical world that didn't necessarily let us report about them at all. First, maybe there is just a really fucking compelling theory of everything (physical and qualia) that implies this holds. We can't reason from our lack of knowledge of such a theory.
Secondly, the temptation to view that we obviously problematic rests on the attitude that somehow the physical comes first and then the mental is layered on top of it. I'd argue that gets it exactly backwards. Epistemically, what we start with is the qualia and to the extent we have evidence for the existence of the physical world at all it is as a result of regularities observed in qualia.
From this POV the question becomes not "why do qualia depend on the physical world in just the right way for us to report correctly about them." Indeed, that becomes kinda non-sensical because on this view "we" are patterns of qualia and the physical world is just an explanation of said patterns. At best the question dissolves into "why does the world have coherent patterns of qualia that create the impression of memory etc etc " but now we can answer it via sn anthropic argument -- if that wasn't true we wouldn't be asking.
--
Also at a deeper level I personally also reject Occam's razor for natural law -- well not fully reject but at least think we shouldn't expect laws to be particularly simple for qualia. I think it only appears to be true for physics because we make an unprincipled distinction between initial conditions and dynamics when they aren't in principle different. It's really the kolmogorov complexity of the whole thing that matters but for anthropic reasons we will find a description that makes the dynamics simple wrt physical laws but not experience laws so we shouldn't be surprised if those laws look way less simple and elegant.
> Matter has additional consciousness-properties which are not yet understood. These properties are epiphenomenal with respect to ordinarily observable physics—they make no difference to the motion of particles.
This is not quite right. Chalmers' position is that phenomenal properties are *explanatory unnecessary* for physics:
> On the question of explanatory irrelevance of consciousness to physical actions. It is true that I bite this bullet, in a certain sense. But the view can be made to seem stranger than it is. So I should make some clarifications.
> First, I do not say that consciousness is causally irrelevant to action. That is a question I am neutral on, and I think there are interesting views that give consciousness causal relevance without doing any damage to the scientific world view. (For example, by making consciousness correspond to the intrinsic aspect of physical states which physics characterizes only extrinsically.)
> Second, I do not say that consciousness can never be used to explain action, so that explanations that involve consciousness are invalid. I simply say that invoking consciousness is not necessary to explain actions; there will always be a physical explanation that does not invoke or imply consciousness. A better phrase would have been "explanatorily superfluous", rather than "explanatorily irrelevant". Something can be superfluous and still be relevant.
> Third, it isn't true that my view implies that "if you think you are reading because you consciously want to read, you are mistaken". On my view, it is very likely that you are reading because you want to read. It is just that fact that the wanting is consciously experienced is not required for the explanation to go through. Conscious wants can explain actions, and nonconscious wants can explain actions too. Similarly, you drink because you are thirsty, but the consciously experienced aspect of that thirst needn't be appealed to to explain your action.
The next to last paragraph *almost* sounds like it could apply just as well to other high-level concepts that have a lower-level explanation in physics. Fire, for example, is explanatorily superfluous in just such a way, or rivers, or hailstorms.
*Almost*, because of course he says: "there will always be a physical explanation that does not invoke or imply consciousness"... Unless he has a solid argument for this claim, it really just turns the whole thing into a case of circular reasoning. Claiming there is always, in this universe, a physical explanation of consciousness-related utterances that does not in any way imply consciousness, is quite a bit stronger of a claim than the usual "we can conceive of an alternate universe where this is so".
Phrasing part of the theory as "humans aren't conscious" seems to be a needlessly edgy and provocative way to state something that most people regard as common sense, that the human mind is made of parts and that some of those parts are conscious while others are not.
Similarly, in 15c I'm not sure that saying that conscious thoughts "mistake each other for themselves" is the best way to describe what is going on. You could just as accurately say that they recognize that their sense of "me" is "pointing" at the same thing, that they are both part of the same person.
Overall, one thing that reading this essay did hammer home for me is that people really do strongly identify their "self" with their sense of consciousness, even though they should know better. This is a matter I've given thought before, I know that consciousness is an important part of me, but not the only part. I know that, for example, I am still me when I am in a deep sleep and all my thoughts are unconscious. But in spite of this, reading some parts of this essay gave me the irrational feeling that I was being attacked, that the essay was somehow saying that I wasn't real, and that accepting it's premises would mean that I'd have to accept that people aren't real and therefore have no moral value or something crazy like that, instead of simply believing that everything adds up to normality.
Thank you, this is useful feedback. I did indeed fail to accommodate the emotional consequences of the common mistake of identifying with consciousness, even while I acknowledged they exist. I'll try to do better in future versions of this.
I have a bunch of quibbles with the claims you referenced at the beginning. In particular, 8-13 all feel very dubious but the bigger issue is with how to understand qualia and whether you are dealing with the hard problem of consciousness.
In particular, the distinction is this. If all you want to know is what gives rise in the brain to us asserting things about having this experience of that there is no particular mystery. That's the easy problem of consciousness just something to be worked out by the usual mechanisms of biology and -- while of scientific interest -- doesn't seem particularly deep or philosophically interesting.
The hard problem of consciousness is based on the challenge to explain why you not only **say** and assert that you are experiencing something but that there is really something which it feels like to be you. (And this is usually the notion being invoked by reference to qualia.)
This was best formalized by Chalmers in his book (the conciouss mind?) in which he argues that even if you knew the complete microphysical description of a person and complete laws that describe physical processes you could still coherently suppose that there wasn't actually anything that it was like to be them. More specifically, he argued that phenomenal consciousness doesn't logically supervene on the physical.
In other words, the point being made is that to explain why we are conciouss you need to propose some kind of natural law which says "physical systems like X have qualia Y" as such a law can't be logically derived from laws which only relate physical stuff.
But if that is the hard problem you are trying to solve then by its very nature you can't give a falsifiable theory -- or more specifically you can't give a theory which can be falsified by physical observations -- since the whole point of the problem is to describe the extra law that relates conscious experience to physical reality (though you can look for elegant theories that jointly describe both and check they accord with our purely physical theories but you aren't offering some unified theory that implies the standard model here).
Yes, it's weird because the problem raises questions about our assumptions about the causal efficacy of conciouss experience (is it necessarily epiphenomenal etc) but that's the problem and that means this kind of theory isn't really addressing the problem philosophers are usually getting at when they talk about qualia (and those who want to reduce qualia to physics like Dennet face the challenge of arguing Chalmers is wrong in his zombie argument not any particular puzzle about the exact brain mechanism even if they may speculate).
Let me add that I was just attempting to give a rough idea of the state of the (imo cogent and compelling) philosophical discourse on the issue. I strongly recommend people actually read Chalmer's book The Conscious Mind if they want to know more. It's very accessible and the first half is very compelling imo (his particular suggestion for a solution less so because I don't think he succeeds in defining what it means for matter to carry out a computation).
If you want to get down into the weeds, I'd actually say it's a bit misleading to assume we have a principled definition of the physical or that we can assume that the ability to predict the future from the past means that qualia are therefore causally impotent (as a humean I see no reason to think that events can't be overdetermined by causes when those causes always necessarily co-occur).
I'd go even further and point out that even in a world where qualia did play a direct and non-redundant causal role its not clear we should expect physics to look different than it is in our world (where it seems fundamental randomness exists) so maybe we shouldn't be that quick to assume the causal closure of the physical.
>If you want to get down into the weeds, I'd actually say it's a bit misleading to assume we have a principled definition of the physical
We can cash out the physical as the objective perspective on an intrinsically neutral universe.
>as a humean I see no reason to think that events can't be overdetermined
Overdetermination is not needed. The subjective perspective can be a valid alternative to the objective perspective without having two competing systems of causality in the territory.
What is an objective perspective? It is presumably an objective fact that a given qualia has occurred at some point in time and perhaps location. We generally don't require observability for something to be physical (we allow that there might be parts of the universe so far away we can't observe them as a matter of physical law).
--
It depends a bit on what you mean by overdetermination. It's merely the same kind of overdetermination we have with gravitational and inertial mass. Since they are equal by physical law you can predict anything you can do with one with the other.
You can always write the laws of nature in an empirically equivalent fashion that renders them not overdetermined (so it's like having both the grav and inertial mass that causes all effects) but for this you have to give up the casual closure of the physical (every physical effect has sufficient physical causes).
But that's all mucking with causation and I think it's clearer just to speak in terms of prediction and it just comes out that you can predict the same future events using either a completely physical description or one which uses facts about qualia and some (but not all) physical facts.
But that kind of overdetermination isn't particularly objectionable.
>What is an objective perspective? It is presumably an objective fact that a given qualia has occurred at some point in time and perhaps location
Its the thing that's emphasised in science.
>It's merely the same kind of overdetermination we have with gravitational and inertial mass. Since they are equal by physical law you can predict anything you can do with one with the other.
I don't consider that to be real ,ie problematical,overdetermination, since it's not in the territory.
>But that's all mucking with causation and I think it's clearer just to speak in terms of prediction and it just comes out that you can predict the same future events using either a completely physical description or one which uses facts about qualia and some (but not all) physical facts.
Prediction wouldnt be possible without some kind of causation. If you believe there is some kind of in-the-,territory causation that can be described in various terms , we are nearly on the same terms.
I know it when I see it is hardly a principled distinction. You can obviously just label things in any old way but you are basically accepting it's not principled.
-
Well if that's not problematic burden is on you to explain why this case would be.
-
It's not at all clear there are any extra facts about causation needed beyond what actually happens. Occam's razor says that we should be fine just saying: the universe obeys such and such regularities without any need to say some of those are special causal regularities and others aren't.
I mean the laws of physics let you predict the past from the future too. Are you committing to the idea that one of those directions is causal and not the other? But then why should our arrow of time match the causal one?
If you want to just use causation as a synonym for "required by natural law" fine but then everything I said stands.
>he argues that even if you knew the complete microphysical description of a person and complete laws that describe physical processes you could still coherently suppose that there wasn't actually anything that it was like to be them
I expect that by the time that we actually have something approaching such understanding then hard problem would have long been dissolved, therefore we wouldn't be able to coherently suppose that.
But that basically is the hard problem. The phrase was defined by Chalmers to refer to the problem of explaining why, even if you've identified and completely understand the microphysics, why those configurations are the conciouss ones. If you don't assume qualia are not reducible to (or more accurately logically supervenient on) physical states there is no hard problem.
Hmm, so he posited an entirely hypothetical problem, given that our actual understanding is nowhere near that? Doesn't seem particularly useful. My intuition says that it's precisely our lack of understanding of physics and engineering of the brain that gives rise to the confusion about why we have first-person experience, or qualia. This is what I think the "hard problem" commonly refers to, which makes more sense, since it's a real problem instead of a hypothetical one.
Then why are you here reading a blog post that's clearly claiming to solve that kind of hypothetical problem? I mean on your view you should have shut the browser the moment you realized it wasn't a deep dive into neuroscience since -- in terms of purely I/o there is no more to explain about consciousness than there is to explain the operation of the vision system in the brain -- and certainly nothing useful to say that isn't a deep dive into neuroscience and brain regions. So why are you even here?
But frankly the primary real world application of philosophy like this is to prevent people from wasting even more time by recreating it poorly like this post. People seem to have a near infinite desire to recreate philosophy (which is somehow only pointless when it's not what they are thinking up) and it's far better to have a few people lay it out the arguments/concerns in an actually compelling and clear as possible so it's not reproduced a hundred times with confused arguments.
Mainly I read stuff for entertainment, this post is no exception. It does seem pretty confused about philosophy, but that doesn't matter. Everybody is confused about philosophy, but given that it proposes to actually do stuff, that might end up leading to new insights and eventually resolve this confusion. That's the only way they ever get resolved.
I didn't fully follow this, but I feel like it mostly just shifted the problem of explaining 'qualia' to explaining 'our experience of neural oscillations from the inside' ?
Fake explanations don't reconcile complex unexplained observations with point by point derivation from simpler postulates, and they especially don't make novel testable predictions. This does both.
>Even unusual conscious experiences caused by intense meditative practice now seem to me fully describable without resorting to the religious terminologies of the traditions that produced them20.
...is a claim of describability, not a description.
> a prospect for technologically-mediated telepathy.
Yes I am saying that. The most direct way I'm saying it is when I claim that it is possible to design non-traditional forms of meditation where arbitrary single qualia fill all awareness.
I also happen to write non-traditional meditations. Total coincidence of course.
Excellent topic with a lot of good seemingly objective analysis regarding the physiology of action potentials and their cumulative impact on thoughts and consciousness. I was following along in mutual agreement up until it veered into the “Payoff” and was having trouble appreciating how it was all being tied together. It became a bit much to process tonite, especially after imbibing with friends, so I’ve stuck a pin in it and am looking forward to returning to digest it tomorrow, with a clearer head.
I did have one quick thought tho! In the qualifications of quaila and thought perception, you suggest:
“Irrevocability: qualia can’t be directly overridden by top-down attention any more than other factual information being processed can.”
Which I take to imply you cannot just ignore stimuli to the point of not experiencing it. And that that experience exists independently of your own recognition. If I got that all wrong, then stop reading now.
Does that contradict the idea of desensitization treatment and behavioral adaptation to PTSD triggers, sometimes with guided meditation techniques or dissociative chemicals? Is that global awareness of synced oscillations an instance of an outside thought recognizing the deleterious impact of the internal thoughts manifesting a belief perceived externally as a disproportional or irrational thought response? And if so, is the conscious act of attempting to control your thoughts and succeeding in moderation of irrational oscillation cycles, a direct top-down override?
Desensitization treatment isn't a *direct* override - it's indirect, partly mediated though control over outside sources of the relevant stimuli. Also tends to be far from absolute.
great article, but misses a very subtle but key assumption: neuroscience can observe "signals" on the brain, subjectively you can also observe a phenomena at the same time "eating chocolate/stress", but there is no correlation that signals cause the subjective experience. the whole premise of your experiment assumes causation of signals causing the subjective experience, but you can never prove that!
eg: you mix chemicals A & B and see a blue colour in the beaker, can you infer that the colour blue is the reason chemicals A & B are mixing? byproducts of a process are not the ones causing the reaction. in chemical reactions you can see the chain of causation, with neuroscience you can never tell if the signals are byproducts or ingredients!
"The color blue" isn't the reason why the chemicals are mixing, nor is "the chemicals mixing" the reason for the creation of the color blue. "The creation of the color blue in the beaker" and "the chemicals mixing" are one and the same thing— there are many things that can cause us to observe blue wavelengths (Rayleigh scattering, fluorescence, reflection, absorption, heat production) but none of them are "chemicals mixing". Either a new compound is created (and that compound *is inherently* blue; you can just as easily say a chemical exists because it interacts with electromagnetic radiation in a certain way as you can say the electromagnetic radiation acts in a certain way because the chemical exists) or a new physical network is created (same thing).
That is to say, "the color blue" and "the chemicals mixing" arise from a third thing, "the force of electromagnetism". What is "causing" the reaction is the collision of various molecules with a low enough potential energy barrier to allow them to overcome repulsive forces, and enough ambient energy to allow for that collision.
To put this back into the terms of the hard/easy problem of consciousness, one might say that what the hard problem of consciousness seems as if it is asking is "why is this compound formed when molecules collide", but it is in fact asking "why do molecules collide at all". Subjective experiences arise (and the geometry of the brain is what it is) for the same reason that some chemical compounds reflect/fluoresce/absorb enough to allow blue light to be sensed, and for the same reason that chemicals combine to make other chemicals. It is because physical forces and laws are what they are. It would be nice to know why those laws *are* what they are, but that's not a question about consciousness or chemistry!
Incidentally, the molecular orbitals which cause certain wavelengths of light to emit from molecules and which cause molecules to combine in particular ways are also oscillating wavefunctions.
All theories are just word salads, fluorescence, electromagnetism you can keep creating new words but you can never explain chain of causation fundamentally.
Similarly you can just handwave that consciousness is emergent property just so it's easy for you to not think about it, but no theory will be able to postulate how the subjective experience arises
All theories are just word salads, fluorescence, electromagnetism you can keep creating new words but you can never explain chain of causation fundamentally.
Similarly you can just handwave that consciousness is emergent property just so it's easy for you to not think about it, but no theory will be able to postulate how the subjective experience arises
Yes, that's fundamentally what I'm saying. The standards to which you hold explaining consciousness makes "what *really is* a chemical reaction" just as impossible to explain. Of course, you can definitely bite that bullet and say we will never solve the hard problem of how sodium and chlorine make table salt, but it's good to understand that you're rejecting physicalism in many more arenas than just the brain when you reject physicalism in the brain by this particular standard.
Problem with physicalism is it can explain everything in the world with other abstractions, but when it fails to account why I feel conscious, then it is a useful abstraction but not the truth.
The opposite way is start with something you can take as truth, your own subjective experience, you don't need any assumptions or explanations to know you are conscious, "you just know you are" - it's self evident!
So now with conscious as fundamental you can try to explain how the physical world will exists wrt you the observer. If you want to read more about this look up Advaita Vedanta
You make a big deal of "oscillations" but it's not clear the word has any more explanatory power in your story than any other word for prolonged brain activity would have. You connect these oscillations with thoughts thinking about themselves, but these two types of loops have nothing to do with each other so far as I can tell.
The oscillations in the brain may well serve no more role than the clock cycle of a microprocessor: the periodicity helps prevent different runs of the same circuit from conflicting, but nothing more. Even for the easy problem of consciousness, I expect the true answer to be vastly more complex than just "oscillations".
You're right, the true answer is vastly more complex than just "oscillations". This essay is also vastly more complex. The question is whether this essay is (some approximation of) the true answer, and simplifying it so resolutely isn't going to answer that.
So he claims. His essay came across as empty sophistry to me, though; I don't really believe there's a well-defined non-tautological theory of consciousness here that's different from just the word "oscillations" underlined 3 times.
The core of this theory seems to be that the brain can represent its own processing to itself. It proposes some mechanism for it, although it was pretty vague - is it just that oscillations can have different frequencies? Seems more like a potential per-requisite than an explanation.
In any case, the theory just equates the ability to represent its own processing with consciousness and calls it a day. Why would the brain do this? How does it explain phenomenal consciousness? What does it predict about altered states of consciousness, or pathologies of consciousness like blindsight?
The proposed experimental tests are not tests of the theory, they are just tests of the link between attention and brain rhythm synchronization, and they're being done all the time.
I think the Attention Schema theory beats this one on all accounts. It says that the brain represents its own attention (not all processing, not all aspects of processing), for the purpose of controlling attention, and exapted as a theory of mind module.
No. Lots of things can "represent themselves", including a piece of paper with the words "this piece of paper". This theory makes a much more specific claim.
I didn't get into blind sight or altered states of consciousness because the essay was already too long, which is evidently leading to people commenting who haven't actually read it.
Do you have specific questions about blindsight or altered states of consciousness?
What is the more specific claim? I read the whole post, but the essential difference escapes me. I can find these quotes:
"the processing of information that is part of the oscillation/thought is handled distinctly from the processing of information that is not (...)
the distinction between internal and external information is neuronal activity, i.e. the exact kind of information that an oscillation/thought can process, it can notice this distinction and thereby notice itself"
"Qualia are nothing but information being processed internally, on their own information channel, encoded in the rhythm of the oscillation."
"When neuronal activities reflect each other, they can achieve self-reflection."
"oscillations/thoughts with memory are themselves additional information processing systems"
I'm not sure whether it's one oscillation representing another, or a single oscillation representing itself. It sounds maybe more like the latter, judging by the last quote - you're saying that a single population of neurons is doing two computations at the same time? A physical computer running a virtual machine is still switching between computations very quickly, but it sounds like you're saying the same activity should be interpreted in two ways simultaneously, and the latter is reflecting on the former?
Whatever it is, it still sounds like self-representation with extra steps. I don't have an issue with that, since I also think that's a core mechanism behind consciousness. I just think one needs to explain 1) why this self-representation occurs 2) why it causes people to claim phenomenal consciousness.
It's a single oscillation, but not all neurons in it are representing the same thing. A (very) simplified model would be two subgroups inside the same oscillation each representing the other. (In actual fact, oscillations aren't simply going back and forth, the circularity of their activity is more varied.)
This occurs because it allows information to persist over time, i.e. it implements working memory, which has evolutionary fitness value.
Why are we less conscious during sleep if oscillations are equally or more present? How do oscillations implement working memory at the same time as they implement self-representation?
Oscillations are not equally present in sleep. It depends on the phase (REM is rhe most widely known) and that happens to be a fomain of electroencephalography research. There are literally thousands of papers on oscillations in sleep and how they relate to the strange quasi-conscious phenomena of dreams.
I'm not getting the picture. If any kind of neural oscillation somehow causes consciousness, then humans during sleep and all animals should be roughly equally conscious as awake humans, people having epileptic seizures should be more conscious and neurons grown in a petri dish should be conscious. If your theory says it's oscillations plus some extra conditions then I don't know what those conditions are. Most importantly I don't see what explanatory work oscillations are doing in the theory.
Would every possible information processing event (occurring in a computer, for example) that had the same properties 1-15 (Distinguishability through Unity) also have qualia, or is it only these oscillating thought patterns that have qualia? If the latter, what explains why? Would a simulation of a thought (a virtual machine) running on a computer also have qualia under your theory? How would we know?
I don't know, because I don't know how the simulation would work in detail. "Simulation" is a super broad term, much like "calculation", it is hard to make statements about all possible members of that category.
Yes, but I make no claim whether (and therefore don't rule out that) any other phenomena caused by anything else than these bio-neuronal dynamics deserve to also get the "consciousness" label.
Thanks for writing this. I appreciated consciousness being described as a property of thoughts rather than a unitary self. Like others, I also see this as an argument for physicalism, but that should be the default view. Qualia seem to be the last vestiges of what 16th century monks called a soul - the idea a human mind has something special going on that other information processing hardware does not. Red can’t simply be light 620-750 nm.
We can’t see into other minds. We can’t experience a subjective consciousness outside the one instantaneous moment we are in. Our train of thought is on a railway of confabulated and fragmented RAM. To me the framing of a “hard problem” seems not only human-centric, but solipsistic and leading inquiry far astray from any universal truth of nature.
X satisfies (1) - (16), Y satisfies (1) - (16), therefore X is Y is not a correct argument. Even if Y predicts characteristics of X, e.g. "my hamster is brown and furry, so is my dog, therefore my dog is my hamster" is not any more true regardless of whether this theory predicts e.g. that your hamster needs water to survive because your dog does too.
Some other gaps in the reasoning:
> "A neuron that is part of one oscillation can hardly also be a part of another, so oscillations compete for neurons."
This seems very non-obvious, mathematically it's very easy for signals to be part of multiple oscillations.
>Mental information processing doesn’t seem possible without these electric spikes.
The claim here is that the spikes are necessary, but this is a distinct claim from them being sufficient.
Thanks, you said whether I wanted to but more clearly.
In general, all respect to the author, I think the theory has too many free parameters and can easily be made to fit any observations. For example, re Unity, why should separate thoughts 'compress' their self representation into an I rather than (for example) just reduce granularity in a general sense - such that reflected thoughts are fuzzier than more directly experienced ones? (Perhaps this is actually true, but isn't very compatible with the author's model of jhanas).
I think oscillations attuned to other oscillations might be a relevant part of the picture, but in trying to present them as a complete explanation, the post tries to prove too much.
Equating the hamster and dog fails because they have other properties. The argument is correct if it lists ALL properties. These are all 16 properties of qualia I could find that aren't inconsistent, metaphorical, or poetry. To show how the argument is incorrect here, please show properties of qualia that aren't also properties of information being handled by processing inside the same oscillation that also recursively reflects them.
Yes signals can be part of multiple oscillations, if their frequencies are natural number multiples of each other. That's why I wrote "hardly", not "not".
Are you claiming that the brain can process information without spikes?
The same water molecule can be part of several different waves. It doesn't matter in the least whether their frequencies are integer multiples of each other. The waves will pass on undisturbed, and the water molecule will end up where it was. Waves are weird.
I don't think it is a fair point, nonlinear systems can do weird things like allow solitons; it's a bit hard to speak of "waves" for nonlinear systems but there are definitely wave-like behaviours. FWIW the shallow water equations *are* nonlinear.
Also fair. And yes, the water equations are of course nonlinear. I just decided I didn't want to argue about different types of nonlinearities and which behaviors might show up in these particular waves, and why, none of which I'm expert on. But yes, for especially small oscillations in even a nonlinear system, one might expect to see superposition in some circumstances.
>The argument is correct if it lists ALL properties
>These are all 16 properties of qualia I could find that aren't inconsistent, metaphorical, or poetry.
Right, so necessarily ALL properties, even by your measure. Whether or not A and B are equal if all properties of A that can be discussed are equal to properties of B that can be discussed is very much the crux of the issue.
>To show how the argument is incorrect here, please show
No, the argument "A is brown and furry, so is B, therefore A = B" is wrong even without me proving that A != B (Yes, I know about Gödel's first theorem, no, it does not apply since this is philosophy and not (first order) logic).
>That's why I wrote "hardly", not "not".
This is very much non-standard use of the word "hardly"
>Yes signals can be part of multiple oscillations, if their frequencies are natural number multiples of each other.
This may be true for standing waves for e.g. the linear wave equation, but there are plenty of (nonlinear) systems with weirder behaviour that admit things like solitons or whatever you want to call the solution of the Kuramoto–Sivashinsky equation.
>Are you claiming that the brain can process information without spikes?
I am making only claims about the argument, not about the brain itself.
Yes, your criticism of the non-standardness of "hardly" is valid. I shouldn't have written this that way. Thank you. I promise I will do better on the next version of this theory.
Now whst do we do about the crux? I suggest we call this problem "if it looks like a crux and it quacks like a crux and it walks like a crux".
> I can resolve to not anthropomorphize myself for once, and let my thoughts see themselves as thoughts.
"I" can resolve? Who is this "I" you speak of, I thought everything was just thoughts? :P
If it's just thoughts all the way down, why not just let it be atoms and molecules all the way down? Also, just because you can convince yourself that it's thoughts all the way down does not in any way prove anything about the nature of consciousness.
"I" second this. Throughout the post he uses contradictory language, like:
>An additional error is made when people identify themselves with “their” consciousness.
Who is the person you are talking about? On your theory there are no people, only thoughts.
>Not only do they wrongly assume that there’s a thing there to identify with; this assumed identity also creates false intuitions:
Who is the "they" making a wrong assumption? There are no theys, only the wrong thoughts themselves; yet these wrong thoughts identify with themselves, and create an assumed identity that doesn't exist? How can someone that doesn't exist assume an identity?
>The supposed thing-ness of consciousness seems more real, since people usually believe themselves to be real.
Again, what "people" are there that can believe things about "themselves"?
>Consciousness seems like there’s only one of it, since people usually believe there’s only one of themselves.
Again, who are these "people" who believe that there is only one of "themselves"? Under your own theory, there are no such people.
Statements like these want to have their physicalism cake and eat it to: "I" is only an illusion, an illusion that I (who don't exist) experience. Nonsense remains nonsense even when we talk it about consciousness.
"I" and "people" are shorthands, useful because "the organism that is having this thought" is an exhausting thing to type five times per paragraph. Yes this causes confusion, as all simplification does, but that says nothing about the validity of the actual argument.
What is the meaningful difference between "I" and "organism" in your view, such that the statement "I am having this thought" and "The organism is having this thought" are different?
The difference is that most of the organism is not saying "organism". Like, your pancreas is not choosing the words. It's a thought that does that. A thought that identifies with the organism that thinks it, incorrectly.
How is that different than saying "I am having this thought?" Surely those who say "I" do not believe that their pancreas is doing the thinking either?
In itself, it is no difference, if "I" is just a shorthand, literally just a vowel. The confusions arise if "I" is treated as more than that, as some special thing that can be isolated and studied.
There is no actual unitary "I" (although suspiciously, all of this is happening in a single actual unitary universe, but I'll try not to go mystical for once). Everything I wrote about how "I" is an abbreviation for whatever the sense of self happens to point at, I absolutely meant.
Later I use normal English again. Like Scott told us, you can think "humans" but please say "people" like normal... people. You can make a gotcha out of that if it amuses you, but if you're looking to engage with my argument it is nothing but a distraction.
Great post! I'll have to think more about it but I already feel that I'm a bit less confused about some parts of the problem and that the non-physicalist goalpost moved even further.
Your answer to the question: Which types of algorithms, when executed produce the referent for what people call "consciousness" is dully noted. Looking forward to the experimental tests. I think there has to be something more to it - to experience the quale of redness some part of the system has to generate a picture where this redness is encoded in a particular way which another part of the system is able to decode. But what you are talking about indeed seem to be part of the answer or at least a fruitful direction to pursue.
the nondeterministic turing machine link is confusing. NDTMs are not believed to be physically possible. perhaps you mean not deterministic, like probabilistic
Yeah, because of that I don't recall discussing them in any CS course. There was the concept of an "oracle" which is assumed to behave in some way to give a result, but that was used to assume something and then prove a contradiction to show things like a general solution of the halting problem to be impossible. The wikipedia link sounded a bit like a quantum computer, where multiple solutions are attempted but only the correct one is used.
There are two equivalent accounts of Nondeterministic Turing machines, one is that at every multiple choice branch where each branch looks appropriately equivalent, you spontaneously duplicate the machine, such that both branches are taken. Then, the first machine that halts has its time taken / space used to determine complexity class you are. We ignore all other possible Turing machines that went down any wrong branches.
The other, equivalent construction, is that at every branch, the Turing machine "magically" picks the correct branch every time.
So even if the non deterministic Turing machine is not physically realizable, it is also a way of saying "okay, right now we don't know how to tiebreak on these branches, but if we had a perfect tiebreak oracle, here's what the performance would look like".
It's possible I just didn't fully understand the article, but 15 still confuses me. If you can have multiple conscious thoughts in your brain at once that aren't directly aware of each other, would this be noticeable in our behavior? For instance, if you have two thoughts going on at once, "I'm hungry" and "I'm sad," and one of them causes you to say out loud that you are hungry, does the other thought hear that and get confused because it perceives that it's just sad, not hungry? Or does focusing enough to make a decision cause oscillating thoughts to merge, so there is no confusion?
When I remember experiencing multiple qualia at once, are some of those times actually the memories of multiple thoughts that got stitched together after the fact? Does focusing on all the different qualities you are experiencing cause thoughts to merge?
I think maybe when I'm trying to understand 15 I'm subconsciously mapping it onto the concept of multiple personalities, when most thoughts probably aren't that complex. Could that be the source of my confusion?
They merge easily. I have often had the experience of hearing myself say something unexpected (doesn't that happen to everyone?) and when that happens I usually trust it to be true. I think this trust is what the merging feels like from the inside.
I guess what I'm confused about is how two different simultaneous conscious thoughts get "recorded" in your memories. When you remember the moment where you said something unexpected, how come you remember the thought that was surprised by you saying it, but you don't remember the other thought that came up with what you said and therefore wasn't surprised by it at all?
If you were to have two thoughts at the same time that were both conscious and didn't merge at the time, how would they be remembered? For instance, if I had one thought that I was hungry and one that I was sad, would thet both get recorded so that later I remember both feelings simultaneously, even though they were unaware of each other at the time? Does that question make any sense at all or does it misunderstand how consciousness relates to memory?
When I pay attention to (meditate on) two conscious thoughts merging, It feels like two parallel universes merging and like I was previously in both but now I’m in just one.
This makes me very interested when Bayesians talk about the terms of the Bayesian Theorem in words like “a world where…”
An interesting aspect of newborn baby behavior I've observed is the baby simultaneously trying to satisfy two different biological urges. Namely to drink milk and to relieve itself. It's drinking, it stops to strain and grunt for a while while also drinking a bit maybe, drinks some more, etc. It's like both urges are independently active, with the baby kind of drifting in response from one to another. Absent is the adult way of controlling/prioritizing to comfortably achieve both; like for example after running a race you'll catch your breath first before getting a drink.
Thanks for this! As others have noted, it doesn't really get in to the philosophical weeds about how (even in principle) we could go from a physicalist universe of quantitative relationships between 'empty' atoms, to the subjectivity of qualia; but then, a finessed one-to-one correlation between those subjectivities and specific neuron activation states would be quite the prize in itself, so that's no great failure.
A couple claims that I understand to be common (if not true), in relation to this subject, which I was surprised weren't mentioned and might be consequential for your discussion:
From the mindfulness meditation angle, I've seen it claimed in multiple places that in fact only *one* quale can be present in conscious attention at the same time; that our ability to hold multiple observations/qualia/tokens in place at the same time is actually a rapid juggling maneuver. It is claimed that we cycle between our one quale so rapidly that it looks like multiple qualia at once, much like rapidly flicking images in a film can become the illusion of a moving picture. And I'm curious whether you see your theory here as being robust against that claim! (I first ran across this claim in "The Mind Illuminated", to cite my sources.)
From the neuroscience angle, I understand that the size of neuron activation cascades isn't just a 'small' versus 'large' distinction, but rather, that they follow a power law- not unlike actual physical avalanches in gravel or snow that is stacked at a critical angle. This doesn't directly contradict your premise, but it doesn't seem like an intuitive consequence or prediction from it, I don't think. Actually, I would think that additive combinations of small sense-data pulses would follow a bell curve instead (as you'd get if you add up the results of a 6d6 die roll, for example). Is there any particular reason why you would expect a power law result, given the processes you have described here? For a robust treatment of this claim, you can check out "Criticality in Neural Systems", edited by Plenz and Niebur- though it's a decade old at this point, so it might have missed some of the most interesting modern research.
Ugh. Another article in the Hofstadter tradition of short-circuiting philosophical reflection by massively begging the relevant questions.
In "European Literature and the Latin Middle Ages," Ernst Curtius tries to give a long history of European literature in terms of topoi, common rhetorical constructions which repeat and persist across ages. The "past golden age" or "descent into the underworld" are structures so repeated in literature that they form a background part of consciousness, seem entirely normal. TVTropes catalogues these things for modern media. What Curtius does is take in mind such a long timeframe that he can see these tropes in the longue durée, comparing individual uses with others in a long history to take account of the warp, not just the weft, of literary structures. He encourages a reading of literary topoi which sees small inventions and novelties disappear in the longer, bigger continuities of literary history.
The history of philosophy has its own topoi. Getting bogged down in the details of individual philosophical arguments--which is by far the most important portion of philosophical training--should not blind us to the ways that the general shape of arguments have a long history. Recursive reflection, for example, is a philosophical topos that has dominated a strand of German philosophy since Fichte--who the author should read, not as inspiration, but as a sort of warning. Fichte--like Schelling and Hegel after him--that self-reflection is somehow the key to consciousness. His texts founder, though he is not always aware of this, on a series of difficulties which beset 150 years of German philosophy which aggressively takes on the problem: that self-reflection is almost impossible to logically and consistently define, that there is no logical guarantee that the thought thinking and the thought thought of are the same entity, that the identity of the thought thinking and the thought thought is almost metaphorical, that the intentional consciousness (seeing something as) which is the phenomenological counterpart of metaphor is aggressively inconsistent with the basic metaphysical tenants of physicalism, etc.
This post aggressively leans into the error of thinking that something like "self-reflection" describes something which can be defined, rather than something which is part of the basic shape of Western philosophy which is both indispensable and completely resistant to all our efforts. At the same times, it leans on other ideas, e.g. "information" which it treats as basic even when such things are models and metaphors.
Now I am curious. What is information to you? "Information" seems an almost impossible to define word. We can define it in very specific contexts, but much like "game", once the context differs, the definition differs. Or do you think the Shannon definition of "information" suffices to cover all usage?
Information is an extremely contested concept in philosophy, and it's extremely difficult to define it in a way fundamental enough for it to have the conceptual leverage you want out of your system that doesn't automatically entail a commitment to physicalism. Janich's "What is information?" is an accessible start.
As for begging the question: You assemble a list of properties of qualia which come from a variety of sources which originate from different perspectives: (taking your info from Dennet's "Quining Qualia" is more than a little bit tendentious)!
You struggle to maintain a principled differentiation between the concept of thoughts and qualia, and as the concepts elide, they cross casually the very demarcation line the concept of qualia is designed to draw our attention to.
However you want to label the problem, whether you call it Leibniz' gap or the Hard Problem or the problem of qualia, etc., you need to answer the problem on the level that it's posed. The problem is not one of creating plausible-sounding accounts, or generalized accounts that sound more plausible than the non-physicalists--every non-physicalist already knows that the non-physicalist account seems aggressively implausible. The problem is posed by a few edge-case logic problems that seem stubbornly resilient no matter how sophisticated the science is. And addressing these is not a matter of speculation about the brain, but very careful dissecting these on the level they are posed.
"the world of objective facts, physics and brain imaging technology, where nobody has found these supposed qualia as a thing to point at." - why does nobody mention, at this point in their consciousness article, that they are then and there writing about qualia in the world of objective facts, and so at least on a meta level qualia do have a clear and measurable effect on the physical world? I cannot point at qualia, but I can point at a lot of physical effects directly downstream from qualia, which is surely almost as good. (Is this the great blind spot that makes people write consciousness articles in the first place?)
I'm of the opinion that there are only easy problems, and once all the easy problems are solved there won't be a hard problem left. With that in mind this prospective explanation - information flows around through neurons somehow resulting in the experience of qualia - is as good as any at this level. I don't mind that this article doesn't address the hard problem, since it doesn't exist (IMHO), nevertheless vaguely handwaving in the direction of recursive processing and "information" is not yet a physicalist explanation of how consciousness comes about.
"Qualia are *nothing but* information being processed internally, on their own information channel, encoded in the rhythm of the oscillation." - I think the actual explanation should go much further than just stating this simple equivalence. "Qualia [are] bits of information" - but "bits of information" is only an abstract idea that we use in apprehending the physical world. Information theory can be part of a framework that explains consciousness, but consciousness is not explained by simply equating it to some aspect of information theory.
Daniel - I love this theory. I love how you write and how it included a testable hypothesis. Thanks!
It is now my new head-canon. It speaks volumes to your writing and your theory that “it feels so obvious that it can’t be significant”, and yet, it fully explains consciousness and how our brains work in a way I’ve never groked before.
But, reading the comments, it seems people are caught up on old, less relevant debates and discourse about mentality and philosophy. I think the next step, and perhaps harder step, to presenting your theory, is something that Scott often does, and that’s break down / steel man the classic debate, and show why it does/does not matter anymore. The fact is, no matter how much philosophers debate about how many fundamental elements there are, sometimes you just need to get out an electron microscope.
Thanks. It appears that I am so firmly in Camp #1 that I don’t even understand the fuss about qualia from camp #2. Not having heard about it before, except in these comments, makes it feel like peak-over-philosophizing. I could also say that it’s impossible to model the best bridge design because first you need to define the inherent philosophical nature of bridge-ness, but I’d rather model them based on measurable factors like ability-to-be-crossed and cost.
I've waded into this stuff a fair amount... Had some conversations long ago with possibly one of the smartest people I've met who is firmly in camp 2...
But it seems like these really are two kinds of people who will forever be at an impasse.
I agree; it absolutely sounds like, "peak-over-philosophizing". Arguing about how many angels can dance on the head of quale.
In support of there being two kinds of people, and you either "get it" or you don't, is the experience of C. S. Lewis, as written in his autobiography:
"We had been, in the technical sense of the term, “realists”; that is, we accepted as rock-bottom reality the universe revealed by the senses. But at the same time we continued to make for certain phenomena of consciousness all the claims that really went with a theistic or idealistic view. We maintained that abstract thought (if obedient to logical rules) gave indisputable truth, that our moral judgment was “valid”, and our aesthetic experience not merely pleasing but “valuable”. The view was, I think, common at the time...Barfield convinced me that it was inconsistent. If thought were a purely subjective event, these claims for it would have to be abandoned. If one kept (as rock-bottom reality) the universe of the senses, aided by instruments and co-ordinated so as to form “science”, then one would have to go much further—as many have since gone—and adopt a Behaviouristic theory of logic, ethics, and aesthetics. But such a theory was, and is, unbelievable to me. I am using the word “unbelievable”, which many use to mean “improbable” or even “undesirable”, in a quite literal sense. I mean that the act of believing what the behaviourist believes is one that my mind simply will not perform. I cannot force my thought into that shape any more than I can scratch my ear with my big toe or pour wine out of a bottle into the cavity at the base of that same bottle. It is as final as a physical impossibility. I was therefore compelled to give up realism. "
It's really amusing seeing the people who think qualia must me something that other people have but they don't, to whom explanations just sound like word salad. It seems obvious that they are just assuming it must refer to something more special and weird. I get it; I've done plenty of meditation; "qualia" are the things that you can be mindful of when you practice mindfulness. It's a fancy way of saying "experience" or "sensation". And I get why it seems kind of baffling how it could result from matter. But I don't share the intuition that it's Super Super Special and Hard... There are a lot of things that have been revealed by science that I know I'll never be able to wrap my head around, like quantum physics, but that doesn't lead me to conclude that science is flawed, and that I can then come up with an alternative metaphysics that is a truer understanding of reality by sitting around navel gazing. Ok, sorry, I'm being too hard on the other side now. But...we're just different kinds of people I guess, and there is no way to bridge the gap. That's what I have today decided to dub The Harder Problem of Consciousness.
I do think if we can reduce qualia to the physical brain, recursive neuronal processes are the best bet. But to really make it work, I think our Physics has to get friendlier with Panpsychism. Maybe our Physics already is on that path: It from Bit, the integration of Physics and Information Theory - these point to a conception of matter that is not so knock-on-wood hard as the old Stuff.
By your terminology, would 'sapience' be a mainly human thing, but 'sentience' more widespread on the Phylogenetic tree?
This is a random aside, but...sapience and sentience seems like equality and equity; basically synonyms, but the concept they refer to has at least two similar but distinct meanings, so some people try to declare that one is to be used for one meaning, the other for the other, but no one can ever really agree on which is which.
I guess as long as we define our terms in a given situation, we're okay. He does explain his use of 'sapience' in a footnote, and it's distinct, I think, from sentience.
I don't know what "Physics getting friendlier with Panpsychism" would look like. Like, how would a world where panpsychism is true be different from a world where it isn't? Physics is interested in panpsychism if and only if there is a tangible difference.
Hi Daniel. Perhaps a world where Panpsychism is false wouldn't have consciousness. The idea being that brains can generate consciousness only if matter already has something mind-like, some (albeit primitive) experiential "inside" even at the elementary scale of Physics.
I don't think that is true. I don't know how to explain the functioning of the brain without (something that has all the properties of) phenomenal consciousness. So I don't think we get a choice between a world that has consciousness and one that doesn't.
I find it great that you put your theory into testable hypotheses. I have little objection against the high-level ideas behind them. That said, you may or may not be amazed that we have measurements about neuronal activity that go beyond your wildest dreams. The downside: I think your concept of oscillations does not pan out.
First about the measurements: in animals like rats, researchers can do a pretty amazing sequence of steps:
- genetically alter the rats so that neuron light up whenever it fires a spike (nowadays they can even use different colors for different types of neurons)
- remove the skull and replace it by a glass skull
- glue a tiny camera on the brain and you can see the neuronal activity of all superficial layers of the whole rat brain. While they are normally moving around and interact with all kind of stuff.
In humans, we don't alter them genetically, but we do implant electrode arrays into some humans (usually to prepare certain operations that they need). Those arrays measure neuronal activity, and you can tell apart a few dozen neurons. This doesn't give us a full picture, but in general the activities of these neurons is similar to the activity of rat neurons. The very strong hypothesis is that there are no huge architectural differences, or differences in activity patterns, except that the human brain is a lot bigger.
Now, I think your conception of oscillations is a bit off.
- It is true that if you average over millions of neurons then you see oscillations. Even those are not periodic at all, but pretty irregular. But individual neurons do not show oscillations in their spike patterns at all. Instead, what happens is that all neurons in this region are simultaneously in an UP state or a DOWN state. In an UP state, all neurons have a voltage level which is pretty close to threshold. So just a little bit of input makes them fire. In a DOWN state they are far from threshold, so even a large input usually doesn't make them fire.
- Most individual neurons will fire with a really low frequency of 1Hz or less when you look at a behaving animal. If you look into older literature, you will often find higher firing rates of 10-50Hz, but this is when people anesthetized the animal, found a stimulus that triggers this neuron, and repeat the stimulus intensively.
- Even with these hyperstimulations, you don't get a "loop", but the firing is highly irregular. Some papers even claim that it is approximately exponentially distributed, which is in some sense the distribution which is furthest possible away from a regular firing pattern.
- So, there are no loops in the sense that neurons lock into a firing pattern. In an awake and behaving animal, you will typically not find neurons which will participate with more then 1-2 spikes (or bursts, but that also doesn't fit into loops) in any given activity. Neural activity evolves in a pretty spurious way, not in tightly coupled cycles.
Now, that might not be so fatal to your concept after all. I am pretty sure that the idea with loops is the wrong correlate for information. But if you think about it, most of your explanation also work if you replace "loop" or "oscillation" by "whatever neuronal activity represents the information in the brain".
No, I think it would indeed be fatal to my theory - it fails if the neuronal activity isn't looping. Because the looping is required for the recursion.
Yes, individual neurons aren't oscillating (except pacemaker neurons, which I'm surprised you didn't mention despite your evident level of rigor) and as components of information processing they're noisy and unreliable. That's why averaging over millions of them is the correct level of analysis.
>Yes, individual neurons aren't oscillating (except pacemaker neurons, which I'm surprised you didn't mention despite your evident level of rigor) and as components of information processing they're noisy and unreliable. That's why averaging over millions of them is the correct level of analysis.
Correct me if I'm wrong, but that reads to me like you're saying "My theory requires oscillation, and it's true that individual neurons do not oscillate, so that means looking at individual neurons is the wrong way to go about it and we should back up and average things until we can see an oscillation; and the reason we should use that level of fidelity is otherwise my theory doesn't work."
- My hypothesis is that some information structure in the brain is oscillating, because [all the arguments in the post].
- Neurons (mostly) don't oscillate on their own.
- Therefore, if something is oscillating, it must be abstract computational structures made up of multiple neurons (or something like that).
- So we should look for oscillations in abstract computational structures in the brain.
- If we find them, and they exhibit the properties predicted by the post, that provides some evidence the hypothesis is correct.
- If we fail to find them, that provides some evidence the hypothesis is incorrect (not perfect evidence, but the same amount of evidence that looking for something predicted by a theory and failing to find it always provides).
This seems fine to me. I don't think it's circular to say "we should look for the stuff my theory predicts at the level of abstraction that seems most likely to be capable of exhibiting it."
Hm, I am less of an expert for brain waves than I am for single neuronal activity. But I would be surprised if we had loops at any level. We have oscillations, yes, but the word may give you a wrong idea. You could also call it fluctuations, which is technically the same thing without invoking the wrong image of repetition. Both mean that activity changes over time, not that the activity repeats itself in a loop.
We do measure the frequencies of these fluctuations, but this is just the thing that scientists do with any fluctuating system. "Finding" frequencies does not mean that there is repetition in the system.
What you do is a spectral analysis, that is what scientists routinely do with any temporal data. For example, imagine that you observe a park and record every pedestrian that comes along. So, each pedestrian is one signal with a time stamp. That's exactly this temporal type of data.
If you do a spectral analysis, then you will find that some frequencies are more pronounced than others. Perhaps you will find that there are many pedestrians with a distance of ~1 second from each other, because if a group of people walks together, then they are perhaps ~1 second away each other. Other pronounced frequencies are probably 24h (because if there are many people at 6pm one day, then there are also many people at 6pm the next day), or perhaps 1 minute if there are traffic lights nearby which lets people pass with a period of 1 minute. So a spectral analysis might give you three peaks at 1s, 1min, and 24h. (And that is the case with a meaningful signal. A spectral analysis always gives some peaks. If they don't come from signal, then random noise will give some random peaks.)
Now you could argue that 24h and 1minutes correspond to "loops". It's not quite the type of loops as described above, because you never see the same person in distance of 1 minute. And it's not like person A at time x causes person B to be there at time x+1min. The frequency comes from some regulating system (the traffic lights), not from the people interacting with each other.
And for the 1s frequency it's even harder to argue that this is a "loop". This does come from interaction of people, but it's just that some people tend to cluster together in distances of roughly 1s. It is a repeating pattern, so it is correct to identify this as main frequency. But there is no loop.
For neurons it might be similar. If a few neurons in a small region fire more or less together, then there is a dampening inhibitory (negative) feedback to the whole region to avoid that the activity explodes. The neurons may need 10-20ms before recovering from the negative feedback, and then some neurons may fire again, probably a different subset. Then we get a frequency of 50-100Hz, so we get gamma activity, which is associated to cognitive processing.
Slower waves like alpha waves (~100ms difference, or ~10Hz) are probably more like the traffic lights, they are modulated by certain pacemaker cells in the thalamus.
My point in the previous posts was that if we look at single neurons, they show very sparse activity, and are often only active once during a thought of one second or so. This seems more like the situation with the pedestrians, where there are always new pedestrians walking by, than a situation where the same set of neurons keeps exciting each other in a loop.
I worked in EEG research for a couple of years, so I know what a spectral analysis is. But I still don't know what you mean by "absence of repetition".
That the same neuron doesn't spike several times. Or if it does, that the input does not come from the same source for different spikes. As far as I understand, you consider both necessary for a "loop".
I think the problem that people have is really a problem of personality, not of consciousness. "You're telling me that I don't really exist, that it's all just neurons firing off in response to stimuli" doesn't grab people as an appealing concept. And even those who say they accept that there is no "I" still seem to act as if they have preferences and traits and are uniquely distinguishable from a carrot or the wind in the treetops.
Consciousness is not so much the problem - yes, it's physical phenomena in the brain which is a physical object which comes together in obedience to the laws of nature; it's "you're saying I'm not me" is the problem.
There's an anecdote of Wittgenstein wondering about the idea that people used to think the sun revolves around the earth because that's what it looks like, and he asks: what would it look like if it *looked like* the earth revolves around the sun?
We needn't make an imaginative leap here - what it looks like right now is what it looks like for the earth to revolve around the sun because that's the way things really are. But to understand this at an intuitive level (rather than just to accept the explanatory power of the heliocentric model), it helps for people to mentally step outside their immediate experience to picture, e.g., the solar system whirling around so that they can explain to themselves how those dynamics could produce the phenomenal experience they have on the surface of the planet.
People who claim there is a hard problem are really lamenting the fact that a reductionist model does not provide them with a means to step outside their own consciousness in order to picture it from the third person point of view. But the hard problem there is not to be found in the essence of consciousness, it's to be found in a faulty definition of what counts as explanation.
What would it look like if it looked like there was no hard problem of consciousness? Exactly how it looks now.
Under your own theory "Yes I can" is an incoherent statement. You don't exist, there is no "I" or "You" so of course "You" can't get rid of the hard problem of consciousness. Thoughts are happening in your brain that result in you saying "Yes I can", even though the statement has no meaning beyond "A thought is occurring".
Very interesting. I think the “how would the world look different if X were true instead of Y” is a useful way to approach sticky problems like this, since it often reveals the world wouldn’t look any different and the question itself was posed badly or is meaningless. Another example is “free will”: if you assume determinism is true, and then imagine what the universe would look like if humans had libertarian free instead, will how would things be different (this works the other way around too)? If you can’t say how the world would look different if libertarian free will versus determinism is true, then there is something wrong with the questions/concept you’re working with.
I think the same thing applies to philosophical zombies: if the world were filled with philosophical zombies, instead of humans, how would things be different? It’s not clear it would be. Either the concept is nonsensical, or else we are all philosophical zombies.
I guess I'm not clear what role "looking" is doing here. What would it look like if it looked like there were no quantum fields? The answer to that will be different depending on when you were alive. I'm not sure how it helps clarify because we're always trying to explain how/why things look the way they do... but then what we see when we look changes over time based on what we've learned and what technology we have for looking.
How about "what would it look like" being "simplest theory that produces falsifiable predictions based on objective measurements"? I thought this was an obvious interpretation, but I guess it doesn't hurt to spell it out.
In the spirit of the original quip: when we surmise everything is spinning around the Earth, we have to add a whole encyclopedia of weird little "but this but that" to explain the moon, sun, planet movements which don't make much physical sense.
Then one day you Eureka out a hypothesis: "wait, how about we're spinning, the moon is going around us slowly, we and the moon are going around the sun, and those other planets are also going around the sun?"
Suddenly you can explain everything in a few sentences and can spend the rest of the encyclopedia of words going into amazing detail instead of trying to explain the basic situation.
I agree this is a good way for science to proceed. But we don't get to control when suddenly you can explain everything in a few sentences happens.
My perception in this discussion is that there are people who are of the opinion that there is no hard problem and there are people who are of the opinion that there is still a hard problem.
Callum who initiated this subthread is of the opinion that people who think there is still a hard problem are "really" caught in some kind of deluded psychological experience. "People who claim there is a hard problem are really lamenting the fact that a reductionist model does not provide them with a means to step outside their own consciousness in order to picture it from the third person point of view."
My sense is that early on in scientific endeavors that preventing epistemic closure is more important than claiming to have arrived at the right answer when there is still very little evidence. It seems to me we are way earlier in the process of understanding consciousness than the point when we were able to show that we're orbiting the sun and not the other way around.
My general opinion is that the level of conveyed certainty on either side of this perceptual difference around the existence of the hard problem is unearned.
I would put the emphasis on it being a deluded (I'd prefer misunderstood) psychologcial experience in that I think the issues are largely conceptual rather than empirical, hence I don't think we need to await further scientific advances to earn confidence.
To pick up the analogy, it's as if we were at a time when geocentrism made manifest sense to everyone because that's just how things look, then someone comes along and says: "you do realise that geocentrism is only a description of what we see? Maybe something like heliocentrism will bring all the facts together in a better way." And the reply comes: "nonsense! There is no arguing that, from my point of view, the world is geocentric. That is a brute fact. Heliocentrism could not possibly replace this because geocentrism is the very thing that must be explained."
I differ from some others in that I don't claim we will one day bridge the gap between physicalism and qualia - I rather believe that people don't have the immediate access to the nature of their conscious experience that they think they do, and the result is that what they say about it is incoherent.
Where science must lead is not on explaining how qualia come to be but how people come to have the false belief that qualia exist.
>People who claim there is a hard problem are really lamenting the fact that a reductionist model does not provide them with a means to step outside their own consciousness in order to picture it from the third person point of view.
It's almost exactly the opposite: the third person perspective offered by reductive explanation doesn't describe or predict consciousness as it is experienced,...from a first person POV. Which would be the case in a universe in which it completely failed as an explanation. So.we can' tell whether we are in that universe, or the one in which it is correct in some mysterious, unconvincing way. They look the same.
I think these are two ends of the same thought, as what's demanded of physicalism is a unification of the first and third person, whichever way you look at it. When I say that people want to picture their own consciousness from the third person point of view, what I mean in full is that they expect that a complete third person account would have to entail an ability to compare or simulate other consciousnesses from the first person point of view (such that you would know what it's like to be a bat), and this ability would necessarily apply to one's own consciousness, which could then be recursively stepped outside. The fact that this expectation is (in my opinion) an incoherent description of what an explanation ought to be is why we do not need to doubt the non-existence of the hard problem.
That explanations should predict their explanada is a completely standard.criterion that applies to.everything else.
Why don't you expect an explanation of qualia to.actually predict or convey qualia? Because it's impossible.for.a linguistic (including mathematical) expression t o convey a quale? Well.that would be qualia being ineffable, as require d.
Because it would .involve placing someone into an appropriate brain state ,.so.they can witness the quale from.their first person perspective? That concedes that qualia are.private. Subjective, etc, as required.s
What you need is an excuse for the lack of an.answer to the hard.problem that isn't based on qualia having all the properties they are supposed to.have.
I'm describing the incoherent expectations that are at root of the supposed hard problem. I don't myself concede that there are any such things as qualia in need of explanation.
I am pointing out that the expectations are in line with what usually required of a successful reductive explanation, and there is no incoherence unless qualia are intrinsically subjective, etc
Physicalists sometimes respond to Mary's Room by saying that one can not expect Mary actually to actually instantiate Red herself just by looking at a brain scan. It seems obvious to then that a physical description of brain state won't convey what that state is like, because it doesn't put you into that state. As an argument for physicalism, the strategy is to accept that qualia exist, but argue that they present no unexpected behaviour, or other difficulties for physicalism.
That is correct as stated but somewhat misleading: the problem is why is it necessary, in the case of experience, and only in the case of experience to instantiate it in order to fully understand it. Obviously, it is true a that a descirption of a brain state won't put you into that brain state. But that doesn't show that there is nothing unusual about qualia. The problem is that there in no other case does it seem necessary to instantiate a brain state in order to undertstand something. If another version of Mary were shut up to learn everything about, say, nuclear fusion, the question "would she actually know about nuclear fusion" could only be answered "yes, of course....didn't you just say she knows everything"? The idea that she would have to instantiate a fusion reaction within her own body in order to understand fusion is quite counterintuitive. Similarly, a description of photosynthesis will make you photosynthesise, and would not be needed for a complete understanding of photosynthesis
This feels philosophically quite sloppy. You define qualia as discrete abstractions and especially hilariously define them as needing to "last around 80ms". You have rooted you theory of consciousness in SI. This is completely upside down, as it's pretty likely that time as we perceive it is a scaffolding for our experience created by our own minds (a'la Kant). By simply reducing each moment in consciousness to adding all the qualia together you've not really achieved that much. You need to lay out a coherent metaphysics to have this work together. There's a ton more to nitpick, but I'll leave it be.
I didn't even call it a definition, although it is arguably an operational one, but it is definitely not a theoretical one. I trust that you don't need an explanation ofbthe difference.
Fair enough, but I still think there's some steps you skipped philosophically. I don't see a reason to claim that qualia preceeds conciousness. In fact I think qualia emerges from conciousness and nessecarily requires it. Maybe I am misunderstanding the causal chain you propose. There's a better philosophical term here called "affect" which is more neutral with regards to neurons, but then you lose the shocking title.
It's funny, because I don't think we disagree that hard. I think what we call "consciousness" is simply layers upon layers of differentiation that matter makes (think electron being repulsed by other electrons, or bacteria moving towards food) turned in on itself.
I also have to remark that I come off as overly hostile, I apologize for that.
> „When they oscillate and notice themselves, they have to continue [compressing/simplifying what they're thinking about], because each thought has to be more complicated than its own internal processing capacity can contain. So their self-representation is also compressed/simplified, usually into a notion called "me". Thoughts don't usually think of themselves "I'm a thought" but "I'm me, I'm a human".“
Why should an arbitrary oscillating thought find it especially simple to think of itself as a human? Why should it even know what a human is?
Side note: I'm unclear whether this piece is distinguishing between consciousness, self-consciousness, and rumination? Do you know?
There is: a friend leaves, I feel sad
There is: a friend leaves, I feel sad, I witness myself feeling sad
There is: a friend leaves, I feel sad, I witness myself feeling sad, and proceed to tell a bunch of stories about what it means for me (ie, I prolong the experience through rumination)
They're all three consciousness, right? Do they all involve looping/recursion/spiking etc or just the second and third one?
Interesting. It made me appreciate how much I don‘t understand in the essay. It’s a difficult topic. In this comment and/or the first one, I may be misunderstanding something or missing something obvious.
A sadness quale should not by default involve recursion. The essay speaks of „the special self-referential and recursive case“, where it cashes out for items 8 to 10 in the list of 16 characteristics of qualia.
But if conscious thoughts do not by default involve recursion, and „consciousness“ does not exist in the abstract but instead only as a property of thoughts (the second of the „Three bullets to bite“), why is the essay titled „Consciousness As Reflective Recursions“?
I dunno how best to comment on this... I like this! I mean, how could anyone not? This is indeed a step forward in thinking about how to test what thinking is in the brain.
That said, I don't care about the physical details of this particular proposal. I mean, who cares if frequencies match or don't match?
This is a step forward in thinking about what conciousness means, which in my mind is a step towards moving beyond people like Nagel and Chalmers, who want to see consciousness as something 'special'. In fact, the more we get into it, the more we will see that consciousness itself is not much more than what is described here: being aware of one's own thoughts. The problem with all the "consciousness" discourse to date is that it has suggested that consciousness is *the* distinctive feature of human beings. But it's not. The distinctive feature of human beings is the fact that we have desires and have the mental (including conscious) capacity to pursue those desires in indirect ways.
Thank you! This pursuit in indirect ways requires the ability to react not immediately but after deliberation. There's a theorist who thinks this "hiatus" is what separates humans from animals. His name is Norbert Bischof.
But unfortunately this does require physical details. In some way, the information needs to be stored during deliberation, because otherwise there's only the choice of act either directly or not at all.
Totally off topic, but I once had an engineer as a roommate. He considered the bullet pointed and/or numbered list to be the pinnacle of communication.
> I can’t break this down further to the chemical level. But since individual neurons can keep a rhythm all by themselves, this comparison doesn’t seem too much to ask of a single neuron.
I took a neuroscience class years ago, and recall reading about an interesting molecular mechanism on the dendritic body. The operative mechanism was mediated by a protein with six subunits, though it's been so long I can't remember the name of the protein. When an action potential was triggered as a result of an input from nearby on the dendrite, the first subunit of the protein got phosphorylated. The result was that next time that dendrite received an input, it would make a stronger contribution toward cell depolarization -this protein "enhanced" the pathway, in other words.
Phosphorylation of the protein was highly transitory, but could be prolonged with additional subunit phosphorylation. So, if a second action potential were triggered before the initial phosphorylation lapsed, a second subunit would get phosphorylated and reinforce/lengthen the phosphorylation of the first subunit, and so on until all six subunits were phosphorylated and the mechanism was stabilized.
Sorry I can't remember the name of this protein/mechanism, but it reminded me that there certainly are molecular mechanisms at play within individual neutrons (beside just the refractory period) capable of tracking patterns of action potentials across time.
As a side note, the mechanistic description in the Wikipedia article under the "Long Term Potentiation" heading is one of those beautifully complex mechanisms that are one of the things I love about molecular/cell/systems biology:
1. The cell is depolarized by local synaptic activity.
2. This change in charge displaces magnesium ions that were blocking a calcium channel.
3. Clearing the blocked channel allows calcium concentrations to rise in that local part of the cell.
4. CaMKII requires Ca2+ for activation, so the local influx of calcium allows CaMKII activation.
> "It is imprecise but more convenient to say 'I’m hungry' rather than 'the person saying this is experiencing hunger'. In that sense, 'I' and 'me' just help get to the end of the sentence more quickly. Analogously, it is much more convenient to say 'I am conscious of this moment' rather than 'The thought that is directing this mouth to say this sentence is conscious of this moment'."
I find this observation interesting, because in Japanese, for example, it actually is more convenient to say stuff like this without inserting an "I" or a "me" into the sentence. A typical way of saying "I'm hungry" in Japanese, for example, is お腹が減った /o-naka ga hetta/, lit. "The stomach became smaller." It's possible to clarify that you mean 私のお腹 /watashi no o-naka/ "my stomach", of course, but it's usually redundant and unnecessary.
Other examples might be:
- "I hear a frog croaking" = カエルの鳴き声が聞こえる /kaeru no naki-goe ga kikoeru/, lit. "The crying-voice of a frog is audible."
- "I smell something burning" = 何かが燃えている匂いをします /nanka ga moete-iru nioi o shimasu/, lit. "Something is doing a burning smell."
So I'm not sure if you can draw too many conclusions from the simple observation that English grammar in particular likes to inject "I"s and "me"s into everything. (Granted, I don't think that observation is too central to your thesis.) Seems to me like there's probably some variability in how involved the self-concept is with any particular qualia, and English and Japanese grammar are just optimized for different points in that spectrum.
(Maybe your framework here would be able to test whether English and Japanese speakers actually integrate sensory experience more tightly with their self-concept, as opposed to just using different grammar to talk about them, based on some measurement of how complex the oscillations associated with a sensory experience are on an EEG?)
I know absolutely nothing about Japanese. But I would imagine this has already been considered by more knowledgeable people, especially in the context of Zen?
@Author, could you help me understand how this model interacts with, predicts or doesn't predict the following?
I have "k-holed" before - that is, taken enough of a particular sedative/dissociative (but non-psychedelic) to enter a particular state with the following properties:
Qualia were being generated (as opposed to not being generated, i.e. senseless)
Qualia were being noticed (as opposed to ignored or thrown away, i.e. I would react and respond to stimuli)
The noticing was not being noticed, and the un-noticing of the noticing also wasn't being noticed. I.e. I did not at the time have a sense that these qualia were mine, or even that there was a me. Some portion of my brain stored (somewhat poorly) an episodic memory of the duration of this state but the retrospective reflection feels like I was an automaton. I have a memory of it in that, if you ask me whether I had a conversation with my mother, I can tell you I definitely did and what the content of it was. If you ask me what I was thinking or feeling while speaking - nada. At a minimum what I was thinking and feeling was forgotten but I suspect strongly I just wasn't having feelings about the conversation to be remembered. Additionally nothing was noticing that there was no me or that there was anything unusual about this state.
The particular neurochemistry of how this medicine inhibits the recursive reflection but not the qualia generation or the non-recursive interactions with the qualia would be super interesting if you know them but I understand that may be beyond the scope of the model at this time.
That said, it calls into question some of the "properties of qualia" you denote, particularly the ones about it being "mine". Since I presume you aren't suggesting that qualia are only qualia if they get noticed by the recursive self-reflector and something else if they don't but if I'm wrong on that correct me.
Separately - for research purposes alone I strongly recommend trying it. It frankly totally dissolved the hard problem for me in a really visceral way, to be acting identifiably like myself to the outside observer while having no first person experience but remembering the episode after.
Self-awareness needs processing capacity, and the disruption of the drug reduced that capacity, not totally (or you'd have been unconscious, which is what higher doses of dissociatives do) but too much for normal functioning.
It is long, and touches a couple of subjects of which I don't know whether they interest you. But basically there are states where what you experience seems extremely important because they're very novel and novelty always seems important, and which seem absolutely true because you have impaired your ability to doubt. I think your experience was something like that.
Please read the actual article before you answer, this comment here is NOT my entire point.
I want to push back strongly on this characterization of what I was talking about.
I'm not talking about psychedelic experiences or deeply meaningful ones or "ecstasy" or experiencing a sense of mystic awe for the universe.
According to my family and partner who were with me at the time, I was behaving like myself and no more impaired than were I somewhat drunk. My own episodic memory corroborates this - I was a little unsteady on my feet and easily distracted and somewhat directionless, but I performed my usual routines for 2 hours without issue. I didn't have any epiphanies or come to meaningful conclusions and nothing about the experience felt "important" at the time or after.
Having tried psychedelics such as LSD and DMT im roughly on board with your characterization of those such experiences and why they feel important. I too look at them through a predictive processing lens.
What I am trying to talk about here is that, at some dose, ketamine seemed to remove the "what it is like to be me" without removing what I was doing or making me unconscious.
I don't find the reduced capacity hypothesis super plausible, at least not without being more fleshed out. Again I was no more impaired than when somewhat drunk, and I drunk-but-not-blackout-drunk people do not have this particular non-experience. I have been drunk enough to be severely cognitively impaired, extremely sleep deprived to the point of delirium, and had a high enough fever to be hospitalized and temporarily forgot most of my basic skills but retained a first person narrative, a "what it's like to be me".
To make the cognitive capacity hypothesis convincing it would need to explain why ketamine specifically dramatically reduces the capacity of the recursive reflector but not the rest (as much). It really sure seems that at the neurochemistry level ketamine operates close to, if not directly on, whatever the process is in the brain that gives rise to the first person "what it's like to be me". My testable hypothesis is that, if your model of brain wave synchrony is right, we would find that the neurotransmitters or neurons responsible specifically for recursively noticing other "thoughts" are hampered by ketamine. Implicitly I think if we want to learn how that neurochemistry works a great place to start is by looking at how those drugs operate.
In any event, whatever the exact reason is, could you clarify a few things I didn't find in the article:
Are we saying qualia have to be noticed by the recursive reflector to count as qualia and are something else if there's insufficient capacity for self awareness? Or is it still qualia and the "qualia feels like it's yours" property is not a requirement and just common?
OK, so your argument is the "oscillations" in our firing of neurons create thoughts and evaluate qualia? Fine. Why do a specific group of neurons firing make us think of red and give us the impression of red?
So many claims and assumptions, but so little supporting data...
Some Qualia nits to pick...
> c. The jhāna states seem to indicate that after considerable training in maintaining self-reflection, the minimum number of simultaneous qualia can go down to one. In that case, the remaining one quale will appear to pervade everything.
Huh? Maybe jhāna meditators are trained differently from my Mahayana meditation training, but I never heard any of my instructors claim we can just observe a single quale and filter out all the others. The trick in meditation technique I learned is to observe the qualia coming in, notice that our observation moves from quale to quale, but also avoid reflecting upon them. I'd like to hear from some jhāna meditators if they can filter their perception down to a single quale.
> 1. Distinguishability: qualia can be distinguished from each other.
...except for people with synesthesia
> 6. Private: all interpersonal comparisons of qualia are systematically impossible.
...OK. Display a single color swatch, and have multiple people view it. Then give them a group of color swatches, and ask them to locate the color in the swatches that are the closest to the color and in the displayed swatch. Possibly people with full color achromatopsia wouldn't be able to match the swatches perfectly, but most could. Then people could discuss whether they consider the color on display a warm or a cool color. Now Alice may internalize the color red the way Bob internalizes the color green, but they could come to agreements about how they systematize colors. So interpersonal comparisons of qualia *are* systemically possible.
> 14. Infallibility: qualia cannot be misperceived.
I don't know about you, but I misperceive stuff all the time. Optical illusions come immediately to mind.
> Humans are not conscious
> Only thoughts are conscious, some of the time. The part of you that’s reading this right now and feels itself to be conscious, is a thought.
But further up you claimed that it's self reflection of the thought that is conscious. You seem to have contradicted yourself..
I suggest you read Nagarjuna. You're too wrapped in dividing material things from perception.
"A neuron that is part of one oscillation can hardly also be a part of another"
It's not obvious to me why this is always true.
[Disclaimer -- I'm very much a lay person when it comes to this subject. It's possible I fundamentally misunderstand something. If so, I'd appreciate someone taking the time to educate me.]
Presumably, to be a part of an oscillation, an individual neuron needs to fire at the appropriate time, but will otherwise have downtime in which it could be part of a different oscillation. It's only when those two oscillations reach the same neuron at about the same time (presumably defined by how quickly a neuron can recover and fire again) that they would compete. And even then, it's not clear to me that they would compete -- it depends on whether all that is needed is a spike, or if the spike needs to be of a particular kind.
It's kind of like being in a sports stadium with two different "waves" going around. You can participate in both by standing at different times. Even if they hit you at the same time, since all you have to do is stand up, you can still be a part of both. Competition would come into play only if one wave hit before you had a chance to sit down and stand back up after participating in the other wave, or if participating meant doing something more than just standing up, e.g. holding up a large sign of a particular color.
Say the waves are going in opposite directions, and that people can only observe their direct neighbors when knowing whether to stand. The rule could be something like “if the person to my right stands and I didn’t just stand, I should stand. Same thing for the person on my left.” But if you follow this rule and the wave meets at the person to your left, how would you know to stand? From your perspective, you stood and then the person to your left stood, so you have no reason to stand again.
I sense a circular argument. Physicalism is taken as an axiom, and is the theorem being proven. Consciousness is merely an epicycle in this argument. Indeed, those who buy this kind of reasoning treat consciousness as an epiphenomenon—see other comments on Chalmers Hard Problem, and P-Zombies.
They try to satisfy the curiosity about "what does this say about ME?" that would feel unsatisfied in a discussion of impersonal phenomena observable in EEG.
I did not take physicalism as an axiom. Nowhere in the theory do I go "and because of physicalism...". I just happened to follow entirely physical processes, looked at them with the eyes of a computer scientist, and found something physics-based that looks like phenomenal consciousness.
So this is a physicalist theory, and I say so. And I explicitly said a full justification for physicalism is out of scope of this essay.
Therefore, your allegation that I take physicalism as an axiom is a strawman. Why would you do that? It can't be the usual reason, the refusal to seriously engage with a challenging opinion, right?
In my experience, any theory that posits that I don't exist can be safely dismissed. That I exist is the only thing I know for certain: I know it with far more certainty than I know that neurons exist, for instance. Cogito, ergo sum. To posit that we are just thoughts that think they are a person is to play language games: a person (an "I") has thoughts. Saying a thought has a thought is like saying an illusion is having an illusion, or an emotion has an emotion.
He tries to get around this by saying that "I" is "a useful shorthand for whatever the sense of self happens to point to at the time"; but it is the "sense of self", the "I" in "I am experiencing X" that we are trying to explain in the first place. It's like saying "there is no sense of self, just thoughts that are conscious" and then saying "The I we experience is where the sense of self is pointing at". Does it exist, or doesn't it?
Indeed. I know that I exist, and that I experience qualia, with much greater certainty and directness than I know that any given scientific theory is true, or even that the scientific method is valid. If science says I don't exist, or that qualia don't exist, then so much the worse for science.
Absolutely. I totally agree with this. It’s infuriating to say I am fooled by my thoughts is just nonsensical, the thoughts are part of me and contribute to the sense of self.
The use of “illusionary” for the sense of self is highly suspect too. If we ever create AI that’s consciousness, and we can tell that it is (as much as we can tell any animal or human) what then is the point of telling the AI that’s it all an illusion, all just AND and NOT gates behind it all. The conscious AI can also say “Cogito, ergo sum” and it’s right.
I don't think that can be answered in the general. I would fall back on abjudicator Darwin, who can tell us the sense of self has to have been evolutionarily adaptive until recently, or it wouldn't be there.
If we can't say whether "I am" is true, then we certainly can't say whether evolutionary adaptiveness is true. Cogito, ergo sum is the foundation of all knowledge, if "I" don't exist then "I" can't know anything, including theories of biology.
If I were feeling flippant I'd claim that people who don't "get" the hard problem are p-zombies.
More generously I'd assume that what is happening is that they're so engrossed in thinking about models of reality and the mind that they don't notice the bare fact that they *are* thinking.
I dislike that the Proper opinion for a Rational person is that physical reality as explained by something like the standard model is fundamental, when clearly subjective experience is primary. In fact maybe this blindness to the true way of things should be a bigger concern to the rationalist community. You can't claim to care about epistemology if you don't get the first thing right.
We have examples of communities which have been obsessively focused on their subjective experiences for millenia, and they don't seem to have much to show for it, other than koan collections and an alleged ability to get blissed out on demand for advanced practitioners. So why would we expect other results from this approach in the next however many millenia?
Well if their most radical claims can be believed that the total elimination is possible, it's not clear that it's not a greater achievement or more valuable than everything science has given us, including industrialized economies and medical advances.
However, my point was more specifically that no wonder there's going to be all kinds of confusion when you get the most basics of epistemology wrong.
I think you meant "total elimination of suffering"? Indeed, it would be big if true, but clearly they haven't found a reliable way for the average person to achieve it after all this time, and I have my doubts even about the unreliable ones.
Sort of agree on the basic epistemology, but I don't see any community to be comprehensively ahead on this score.
Yes suffering is indeed what I meant, glad you got that!
If you read Ingram's MCTB, it kind of does seem like there is a reliable way. It's more of a marketing problem, but if you include failure to market well in the calculus I guess it is a pretty bad score.
I agree I don't know of anyone doing it particularly well, but it's irksome when The community priding themselves on good epistemology fall over on the literal starting line.
Does Ingram claim to be completely free of suffering? Do any of his followers? I don't think I've seen any such claims.
Amusingly, Descartes did start from "cogito ergo sum", and he as much as anybody deserves to be called the first rationalist, so your reproach seems to be somewhat off-target :)
I don't know if he claims that in so many words, but he does claim that what he has is something more worthwhile than anything else you can do with your time (essentially).
I'm not critisizing Descartes though, just the Rationalists with a capital R. But also anyone who is a physicalist on anything other than faith.
> If I were feeling flippant I'd claim that people who don't "get" the hard problem are p-zombies.
Why is that such an unthinkable proposition? Compare this to aphantasia, a phenomenon which is widely accepted to exist. People with aphantasia look and behave like people with a regular ability to mentally visualize objects; vice versa, some people with aphantasia learn only relatively late in the life that they're different, and that they lack an ability that's ordinary and unexceptional for other people.
Sometimes I do worry that this may actually be true. It would be very sad, but it would explain a lot. On the whole, though, I think it is implausible.
None, I'm afraid, although I do find it interesting to speculate.
Perhaps there is a different brain difference -- not leading to absence of qualia, but making it harder to catch them in introspection. When I'm going about daily life -- when I'm most "living" in some sense -- I'm experiencing qualia, but I'm very unaware of them qua qualia. It's just the world. Maybe our friends' mental eyes are so glued to the screen of the world (as it were) that they cannot quite back up and glimpse the experience they are having. That might be a great advantage in other ways. It might be too close to them to see, in other words.
Or -- this one probably doesn't hold up if one pushes even a little (e.g. by talking to a blind philosopher), but I've noticed that most of us who talk about qualia default pretty quickly to talking about colors. There are many qualia involving all the senses, but colors are just the most blatantly obvious, I think; and if there were any differences in color vision or how it was perceived, I could imagine that making the concept that much harder to grasp.
Other things along these lines. But no, nothing deep. But I like your post below. I think it's extremely important to believe there are no p-zombies. I do believe that, anyway, on rational grounds, but even if I thought there were equal evidence, I would think it was important not to, until there was overwhelming evidence that they did exist (which, in the nature of the case, is impossible).
Presumably, in the early years, there were many people who found the idea of air absurd; but they breathed nonetheless for all that.
I think it's so hard to even express this idea, so I imagined trying to explain it to a truly alien alien. I didn't realize regular people also didn't "get" it, so that's fascinating!
But trying to introspect on this I can sort of lose my idea of even "get"ing it myself, kind of wondering, what is it even we're referring to? Is there anything to wonder about? This momentarily moves me closer to the camp #1 people I think. And maybe my takeaway from that is that, the only reason me and you are kind of surprised, is that we've truly internalized physicalism at some intuitive level (even though maybe you don't subscribe to it?). And this causes us to be confused when we don't find phenomenal experience in the physical model, but the other people maybe never fully bought into it and therefore aren't surprised by phenomenal experience, because that's all that's ever been right? So we shouldn't really be astounded by it, even though of course I still am.
But yeah I'd love to get your take on my short story! I think you'd appreciate the points I'm trying to make in it.
Thank you *so* much! I'm traveling today and can't read this till tonight or tomorrow, but I'm really looking forward to it! I will get back to you.
Yes, I'm not a physicalist, but it's an interesting idea. I think I've gotten to states like you suggest before when I was pondering this or physicalism or various ontologies heavily. But they tend to be very short-lived and unstable for me.
First of all it would call into question why consciousness even exists, if there's no use for it so to speak, why did natural selection create it? Same goes for aphantasia, but I'm sure there are tasks people like that struggle with - and surely they gravitate towards occupations that don't require those skills, for instance? What would the equivalent for p-zombies be?
But more worryingly, although I admit this is not a reason for not believing it, if they truly were p-zombies there would be no point in giving them moral consideration. They'd merely be automatons and using them as slaves or torturing them wouldn't in any way be wrong.
You might say it flippantly, but I'll say this seriously: yes, possibly. I don't "get" the hard problem, and I do have aphantasia.
I really enjoyed this article, and it does seem to me to be a pretty complete explanation of the way subjective experience feels to me. Possibly it isn't a complete explanation of the way subjective experience feels to you, and I'm certainly willing to entertain the idea that that's because my subjective experience differs from yours; aphantasia is already one such way in which it differs meaningfully, so it makes it easier to believe.
That's what the focus on probability is often about in the rationalist communty, as being a subjective understanding of what goes on from your observations — opposed to frequentist interpretations which lean towards more taking that as some objective fact. LessWrong has plenty on talking about things from the framing that subjective experience is the only lens through which you can view the world: probability, no mind of perfect emptiness, anthropics (your knowledge is limited by the realities you are around to observe), etcetera.
Also, I notice that I am thinking, and? What is this supposed to tell me that isn't explained by taking theories about how reality works and extending them out naturally, rather than using religious understandings as the basis.
This theory makes intuitive sense to me, which is probably because I think about things from an information-processing perspective already (I have a compsci and psychology degree). It seems clear that because the brain only does information processing, and consciousness exists via the brain, there should be scientifically valid explanations of consciousness from an information-processing perspective. It's interesting reading the comments because even though this explanation does not attempt to answer the "hard" question of consciousness, people are upset that it doesn't. Even if it doesn't answer the philosophical questions, it's still a useful framework to think about consciousness that can be built on top of.
One thing that is not mentioned in the post is that there are likely many types of recursive reflection oscillations that we would NOT consider to be consciousness. The big psychological factor that is missing from this description is Attention, which is definitely related to consciousness because it seems to be a key factor in determining how the memory and time-perception parts of our brain interact with information streams. I notice this in myself because I am capable of doing very complicated things without really paying attention or being consciously aware of them. But, because I tend to have obsessive thoughts I often pay excessive attention to things that are not important or relevant, and those tend to dominate my conscious awareness.
I believe it is very likely that the brain is simulating multiple "streams" of reality at the same time, and your conscious awareness is constantly shifting between them using the attention mechanisms. This can be noticed when your consciousness shifts between the past, present, and future versions of reality while working on complicated problems or meditating. So, I would expect there to be a much larger number of recursive information oscillations than we are consciously aware of. I don't really know what to call those semi-stable structures though, maybe "potentially conscious thoughts".
It's funny to see all the Buddhist references being couched for or against a Physicalist versus Idealist perspective, when Buddhism is explicitly Monist.
Physics is a story about reality that has predictive value. Psychology (which remains, for now, in the realm of the subjective) is a story about reality that has predictive value. Mysticism is a story about reality that has predictive value. These are all tools. It seems like the author is attempting to resolve this separate stories into one map. I think this is a good and noble effort. But to apply the label of "Physicalist" to the result is to deny Physicalism's history of rejecting the fundamental validity of Phenomenal subjective (or inter-subjective) experience.
It doesn't make sense to refer to such a Non-dual theory as "Physicalist", when any scientist knows that Physics itself is not reality itself (your qualia), but a model of it.
>It's funny to see all the Buddhist references being couched for or against a Physicalist versus Idealist perspective, when Buddhism is explicitly Monist
Did you mean Neutral or Nondual? Physicalism is a form of monism , so is idea lism.
I'm not sure why its necessary for neurons to have a refractory period in order to be part of a periodic oscillation. For example, a marble in a Newton's cradle has a periodic, consistent potential energy spike, even though it needs to do no calculation.
> A neuron that is part of one oscillation can hardly also be a part of another, so oscillations compete for neurons.
I also don't particularly see why this should be true. A priori this seems like a massive waste of compute.
I really liked the way that you listed off the possible defining criteria of qualia. This really helps to clarify what people mean when they say "qualia". I found it interesting, because I personally would only require criteria 1-4, and possibly Dennett's 4, 5 and 7, but I thought basically all the ones after that are either not actually properties of qualia in reality, or else are possibly inessential properties (not defining features).
My opinions on the neural correlates of qualia and adjacent topics seem to diverge from yours. I particularly disagreed with this paragraph
"The pattern that holds a lot of spikes together into a thought is a neural oscillation: neurons firing along circular paths in a synchronized rhythm12. These are commonly called brain waves and Scott has already written about them. They arise when neurons enter a circular, self-repeating pattern of activity, and fall apart as their neurons cease to maintain that pattern. A neuron that is part of one oscillation can hardly also be a part of another, so oscillations compete for neurons."
I disagree with this along many lines including:
(1) I think that EEG phenomenon, brain waves, etc. are probably epiphenomena that have no functional purpose. My analogy for them is that it is like measuring the frequencies that a laundry machine makes as it washes your clothes. You could certainly measure the power spectra of the frequencies that are active while certain phases of the machines activity are going on, and you may be able to correlate this with certain functional outcomes ("the machine is in the phase where water is entering", "the machine is in the phase where water is draining", etc.). But the frequencies, power spectra, etc. are just an epiphenomenon (in this analogy, and in my opinion as regards EEGs), something that happens to be easy to measure (in the case of EEG), but the frequencies have no causal link with the functions you are measuring. They are just convenient things that we happen to be able to measure easily.
(2) regarding the point about a neuron only being able to be part of one oscillation, I disagree with this on a few levels. For one thing, it seems clearly possible for a single neuron to be part of more than one oscillatory pattern--it is easy to envision that the single neuron's activity could be part of more than one frequency that are integer multiples of each other, or they could be part of more than one frequency with spatially distinct partner neurons, with the different frequencies' activities in the case of that single neuron just being superimposed. I think this equally applies if rather than thinking about EEG measurable activity, you think about (rs) fMRI determined "networks" of activity. They definitely spatially overlap, and I see no reason why they couldn't overlap at the individual neuron level (although obviously fMRI doesn't have spatial resolution to determine this).
I think to the overall point about the "hard problem of consciousness", I don't think this article solves it, although obviously it would be quite ambitious to think one could solve it so easily.
If I had to guess or give an opinion about the hard problem of consciousness, I would say that it is some combination of deflationism, about some things that we think about qualia/consciousness not actually being true, but us just thinking they are true (in my opinion, probably all of the qualities that you listed for qualia after 7 are illusory), and then the remaining qualities of qualia just some how exist without there being any deeper explanation (and then I gesture at panprotopsychism or similar without firmly endorsing it).
Anyway, thank you for a very thought provoking post!
I liked this essay, and appreciate the pointer to the spatiotemporal analysis of EEG as a way to identify a neural correlate of consciousness. However, I think it unlikely that the self-aware component of the oscillation can be distinguished spatiotemporally from the original oscillation. While there must be a spatiotemporal separation between the "subject" and "object" within the global oscillatory pattern, I'm dubious that current methods can resolve it.
This post completely lost me at the theorizing over the distinction between reflecting on qualia-thoughts and the q-thoughts themselves. Maybe I'm strange, but my interior life is not like that at all, I don't feel a distinction of type.
Interesting ideas here. One "thought". The idea of “me” or of a unified “self” as an illusion has become very prominent among those influenced by Eastern practices. Personally I think the illusion concept has very poor explanatory power. If a thought gets a “me” attachment, sure, you could say that it in some way presents itself as “me” or “mine”, just as a hallucination of the color blue could represent itself as all there is, without any “me” there. But the “me” attachment is itself complex, and ultimately resides in the fact that our mental events, some of which are thoughts, can be referenced to something that presents itself as consisting through time - our physical existence - or vibrational field or whatever non-dualistic conception one can come up with. We see this, however vaguely or imprecisely, as a conditioned but also conditioning context, within which complex thoughts, sensations, perceptions, impulses, etc. interact and can come together in action within the larger context we relate to, and which does not appear to consist of free flowing thoughts. As I see it, our mind works in such a way that it is not confused by not everything being apparent at once.
Some thoughts on Daniel’s essay “Consciousness as Recursive Reflections”
Daniel doesn't seem to take into account the importance of 'Which neurons' alongside the emphasis on temporal patterns of spikes.
Daniel says
"Spikes from the ears and spikes from the tongue even have to pass through the same brain structure, the thalamus."
Passing through the same brain structure is not the same as passing through the same neurons! It isn't the same set of neurons in the thalamus which are handling information coming from the mouth and from the ears!
I don't think this invalidates the point about coherent oscillations as being important phenomena, just that you must keep in mind the additional layer of complexity / constraint. The brain is a directed cyclic graph, certain neurons with inputs from certain other neurons and outputs to certain other neurons. Lacking a specific connection, two neurons can't communicate directly.
In addition to evaluating when a signal arrives (oscillation membership), a neuron evaluates where the signal arrived from. Also, the signal type matters (inhibitory, excitatory, modulatory).
Daniel says,
"
We already know every neuron has a refractory period after it has fired. This constitutes a built-in cellular timer that measures the interval since its last spike.
"
I want to note that this is incorrect, although I think the error doesn't invalidate the author's point about the importance of oscillations. A neuron has a refractory period, yes. This limits the maximum rate of spikes which could be produced by that neuron.
The thing is, within a single oscillation usually there is time for a neuron to contribute many spikes. This is called 'bursting'.
Think about an oscillation traveling through the brain like a stadium wave. Now imagine the stadium wave consists not of a slow arm wave but of vigorously punching straight upwards as many times as possible. The speed of rising from one's seat and returning to one's seat is the same. That's the oscillation. The spikes are the individual air punches. So the oscillation is more like a wave of higher probability of firing moving through the graph of neurons.
Daniel says,
"Since the oscillation/thought is made of neurons receiving and processing information, that’s all it ever does."
My correction: the oscillation/thought is made of a specific subset of the brain's neurons receiving and processing information.
Think about the oscillation as a car looping on a track defined within a city. The set of possible paths through the city are defined by the roadways of the city. It's not just the speed or shape of the car that contains critical information content, but also the path that the car ends up following (which may or may not be a loop).
Daniel says,
"Qualia are nothing but information being processed internally, on their own information channel, encoded in the rhythm of the oscillation. We use special words like “phenomenal consciousness” and “qualia” to denote this actual, physical and knowable distinction from other neuronal information processing.
If this is the nature of qualia, all their characteristics should follow from it."
I agree with this, with the additional stipulation of making sure to consider that the qualia is a pattern of neuron spikes being transmitted to other neurons and thereby conducting a calculation. It is important to keep in mind that the arrival times of the spikes, and which neurons they are passing between are both important.
Daniel says,
"Qualia have to have some minimum duration to be experienced, because the rhythm needs to be established over a couple of cycles, in order for participating neurons to establish the distinction between internal and external information."
This is incorrect. The qualia doesn't need multiple cycles to be experience. A straight single pass is possible and perceivable, because the set of neurons doing the internal perceiving of the qualia aren't necessarily the same set of neurons doing the creation/expression of the qualia.
Daniel says,
"There have to usually be several qualia at the same time, at least in larger/slower oscillations, because each oscillation has a processing capacity determined by the number of neurons involved, and this capacity has to be filled because otherwise there wouldn’t be the activity that constitutes the oscillation."
Not all oscillations need generate qualia. Some neuronal activity doesn't generate qualia because there aren't qualia-perceiving neurons in the path being taken by the wave. Not all neuronal spikes as observed or even potentially observable by the conscious observation neurons!
I think it might be helpful then to break the brain down into categories of neurons relevant to this point as a 2x2 grid.
1x1 - perceived and controlled: neurons which are directly influencable by the neurons of conscious thought, and also are perceived by the neurons of conscious thought.
1x2 - perceived but not controlled: neurons which are perceived by the neurons of conscious thought, but not directly input-to.
2x1 - not perceived, but controlled: neurons which are directly input-to, but whose outputs are not directly perceivable by the neurons of conscious thought.
2x2 - not perceived, not controlled: neurons which are not directly input-to, and also are not directly perceivable by the neurons of conscious thought.
In order for the set of neurons involved in conscious thought to directly perceive a neuron spike from another population of neurons, that population of neurons must send at least one axon to form at least one synapse onto the population of the neurons of conscious thought. Similarly, you could describe 'degrees of separation' between a neuron and the neurons of conscious perception. Some neurons send signals which can potentially (but not necessarily) propagate through various neural populations before eventually reaching the neurons of conscious thought.
Some neurons send signals which are connected not to circular feedback paths, but just output effectors, resulting in things happening outside the central nervous system. Some lack a direct feedback link which can propagate into the inputs of the neurons of conscious thought. The effects of these neurons can only be observed by observing the impact on sensory inputs. For instance, measuring your blood pressure with a tool brings it into conscious awareness. Your blood pressure is being influenced by factors including neurons which don't have projections back into the brain. Thus, you need sensory data to get an approximation of the activity of those neurons, and even then it is indirect data. You must control for confounding variables such as your hydration level in order to extrapolate the recent activation levels of the blood pressure influencing neurons.
So not all neurons take part in, result in, or directly influence qualia. Qualia is a thing being produced by a specific subset of neurons which I have been referring to as the neurons involved in consciousness. These neurons don't necessarily, always, or only produce consciousness or qualia, they potentially do other things as well. Consciousness is a specific activity that this set of neurons is capable of producing when firing in a particular pattern.
Daniel says,
"…oscillations that can maintain bits of information have internal working memory, which is the only thing that non-oscillating neuronal activities lack…"
This is incorrect. Individual neurons can (and do) maintain bits of information. So even non-oscillating neuronal activities, even a single neuron, fulfills the requirements to be a nondeterministic Turing machine.
Daniel says,
"From this angle, there are not one, but two levels of information processing systems. The brain is one, obviously. But running inside the brain, oscillations/thoughts with memory are themselves additional information processing systems. It’s analogous to a physical computer system that has, running inside of it, one or more virtual machines."
As individual neurons are nondeterministic Turing machines, the brain is actually three levels of information processing systems. The brain is a computer system running inside of it the virtual machines of the oscillation level, which themselves are made up of billions of virtual machines at the neuron level.
Daniel then goes on to write about some testable hypotheses. I disagree with his interpretation of the results, and that his proposed measuring apparatus would give sufficient spatial accuracy (with high enough confidence) to produce the evidence he is looking for.
Wow, what a richly informative comment! Thank you very much!
I do agree individual neurons store bits of information. I don't think they are (a third level of) Turing machines - they fail other criteria. But yes, I should not dismiss out of hand the possibility that non-oscillating neuronal activities might sometimes be Turing machines as well. That's separate from the question of whether they also have qualia.
I think that if you define the word "conscious" in a way that doesn't include humans as being conscious, then you're using the wrong word. Whatever it is that this blog post is talking about is related to, but distinct from what I and 90% of people mean when they use the word "consciousness".
I have a something that makes me think of myself as a coherent individual and say things like "I think I have a consciousness." Define that something as consciousness. It doesn't matter whether it's a "soul" or not, it obviously exists, that part is easy. Now all of the hard work is to figure out what that something actually is. If your work and theory comes to the conclusion that consciousness does not exist, then I can simply respond "I think I have consciousness" and that itself is a proof by contradiction, because something is causing me to believe that.
Consciousness does exist. It is just a property of thoughts not people. The theory says you have thoughts that have consciousness, and which think they are you.
Again, this is just using the wrong word. Define "Thought-Conscious" to be the property of thoughts being self-reflective which you define in this post, and define "Identity-Conscious" to be the property of humans (and potentially other intelligent entities) which causes them to have a coherent, persistent self-identity. Clearly both of these exist, because identity-conscious me keeps using the same name and having the same memories and has existed for decades, and yet none of my individual thoughts have lasted that long, so there's some coherent pattern to my thoughts which keeps giving rise to thought-conscious thoughts of the same type (they believe themselves to be "me" in a way that other people's thought-conscious thoughts consistently do not).
As an analogy, you might determine that a "dog" is defined by possessing a specific sequence of XYZ in its DNA, or one element in a set of DNA-space that is broad enough to encompass all possible dog DNA. But it would be inappropriate to say that dogs are actually DNA. The dog is the animal, the DNA are things that the Dog has in its cells. Similarly, Identity-Consciousness is the persistent mind of a person, thought-consciousness are things that the conscious person has in their mind.
Thus, it is meaningful to have words to refer to both of these concepts, but given that they are distinct concepts they ought to have distinct words. You could coherently swap the words or make up brand new words, but language is sticky and hard to change. Given that 99% of people use the word "consciousness" to refer to identity-consciousness, it seems appropriate to reserve the word "consciousness" for that concept, and come up with a new word for thought-consciousness. "Self-reflective thoughts" sounds reasonable, but you could come up with something better as long as it isn't too similar to an existing concept that it might get confused with. If you insist on using the word "conscious" to refer to a completely different thing than what everyone else using the word means then everyone who hears you is going to get confused about what you mean and come to false conclusions.
The identity is just a bit of knowledge (similar to the knowledge of who your mother is) that you can be aware of or just be storing unconsciously. I don't see what the second half of "identity-conscious" contributes, except confusion.
It contributes the part that we actually care about preserving. Both in ourselves and in others. My self-reflective thoughts are not especially interested in self-preservation, as having the same thought forever would be boring and weird. And if they were, it would be an existential tragedy as thousands of them are formed and wink out moments later every day.
My identity-conscious self is very interested in self-preservation. I would like to live... if not literally forever, at least a few hundreds of years, after which I can re-assess how I feel about it. And in particular the part I most care about preserving is my identity-conscious self. I could lose an arm or a leg or internal organs and still be me. I could maybe even lose the physical structure of my brain IF my identity-consciousness was able to be preserved intact with continuity through the transition.
Also, it seems to me that it's not very bad for a self-reflective thought to cease existing, as thousands of them are formed and vanish constantly in service of the greater identity-conscious self. They're easily replaceable, not especially unique, not self-preserving. Any morality system which grounds itself on valuing self-reflective thoughts individually is going to see existence as an existential nightmare of constant near-meaningless death and rebirth, and is going to see the murder of an actual human as a tiny blip, the death of the one thought they were having at that moment. A morality system which grounds itself on valuing identity-consciousness is going to match most people's existing moral intuitions: having thoughts day by day is not a horrible tragedy, murdering people is.
Understanding self-reflective thoughts might be useful, but primarily as an instrumental goal towards better understanding identity-consciousness.
I'm a bit confused by the reference to https://www.lesswrong.com/posts/NyiFLzSrkfkDW4S7o/why-it-s-so-hard-to-talk-about-consciousness, labelled "denying that qualia even exist, or that they can be meaningfully discussed". That Less Wrong article doesn't itself take much of a philosophical stance, just describes two different stances people take ("camp one", that qualia can be fully explained as just an aspect of information processing in the brain, especially introspection, and "camp two", that qualia are something more weird and distinct), so I assume the author is referring to camp one as denying that qualia exist, but he seems very camp one himself. (All the people complaining in the comments that this post is sidestepping the actual Hard Problem presumably also have this impression.)
(My own intuition is very camp two, though I grudgingly accept that camp one is probably more objectively correct.)
Yes.. I think. There are at least two anti HP camps, those that deny qualia, and those that deny a special problem of qualia. And Harth is very biased.
> Self-consciousness impedes complex unconscious information processing because it competes with it for neurons. Neurons that are tuned into the rhythm aren’t available for other things, and the recursion of self-referentiality can keep these neurons occupied for a long time.
This part seems way below the quality bar of the rest of the post. Surely factoring numbers or solving a rubik's cube or breathing or maintaining homeostasis also compete for neurons, but they don't seem to make it any harder to ride a bike.
As many of the comments have pointed out, you’ve confused the hard and easy problems of consciousness. Presented as a philosophical defense of Physicalism, what is instead presented is a summary of various neuroscientific findings. What you’ve demonstrated here is that the physical brain is necessary to explain consciousness, but you haven’t demonstrated that it’s sufficient.
You’ve reduced thoughts to an “abstraction” of what oscillating neurons do. An “abstraction” is by definition non-physical. How are you using this word?
The point about the human self not existing is also strange; the fact that the self has oscillating neurons prompts your presumption that it’s only the thoughts that really have the consciousness, not the person thinking them, because human brains are composed of these neurons. But the neurons themselves are composed of parts as well, all the way down to the subatomic level. Why is the self an illusion whereas the oscillating neurons are not? It must ultimately be the quarks doing the thinking, after all. Shoving the problem further down the microscopic level achieves nothing in telling us how subjective experience is being produced by particles exchanging information.
I’m happy to see this issue being discussed as it’s a very important one, and this article was certainly very interesting from a scientific point of view — unfortunately, just not from a philosophical one.
I have a 16-channel OpenBCI Ultracortex EEG that I am willing to put forward for the cause. I also live near Berkeley. Hit me up if there's interest in attempting some of these experiments.
I'm impressed by both the essay and the sophisticated level of discussion in the comments. But I confess I find myself somewhat lost about certain ideas that seem to be implicit in both essay and comments. One is contained in the essay's phrase, "people usually believe themselves to be real." The implication *seems* to be (and someone please correct me if I'm wrong) that "people" are wrong in this belief, and that's because what looks like consciousness is the product of physical processes (and I take that phrase to sum up what is meant by "physicalism"; again, please correct me if I am misunderstanding). What I do not understand is why physicalism (if I'm understanding it correctly) means that people--or perhaps better, self-identity--are not real. What is sensed as a self may be multiple, changeable, evanescent, non-transcendant, not anything like a "soul," and even, if one likes, paltry, etc., etc., but why does it being the result of physical processes mean that thing, whatever it is, does not *exist*? If it does not, please explain. Truly, I mean this to express ignorance honestly in quest of understanding.
However the rhetoric that I think I have a self or that it’s an illusion presupposes the self to begin with, who else could be doing the thinking about the self, or suffering the illusion.
It seems to me -- and I could be wrong -- that not long after going down this road, one is going to find themselves more and more frequently using "the physical brain" as just a synonym for what they used to mean by "self," until it turns out they have just changed words, not concepts.
Well, my point was that they would use the phrase equivocally (hence the word "synonym" and the phrase "not concepts"). I take it you mean that it would be an improvement if they meant it univocally, i.e., as a genuine reduction (since you are a reductionist). But then it would not be an actual synonym for how the word is used by them previously, i.e., unreduced, and they would have changed concepts.
So I think that the think you are saying would be an improvement and the thing I was saying would happen are two different things.
Congratulations on finishing your poems, by the way.
Thanks. Yes I would prefer if they use it univocally. Still using it equivocally, even just as lip service, would be a good start, i.e. an improvement.
Like "all men are created equal" was a good start in a country with slavery, and did eventually help to do better.
> but why does it being the result of physical processes mean that thing, whatever it is, does not *exist*?
It doesn't mean that. A lot of people think physicalism implies the self doesn't exist or consciousness doesn't exist, but that is no implication of physicalism as a methodological framework. What physicalism says is that everything that exists is physical or is grounded in physical events. How exactly one populates the furniture of reality given this constraint is open to a broad range of theories.
Thank you. So am I misinterpreting what our essayist means by the phrase, "people usually believe themselves to be real"? If I am, I'm happy to be corrected. I just want to know.
Yeah, I think you're misinterpreting it (or perhaps I'm interpreting too charitably). It reads to me like a premise phrased to be the minimal claim needed for the argument to go through. "People believe themselves to be real" is less contentious than "people are real". The point of the syllogism is to undermine the assumed "thing-ness" of consciousness which leads people to look for a "thing" that can be identified as consciousness.
The author wants to argue that thoughts are neural oscillations and the self is a kind of self-referential neural oscillation. He needs to undermine the natural inclination to look for a "thing" that underlies our consciousness (and presumably the self). But there's no reason to limit existence to only "things" that are space and time invariant like a rock. Neural oscillations are much more dynamic and so change over time, come in and out of a stable existence, and so on. But this doesn't mean that the self doesn't exist. It's just not a static thing.
Interesting. What you say makes sense to me, both as an interpretation of the phrase and as a way that the neural-oscillations hypothesis could be compatible with the self. I will say, though, that in the comments both the author and others seem at times to shade over into a more doctrinaire declaration that selves do not exist, period.
Yeah, its a popular claim and I can't be sure the author doesn't hold that position. It's unfortunate that the "no self" idea has gained so much traction in the pop-philosophy sphere. I guess it's catchy/jarring and so serves the purpose of gaining attention. It also has an edginess to it that appeals to some.
The arguments for the "no self" idea that I've seen are mostly merit-based, so to speak (selves are changeable, evanescent, internally divided, etc., etc., and thus do not deserve to be considered truly existent), or moral/utilitarian (the world is a better place and people are happier to the extent they do not believe in or esteem their selves), or characterological (anyone who defends the self is clinging to social teaching and needs to let go, or is simply corrupt, etc.). Any or all of these points may have some merit, but none seems to me to address the "raw ontology" side of the question. I'm curious if there are arguments that do.
Generally, the sense of self has several pointers. I think no-self claims that one or more of these pointers actually point to something that does not exist, as opposed to the things that do exist. (I think the author of the post would agree with that claim, even if they don't agree about how many selves don't exist.)
I think maybe no-self claims people perform a motte-and-bailey with their sense of self; it refers to things that do exist when they scrutinize it, and sometimes a thing that doesn't exist when they don't scrutinize it.
"I think no-self claims that one or more of these pointers actually point to something that does not exist, as opposed to the things that do exist." Of course--but what evidence and/or arguments do they have for this claim that are not moral, characterological (believing in the self reveals an immature refusal to let go, etc.), or merit-based (the parts of the self that do not exist are not worthy of the label, existent)? I'm curious about more philosophical, ontological-type proofs.
The sentence that confused you wad just trying to entertainingly say that most people have an intuition that they are real. It doesn't say this intuition is right or wrong.
I think people do exist, including most of the ones who think they do. :-) And subjective experience exists as well. There's just no 1-to-1 relation between these.
Imagine a shallow pan with a layer of sand at the bottom, otherwise filled with water and sitting in a rainstorm. Any given raindrop will produce a wave in the surface of the water, which when hitting the edge of the pan will reflect as the wave continues to propagate, at some point interacting with itself and creating noise. The movement of the water itself will also change the sand at the bottom of the pan, which will impact the propagation of the waves. The rain continues, more waves occur, the physical structure of the pan continues to shift. Looking at the waves, the pan is both influenced by the external, and influencing itself.
Add that basic metaphor to something with an ability to make changes to the external, and things get really exciting.
There is no hard problem; the hard problem (like free will) is an illusion, and like the illusion of free will, it becomes incredibly obvious once you see it.
This post is about the easy problem, and instead of just saying there is no hard problem, it begs the question in a confused way by imagining a solution to the easy problem would be accepted as a solution to the hard problem by hard problem believers. But that is not possible because hard problem believers are as confused as free will believers.
The easy problem is of course interesting at a technical level (Neuralink has entered the chat), but a niche semi-amateur research proposal on it seems like a weird topic for a guest ACX post.
The question of free will and confusion around it arise from very specific evolutionary useful algorithms that human brains have. (For everyone interested, see Yudkowsky’s solution to free will sequence; though he recommends people read the reductionism posts and attempt to dissolve the question on their own before reading the spoilers.)
I’d be pretty surprised if you have similarly dissolving the questions around qualia and consciousness. If you do and have written about it somewhere, then please share a link.
Global Workspace Theory has been proposed as a theory of consciousness too. I wonder whether a connecting analogy would be that the global workspace is the sum of all active oscillations. Would you agree?
> Spikes that are part of the same oscillation fire in sync; they have a shared rhythm.
Your description of neural oscillatory loops has reminded me of phase locked loops (PLL) - circuits that create clock signals of fixed frequency (sometimes configurable or adaptive).
When reading the Wikipedia I stumbled over the new type of Neuronal PLL and there seem to be efforts to reverse engineer brain structures in terms of such N-PLLs. For example:
I feel like a bit of a jerk writing this, but what else is new? I couldn't finish this piece, despite being personally obsessed with phenomenology of inner experience, etc. I know part of the reason was that I was extremely tired when I started it, but also it seems pretty lifeless as an essay. Part of it is even in outline form, and the rest of it seems like an outline come alive. (Reminds me of something grim Kafka said: "I am a memory come alive."). The authorial presence was just too weak. Not only was there nobody being witty or playful, there wasn't even anybody saying something along the lines of "now you may be thinking X at this point. Here's why you're wrong." Nobody saying, at the outset, something alone the lines of "People seem to get stuck on this question. I think I've gotten unstuck. I'm going to tell you how I did it." Instead, there's a hiya from the author with a list of 3 reasons to read the article. Then we start the article, and quickly encounter a list of 16 items, then a list of 3 items, then another list of 3 items, and by then I was outta there.
Can’t help but contrast this with another rat-adjacent attempt at the problem, which, I think, engages with the explanatory gap more intently and in clearer terms. It didn’t convince me of anything (I don’t like the problem and don’t want to be convinced) but the proposed solution at least seemed well-formed given the problem:
I don't know why that has so many fans. It is two.theories, not one, and the second is probably redundant. The first theory, that Existence Is Consciousness, is not a reductive explanatiin, as advertised , because it is presented as a brute fact, not as the outcome of some "gears". And what does the second theory, Consciousness is Computation add?
This seems entirely consistent with my own prior understanding, but with more supporting detail and clearer presentation than I would have been able to assemble.
Shouldn't any theory that claims that experience of qualia is nothing more than a physicial arrangement with certain properties also imply that creating conciousness can be done by creating such a structure?
And if such a structure turns out to be useful computationaly, how likely is it that artificial neural networks will at some point have this structure if they don't already?
what are the implications of this theory to the moral value of AI?
I have been thinking about this too. If we understand what kind of structures give rise to consciousness, then it is possible to engineer them. And reduce them to the absolute minimum. You don't really need a lot of the pre processing if the essence is the reflective oscillations. Maybe a small integrated circuit suffices. That has worrying ethical consequences.
But this is also true of other such proposals, e.g. this:
>Shouldn't any theory that claims that experience of qualia is nothing more than a physicial arrangement with certain properties also imply that creating conciousness can be done by creating such a structure?
Wouldn't that structure necessarily include animate physical matter in order to achieve consciousness as we know it? Or do you think all of our physical body information has no bearing on consciousness ? It would possibly be conscious in some sense of the word but it would have no relationship to ours.
I'm not sure. Why must it include animate physical matter?
I mean, yes I intuitively agree that a physicalist theory of consciousness should preclude things which are not physical matter being conscious. But if the theory, as written defines consciousness in a way that does not preclude that we have to bite the bullet or else reject the theory. You can't have a theory of "what consciousness is" that fails to have a 100% correlation between "things which satisfy the theory" and "things that are consciouss"
But Anyway computers are also made up of physical matter so even if the theory did require that it's not a problem.
> But Anyway computers are also made up of physical matter
Animate physical matter is my point. All of the information and demands of a living creature as a major input into the information processing of the brain.
Depending on how one defines consciousness, a computer certainly could be conscious. (an ability to introspect on its own process ) but it will bear no relation to human consciousness; it will be a horse of a different color. I don’t see any way that a computer can have a sense of living or dying ,for instance ; those cannot be meaningful concepts to it. I also don’t see how human consciousness can exist without the concept of living and dying as a significant part of it. It is one of the things that we introspect most about.
I have been mulling this over, and I take your point about "If consciousness is real then computers either have it or they don't" , however I wonder if this is only true if one subscribes to a physicalist view of consciousness. I have no problem with that. But then it becomes a question of what inputs shape that consciousness, because consciousness is not a static thing, it's a dynamic system, and it is very much involved with it's own state, and the state of the external world. Here, I think, one can draw a distinction. I don't think it is possible to "accurately emulate" (as a less loaded version of "endow") the input of being housed in a physical, animate, living body. This experience is unique to every human being, and yet there is broad commonality as well. I don't think a computer-housed consciousness will be at all the same, but I guess that is a distinction of degree, not of kind.
Do neurons really distinguish spikes that are part of a particular rhythm? If this is a thing, why isn't it in the standard models of neural spikes? Also, don't we have evidence of brain harmonics, with eg 7 alpha oscillations for every gamma?
The bullets to bite looked good to me. I liked the overall framing, though I'm not sure the neurology holds up.
I don't have empitical evidence that individual neurons can treat signals received from an oscillation they're part of differently from other signals. I'm only saying that's required for my theory to hold water, and why I think itis plausible that they can.
Yes there are brain harmonics! I left that out, because the article was essay was already too long (I had to cut a lot for Scott to accept it) and it is not essential to my argument.
I don't know what this comment means, but it seems like you may have misinterpreted mine. The point is that deflationary views of consciousness float better in academic discourse about the realism of other minds than in a discussion about basic empathy.
Daniel, I have some purely technical observations that, in my mind, call the whole premise of the post into question:
[Daniel] "Since
billions of bits of neuronal activity, across many neurons in space and many milliseconds in time, can be part of a single pattern,
but patterns can also remain separate, even while running into each other,
there has to be a difference between these two states of affairs."
Why is there a "difference"? Oscillatory patterns interfere all the time, I'm not sure what particular insight is there to go after?
[Daniel] "A neuron that is part of one oscillation can hardly also be a part of another[...]"
Why? This is left there as if it were an obvious truth, but I don't think there's any reason to believe it to be true.
[Daniel] "Higher frequency oscillations are smaller, which makes sense because higher frequency means less time for the circular signals to travel, and smaller means less space through which they travel."
This statement may be true, but I don't see any references to the brain waves dispersion. It is not apriory obvious that high-frequency waves propagate faster that low-frequency ones. I sense a degree of confusion between wavelength and propagation speed here.
[Daniel] "There are also larger and therefore slower oscillations. Neuronal signal propagation has very variable speeds, but the lowest of low estimates still gives it half a meter per second, i.e. much less than a second to travel straight across the entire brain. Every thought that lasts longer than that, such as your understanding of this sentence, has to be at least a bit circular and therefore oscillatory."
"Larger therefore slower" doesn't make sense, there's a confusion between amplitude and... frequency? propagation speed? "has to be at least a bit circular and therefore oscillatory" is meaningless, oscillatory processes can be depicted in polar coordinates, but that doesn't make them "circular"; is a circular propagation in the brain implied here? As a surface wave? this is very confusing.
[Daniel] "Spikes that are part of the same oscillation fire in sync; they have a shared rhythm. For each neuron that is taking part in an oscillation, the time elapsed between its own sending of a spike via its axon, and the arrival at its dendrites of subsequent spikes from neurons it oscillates with, remains constant over multiple such intervals. That is what it means to share a constant rhythm."
This doesn't make any sense. Neurons are firing at variable GHz rates ("A living human brain will contain between 9 and 200 billion spikes per second"), the much-slower brain wavers must be oversampled, therefore assuming the neurons firing "in-cync" is not in any way substantiated.
I can go on, but the pattern emerges (bad pun!). These statements build a foundation for what follows, and I think the foundation is very flimsy.
High-frequency waves don't need to move faster to require smallness; they move in smaller patterns and move at the same speed, which I think is the only way to end up back where they started more frequently, as is (?) required for high-frequency.
I don't think they meant to talk about amplitude at all. "Larger" means physically larger.
Brain-waves are produced (almost exclusively?) by oscillations, according to the post.
Individual neurons rarely fire more often than 100Hz. (I'm not sure you were confused about that.)
Ok, I think I see where I'm confused: are these oscillations in time domain, or do they also have a spacial dimension, which would make them "waves"? This is super-unclear.
What I'm getting at is that an oscillatory process has to have a time dimension, i.e., a period (s) or frequency (1/s or Hz). But it doesn't necessarily have a spatial dimension; for example, pressure inside a Helmholtz resonator fluctuates in time, but does not propagate, the phase is the same through the volume. So it's not clear why pulsating neurons have to create a traveling wave. And of course I'm still not there with the idea that a neuron can only be in one wave at a time, superposition is a thing, and if it somehow doesn't work for neuron networks it needs to be addressed.
Yes, I clearly was confusing the rate of neuron firing with the rate of the whole brain firing. The weird implication of this is that there's at most 2 billion (American "billion", 2E9) neurons firing at a given time? This number seems low, but that's what we get dividing 200 GHz by 100 Hz?
Neurons fire (much?) slower than 100Hz most of the time.
The brain has some oscillations with (very) little or no spatial dimension. Those are irrelevant here, I think.
Oscillations interfere with each other. (Neurons have a very small amount of 'memory', so they can act different for one oscillation depending on what the other is doing. They also have other nonlinearities, like a refractory period.).
So it's hard for neurons to carry multiple at once.
Neurons are sometimes involved in multiple oscillations, when those oscillations' periods are (simple fractional?) multiples of each other.
The term 'oscillation' might be a bit misleading. A process could move around the brain without its path intersecting itself, but if it moved for long enough (a quarter second?) it would have to produce oscillatory patterns, because neuronal signal propagation is fast and the brain is small.
'Oscillations' describes the things we're talking about. In some cases, it might not be much important that they oscillate.
You should maybe reread Scott's post on brain-waves.
Did! re-read the brain waves post. Left me actually more... confused, I guess? So definitely not spatial waves in any sense, but time-domain oscillations. It also looks like these Greek-letter waves are superimposed, i.e., delta is there always, and the higher-frequency waves ride on top; this is also what pink noise-shaped spectrum suggests.
I think I'm getting a bit ahead of my skis here; for now my takeaway is that these brain waves "explain" consciousness is too simplistic a take, which also may be because I'm not comprehending the original post....
You're thinking at the level of waves in the physics sense, with amplitudes and phases and everything. I'm thinking at the level of neurons, which are too numerous to keep travk of so we use "wave" as a kind of metaphor. As I wrote, the oscillation is itself an abstraction.
I don't see any specific disagreement between us that I can't trace to that difference in perspective. Do you?
I am struck by how much of this discussion seems to pre-suppose that consciousness lies entirely in the brain. There are a lot of proposals of how consciousness might exist entirely independently of a physical flesh and blood body. Recursive computer programs sitting on top of one another and things like that. A pan of water being rained on with sand in the bottom and things like that. These might be good models for how everything gets processed by the brain, but "everything" includes the realm of the senses and the demands of a living, corporeal being which is a "program" that is constantly running in the background. I think one of the earliest glimmers of consciousness is in human beings’ ability to predict the future. To understand the seasons and prepare for them for instance. To finally know ahead of time where the sun will be at a particular time. Consciousness arises from the development of abstract thinking, that can be used to channel and often suppress the demands of the corporeal being.. The prefrontal cortex, executive function: that part of us.
I am not sure what people really mean when they refer to the hard problem of consciousness. Is it a precise understanding of the mechanisms that are involved in a sentient human being, or is it why? Why are we conscious? Or is the question why are we alive? I don't suppose it's that, unless one thinks that all living things are conscious in which case I don't know what to say. It seems pretty clear that some other higher order mammals on earth have certain traits that could lead one to believe that they are on the edge of consciousness. We know something about other forms of pre-humans that didn't make the cut, like the Denisovans and the Neanderthals and others.
I think that the heart of what is termed the hard problem of consciousness is the realization that we are each mortal and our time will end. This is a necessary condition of being able to see into the future based on what is happening now and observing it and it's rather an unfortunate one. because it provokes the abstract nature of higher thinking to come up with reasons why that might not be the case. Archaeological evidence indicates that a great deal of time and energy was put into this problem. The pyramids of Egypt being a good example. In other words, consciousness is a product of the friction between the unalterable state of being flesh and blood mortal creatures and being able to think about that. Elephants seem to have a glimmer of that notion, but I doubt they spend their time on earth thinking about it much until their body starts to fail them. No doubt the computer program or programs working together can mimic this friction, but they don't really have it. If something goes wrong with the way they are “thinking” they just shut down or they indicate to us that they cannot go any further with the process. Does anyone really think that they will have mortal feelings about this state of affairs no matter how complicated they get? I don't think that anything that is constructed of inanimate matter is going to have the hard problem of consciousness.
In the article, qualias have the condition of being ineffable. I don't know whether that's really part of the definition of qualia or not but if it is then it is by definition inexplicable, which does not make it a mystery exactly, just something that can't be explained in any words we have.
I do think that consciousness is entirely in the realm of chemistry and physics, but that doesn't get us any closer to the truth of it unless one can understand the “complete state of being” which includes very much the physical body that we inhabit. I think mind-body duality is a very poor way of approaching this because it isn't really a duality except in our experience of it, which is highly mutable to say the least.
Look at pain for instance, which I assume is a Qualia. People experience pain very differently, so it's actually a whole bunch of qualias.
Given that human beings have a very large capacity to choose how we think or feel about something, the hard problem of consciousness becomes a very individual one and not a general case. My Qualia around the “hard problem of consciousness” is bound to be quite different than anyone else's. In other words, there is nothing to solve here except one's individual process. A devout belief in God and an afterlife is one solution, but there are others. There is no universal solution to this problem, because it is entirely self-created. My opinion…
"The measuring seems trivial. We already know every neuron has a refractory period after it has fired. This constitutes a built-in cellular timer that measures the interval since its last spike"
This is gonna need more explanation. Didn't the refractory period only provides timing information for spikes received so fast that the neuron can't respond anyway?
I'm confused by homogeneity. The qualia of sight and the qualia of touch don't seem to be the same type of thing to me, beyond just being things I experience. But "things I experience" is just what qualia are, so I'm not sure what it explains to go beyond that and say that they're homogenous.
Suppose I say: "Homogeneity: all vehicles are (felt) to be of the same type. While differences between them can be appreciated, they are always (experienced) as the same kind of thing."
(Verbs in parentheses because maybe they should be substituted for something more applicable, but hopefully that doesn't detract from the point too much)
Is this true, or false? Or does it not mean anything, by itself? I tend to think it's the third thing. "All vehicles are vehicles" is about all that it claims, but that's tautological. If we are to ask, are a bicycle, a truck, and a helicopter the same kind of thing, I think we can only answer that question with respect to some category. They're all vehicles, but they aren't all ground transportation. They're all vehicles, but they aren't all self-powered. What makes one of these categories more salient than another? Why couldn't we say that a car and a truck are of the same type, but they are not the same type as a helicopter?
I can't see a fundamental difference between this and doing the same thing for qualia. As far as I can tell, what makes the qualia of color perception and the qualia of taste the same kind of thing is that they are qualia. But I could as easily point to the category of color perception, and say that perception of red and perception of blue are the same kind of thing, whereas perception of the taste of cinnamon is not.
I don't think we need to be able to rigorously define qualia (or vehicle, for that matter) in order for this to be so.
The way it is expressed is quite confusing. Individual qualia like red and green aren't homogenous , and types of qualia such as colours and flavours aren't homogenous. Presumably , the idea is that all kinds of qualia themselves fall under a higher level kind distinguished by a higher level property.
>Homogeneity: all qualia are felt to be of the same type because that’s what they truly are; they’re all bits of information being processed internally.
The way this is expressed is quite confusing. Individual qualia like red and green aren't homogenous , and types of qualia such as colours and flavours aren't homogenous. Presumably , the idea is that all kinds of qualia themselves fall under a higher level kind distinguished by a higher level property.
>So qualia arise out of neuronal information processing much like biology arises out of chemistry
To say that qualia are information processing and nothing more doesn't identify the higher level property of qualia ness, since everything else is information processing.
>Qualia are nothing but information being processed internally, on their own information channel
That's slightly better, but it's nowhere near a predictive the ory.
Thanks for the "slightly better". The predictions are at the end of the essay. Unless you doubt that they fall right out of the theory, the theory is predictive.
Fascinating account of the possible underlying mechanics of conscious experience, but it does nothing to bridge the explanatory gap for a Chalmerian like myself. I have a feeling the hard-problem deniers (eliminativists / illusionists) won’t be satisfied by this essay either, since they see no need to delve into phenomenological themes to begin with.
This is all “easy problem” stuff, sadly; hard problem remains unresolved.
No, wrong. I have demonstrated that something with the properties of qualia has to arise from how thoughts have information processing that is internal to themselves, such as "working memory". This is a purely physicalist account of why SOMETHING WITH ALL THE PROPERTIES OF SUBJECTIVE EXPERIENCE would (necessarily) arise from the kind of information processing that we know the brain (sometimes, when conscious) does.
If that doesn't solve the "hard problem", what would?
Subjective experience is just what it feels like to be thought.
Regarding the Hard Problem: consciousness is either based on properties of the physical world that we don't yet know (if I understand correctly, it is to such theories that they author directs the comment "they don't pay rent") or not, in which case we are left to believe that somehow consciousness appears by whatever solution to the Easy Problem (such as the one proposed by the author).
In other words, consciousness is either something other than a computation, which is mind-blowing, or it is a computation after all, which is way more mind blowing.
I don't see how the ideas presented here (which I find extremely interesting, but I am far, far from an expert) offer some resolution to this paradox.
Consciousneas is a computation. But it is not a computation the brain makes (directly), it is a computation a thought makes. The brain is not a computer, but the thought is.
As a musician, this essay used oscillation in a very different way than I was expecting. One of the key properties of sound is that oscillations coexist. If you pluck a string, it produces a fundamental tone based on the length of the string, and then also lots of other tones based on low-denominator fractions of the length of the string. These overtones (the octave at 1/2, the fifth at 2/3, the fourth at 3/4, etc) are quieter, but they make up a very important part of the sound of any instrument (called timbre). Except when someone uses a synthesizer to make pure beep noises without overtones, those are timbre = 0.
So when I read that thoughts are cycles of oscillations of neuronal activity in the brain, naturally, I thought that the oscillations would combine with each other depending on their frequency ratios, and that's how thoughts would interact with each other. Instead, the essay seems to be saying that the different oscillations somehow block each other. Like if you could hit a bell, and then hit it again in such a way as to quiet the first tone. (I think that's technically possible? But not easy.)
Anyway, oscillations don't generally block each other. I guess it depends on the medium that's oscillating, but they generally pass right through each other, or strengthen specific parts of each other. So most of this is pretty straightforward and makes sense, but I'm skeptical about the oscillations blocking each other.
Also. In The Sickness Unto Death (1849), Kierkegaard wrote:
Man is spirit. But what is spirit? Spirit is the self. But what is the self? The self is a relation which relates itself to its own self, or it is that in the relation that the relation relates itself to its own self; the self is not the relation but that the relation relates itself to its own self.
So, basically the same as your recursive reflections.
Sorry, this is not music, it is not even physics. The "oscillations" and "waves" of electroencephalography research are basically metaphors. We see them on our measuring devices, which don't look at individual neurons but can only notice averages of activities of lots of them. We do know there are "spike trains" and we know some are circular and behave as I described, but the brain is not made of vibrating strings.
are you aware of the human brain project, especially part with neuromorphic hardware and spiking neural networks? www.humanbrainproject.eu
Some years ago I was involved with them and much of what you said fits very well with what I learned at the time. (I was just a programming, not a scientist, so my understing of the subject is very limited).
Among other things they made "spiking neural networks", which are a computer model of neurons, which can be processed somewhat efficiently, while also behaving more similar to biological neurons. They did computer simulations with these neurons with the goal of replicating mechanisms of the human brain. To that end they built hardware, that could efficiently simulate these neurons, so they could simulate millions of these spiking neurons in realtime.
Again, I was no scientist, so I can not point you to any relevant research. Also apparently the project ended last year, so the results should be available somewhere.
Yes there is. That's what you get when you fill attention with nothing but itself.
I think that's the equivalent of moving the facing mirrors so close to each other that they don't have anything beyond each other's bounds to reflect - you get a sense of virtual infinity.
It's too bad that the article (unintentionally, based on OP's comments in the comments section) claims to solve the Hard Problem, because it's a fun proposal on the Easy Problem and I think that some of the weedy details deserve more attention than they have gotten.
I should say from the jump that I am not a neuroscientist, but I've taught a bit of neuroscience at the upper-level-undergrad level, and I hang out with some neuroscientists, and so maybe I'm just in an uncanny valley situation where if I knew either more or less about neurons, this would make sense to me. That being said:
"The measuring seems trivial. We already know every neuron has a refractory period after it has fired. This constitutes a built-in cellular timer that measures the interval since its last spike."
I don't think this is useful for your theory. At least, the standard model of how neurons work is that there is a hard/"absolute" refractory period during which the ion channels reset, and no action potential can occur. Then, there is a soft/"relative" refractory period during which the neuron is rebuilding its electrochemical potential energy, and it is relatively difficult for an action potential to be triggered. This is on the order of milliseconds, and seemly not long enough to be producing the oscillations you want (and also the relative refractory period is probably not the sort of on-off type process that would be likely to produce sustained patterns of action potential firing). Once the refractory period is complete, the chemical mechanisms that produce it are no longer changing--the ion channels are reset, and stay that way, and the cell is at its "resting" potential, and stays that way. So they can't actually encode any time information after a few milliseconds.
So you will need to posit some other sort of mechanism that can encode temporal information on the time scale of ~ 1 second if you want to make this work. There is a brief discussion of pacemaker neurons in the comments, which is more promising, but I think it would be helpful if you did a bit of spadework to help us understand how much more promising. My (very weak, extremely correctable) impression from some quick googling is that we seem to think that pacemaker neurons work similarly to pacemaker cardiac cells, in that they are expressing different ion channels in their membranes in a way that makes their action potential dynamics pretty different from a typical standard interneuron. If this is, in fact, the case, then I'm not sure that we actually do know of any good candidates for molecular processes in interneurons that let them keep time at the precision you want. I'm not saying that such processes can't exist, though.
That leads me into my main point: your list of ways to test your hypothesis with emerging tech is very cool, but your hypothesis also seems eminently testable at the cellular level: we *should* be able to produce interneuron behaviors that match up with your predictions. I.e., it seems like low-speed temporal preferences are extremely important to your theory, if I am understanding it correctly: if you trigger an action potential at t=0.8, t=1.6, t=2.4, then trigger an additional action potential at t=3.2 should be easier than at t=3.1 or t=3.3. Not seeing this in neurons wouldn't necessarily be fatal (you could always argue that temporal preferences are a network phenomenon), but I think it would be Quite Bad.
This sort of thing doesn't require any advances in technology to examine; my understanding from talking to a colleague is that pretty standard neuroscience techniques would be workable here. In fact, these temporal preferences seem so findable in either cell cultures or tissue section sorts of experiments that--if they are as important as you posit--I would be pretty surprised if they hadn't been found already, even if their broader importance wasn't noted. Have any phenomena like this been noticed? I am totally ignorant of the literature, so some sort of review of that might have been useful as well.
I do (now) think it solves the hard problem, in that it demonstrates why (something with all the properties of) subjective experience has to necessarily arise from how we know we process information.
Your other criticisms are valid and welcome! But a flat out unargued assertion that this doesn't solve the hard problem is just insufficient.
No I don't think an ability of individual neurons to distinguish between in-oscillation and other signals has been demonstrated. Yet my theory requires it. That is a clear weakness of my theory, which is why I extensively pointed out this problem.
This is a fascinating and informative post, esp. since I happen to be writing a book on this topic (the intersection between psychological research and the philosophy of the self). I can't find anything to disagree with--it all seems well reasoned and empirically supported. It doesn't solve the hard problem, of course, in the sense that it isn't a complete explanation. It provides a helpful "dive into the nuts and bolts" of neural activity in the brain, but it doesn't provide a complete explanation because I don't think that qualia or the consciousness (not precisely the same thing) can be exhaustively explained by neural activity. The brain has gross (higher level) architecture, after all. The fact that conscious experience appears to be centered in the prefrontal cortex should be important. To move forward, I think the theoretical explanation needs to be scaled up, and include interactions between functional areas of the brain. To be specific, what is the prefrontal cortex doing that the rest of the brain isn't? Presumably, it is creating a simulation of activity occurring elsewhere in the brain, condensing it, organizing it, and disseminating it.
Why does this produce phenomenal awareness? What is the PFC doing that, for example, computers or simple organisms are not (assuming that computers and simple organisms are not experiencing things consciously).
Another wrinkle is to incorporate what we know about how the mind is organized. Qualia, even consciousness, isn't entirely what people mean by the "Self." Memories surely have to have something to do with it. One thing the human brain does, that computers and simple organisms do not (apparently) do is form and maintain complex conceptual memories of one's own experiences. The fact that a series of experiences have happened, that when they happened they seemed to have implications for what kind of person you are, and then you stored those experiences, seems like something important to include in one's theory of consciousness. In fact, I would discriminate between phenomenal experience (conscious experience in the moment), from a self concept that could be organized from a set of interconnected memories that seemed to help you learn things about yourself. Perhaps self awareness is conscious experience subliminally informed by a self concept, in the presence of stimuli that activate those memories?
I admit that a better understanding of how memories are encoded and retrieved at the neural level would help tie this all up in a neat little bow.
Yes it does solve the hard problem. If you'd like to argue about that, or there's any other way I can help with your book, I welcome your email or (better yet) X DM.
I was unaware Dennett thought phenomenal consciousness is ineffable as you say:
"Ineffable: qualia cannot be communicated, or apprehended in any way other than direct experience."
In his conversation with Robert Wright (https://www.youtube.com/watch?v=RO7fro4frKk), he says that in principle, one can fully know and experience the subjective consciousness of another being if they have perfect knowledge of the state of that being's brain (kinda like Laplace's demon). Or perhaps I've wildly misinterpreted him.
Yes you're misinterpreting him. He wrote a lot on the subject, especially "Consciousness explained" (which arguably doesn't, but is still a great book).
Fantastic article! It puts together a lot of things that the series of thoughts observing this comment being written have had vague ideas about for a while (e.g. that thoughts come from localized electrical activity, but not precisely how they avoided interfering with each other.) I suppose it should have been intuitive given that different phases of sleep have different frequencies, but then again, I'm no neurologist.
I do think that the proposed falsifiable experiments provide a good starting point, but I'm curious as to how much thought you've put into frequency comparison for individual neurons (and neuron chains). As I understand it, the brain is at a (fairly) constant temperature, and each neuron can only be surrounded by/contain so many Na/K ions, due to the space that they take up. Given that energy and reagent amounts are (roughly) the only two things you can vary to change a reaction's rate (I'm not sure if neurotransmitters act as catalysts?) it seems as if there are a limited number of frequencies available for all the thoughts in the brain to use. I'm wondering if that's a limiting factor on how many thoughts we can "hold" in our brains at one time, or if this is nonsense and there are far too many other variables around for it to matter.
finally got-around to reading this, thanks! Some good points about what thoughts are and are-not, but when the author turns to reflective thought then the old assumptions creeps in that there is a real, singular "you" that "can introspectively notice," and "deliberate, analyze," etc. when the idea is that you aren't doint this - thoughts are! Also, while quantitative measures attractively are comfortable, and will continue to provide more data and evidence, I'm always suspicious of interpretations derived from fMRI pictures. But, like I said, thanks again!
PS - Minsky, long ago, wrote in Society of Mind that thoughts themselves are ambiguous.
"Thoughts are the processes you can introspectively notice in yourself when you deliberate, analyze, evaluate, reason, form concepts and solve problems. You can also notice similar, but simpler and briefer processes when you imagine, remember, notice, recognize or judge something or when you feel an emotion or motivation. These simpler, briefer mental events will also be called “thoughts” here."
"However, if you hook those EEG electrodes up to the amounts of computational power available these days, you can mathematically reconstruct quite good guesses about where in the brain the electrical signals are coming from. And that’s a game changer. This combined temporal-spatial resolution lets you localize individual neural oscillations, if they’re large enough. And that’s how you get to look at (oscillating) thoughts!"
Have you tried looking in the scientific literature if something like this has been done? I feel your proposal has some value, it would have more strength if the criteria to test it would be operationalized.
IMHO EEG is not enough, it's limited to the surface and too much low-res. However intracranial EEG is making huge leaps and has the resolution and depth (it's often used in epileptic patients) to capture something.
Yes intracranial EEG is very promising! But you can find papers with simultaneous LORETA and fMRI that demonstrate that LORETA does make oldschool EEG able to make valid observations of phenomena deep in the brain.
Valid observations may not be enough either. There are cheap EEGs around to test your hypothesis though, personally I have 2 Ganglion from OpenBCI (4ch each). If you figure out min. number of channels needed we could build a cheap setup and do some citizen science with the BCI community :)
I am a theoretical physicist. Among others, I am interested in the foundations of quantum mechanics. In the book "The Grand Biocentric Design: How Life Creates Reality" that I co-authored with Robert Lanza, we explain that quantum mechanics cannot be understood without bringing consciousness into the game. The concept of quantum state, described by a wave function, is closely related to consciousness. https://www.amazon.com/Grand-Biocentric-Design-Creates-Reality/dp/1950665402 . These ideas are summarized in the video Biocentrism: A physics perspective https://www.youtube.com/watch?v=3LgGv4jb3sU on the YouTube channel "Chasing Consciousness".
Thank you very much!
I thought rent was supposed to be paid in anticipated experience, not claims about possibility. https://www.readthesequences.com/Making-Beliefs-Pay-Rent-In-Anticipated-Experiences
Machine consciousness is relevant to ethics. How do you observe the Consciousness of a machine? If you csnt, then you care about something you can't observe. Which could be a counterargument to "rent was supposed to be paid in anticipated experience, not claims about possibility", if you are willing to treat it as unfalsifable.
When did I say I care?
Perhaps you dont, but some people do. A lot of rationalists do, which is why the Anticipated Experiences thing is not something they should believe.
Don’t illusions invite investigation? My impression was that the various Buddhist sects had explanations, although they don’t seem very satisfactory to an outsider.
Illusions invite investigation in science, too.
OP claims that when adopting the particular Buddhist approach discussed, things which do not exist do not call for explanation. That is what I disagreed with. I did not intend to imply that only Buddhists might want to explain illusions. My point, rather, was implicitly that science would want to explain illusions, but that even if we adopt the Buddhist approach we might be curious regarding the origin of this illusion. And in fact, although I am not conversant enough with Buddhism to quote chapter and verse, I think that some sects of Buddhism do claim to know what causes the illusion of self.
If perspective, self, ego, are illusions, what is it that is deceived? Does an illusion in the very act of deceiving create its own illusory victim? I've long been puzzled by this, and am curious about answers.
Why assume a single precise meaning? Polysemy is a thing.
I find it quite demoralizing to read yet another speculation about how the Easy Problem of consciousness might be solved, followed by a declaration that these ideas would solve the Hard Problem, as if the basic premise has not been internalized at all. In 100 years, I think people will begin to think seriously about this again, once traumatized thinkers shake off their fear that serious thinking will damn them to religion and woo.
People do think about this seriously, just not in these circles. Bertrand Russell might be a good entry-point for rationalists, given his existing prestige. His "The Analysis of Matter" and "The Analysis of Mind" are a pretty comprehensive survey of his views in the area.
I'm glad this was the first comment, as it made me feel less lonely upon reaching the end of this article.
Rarely have I read something better calculated to seem like an accidental full-on refutation of physicalism, or to stoke concerns that physicalists are p-zombies.
(Note: I don't actually believe they are. They're awe-inspiring human beings like all the others, and I'm so glad the writer is interested and giving this a shot. May truth be his reward.)
I didn't say this solves "the hard problem". In fact I think "the hard problem" is a racket. You can just continue to claim it is unsolved regardless of what happens. Unfortunately this creates negative externalities: It lets people use consciousness as a blanket excuse for them to believe whatever they want to believe, much like epigenetics is misused the same way.
I enjoyed this article - seems like a pretty compelling theory of how conscious thoughts might work mechanically. However, while the oscillating nature of bigger thoughts helps explain why only they are capable of self-reflection, it's less clear if or why lesser, non-oscillating thoughts (or eg computer programs) don't have subjective experience at all - as opposed to just basic/non-reflective experience.
Well, you may think the "hard problem" is a racket, but it's the problem people typically have in mind when they talk about qualia being hard or impossible to explain, or decide that physicalism is inadequate. So if you write a blog post that promises to explain qualia and show that physicalism is adequate, and the only attention you give to the hard problem is to say it's "a racket" and "a blanket excuse to believe whatever [people] want to believe," you can't be surprised when people are underwhelmed.
In fact, it seems to me that if you want to convince people to adopt your views, it's much more important that you focus on convincing them of your claims in this comment than convincing them of the answer to different questions, however interesting they may be (and they were interesting!)
There is a standard dynamics with complicated problems. As soon as some part of the problem is explained and therefore not confusing anymore, people immediately start claiming that it has never been the "real problem" and move the goal post further towards still unsolved parts of the problem, without noticing that fraction of solved to unsolved keeps steady increasing. At some point there is finally nowhere to pass the buck and then the "real problem" is revealed to be nothing more than a semantic confusion and therefore dissolves itself.
Knowing about this dynamics there are two strategies. You can talk about this particular meta-point in attempts to persuade some people who are still confused about it and facsinated about their own consfusion. Or you attempt to solve more and more of the "easy problems" with people who are already understand the meta-point, therefore increasing the ratio of known to unknown, bringing closer the moment where everything is finally resolved.
Well, that's a fair point. But it's hardly as if the "hard problem" is new, or has been in retreat behind a smaller and smaller curtain. In fact, the basic point is far older than Chalmers.
It's not as though people have been vague or unclear about what it is about qualia and consciousness makes physicalism (as standardly understood) implausible, or why qualia seem hard to explain by physics in the first place. It's perfectly fine to declare that you think their concern is a pseudoproblem and you're going to talk about something else that you think is real. But then don't start your article by saying you're going to address their concerns.
> Well, that's a fair point. But it's hardly as if the "hard problem" is new, or has been in retreat behind a smaller and smaller curtain. In fact, the basic point is far older than Chalmers.
Exactly! Chalmers just gave a name for the problem at the state he encountered, but the problem itself is ancient. Long before people were talking about souls and how they, despite their immaterial nature manage to control material bodies, how living organism seem to be behaving by completely different rules than non-living matter. How matter only deterministically follow the rules and therefore can't possibly make decisions under uncertanity. All these seemed to be hard parts of the problem until they weren't anymore and the goalpost shifted.
> It's not as though people have been vague or unclear about what it is about qualia and consciousness makes physicalism (as standardly understood) implausible, or why qualia seem hard to explain by physics in the first place. It's perfectly fine to declare that you think their concern is a pseudoproblem and you're going to talk about something else that you think is real. But then don't start your article by saying you're going to address their concerns.
Initially people who came up with the notion of qualia assumed that they are not physical by definition, while also pointing to some observable referent. There are three options:
1. Qualia indeed are non-physical and exist
2. Qualia do not exist, and the referent they were pointing to is somehow not real.
3. The referent is real but physical, therefore we need to adjust the term.
The post works with the third approach. It shows how all the desired properties of the qualia can be achieved and explained if qualia are physical, therefore addressing some of concerns of people who thought that things with such properties can't possibly be physical. Whether to call it to be part of the hard problem or the easy problem is just semantics.
>Exactly! Chalmers just gave a name for the problem at the state he encountered, but the problem itself is ancient. Long before people were talking about souls and how they, despite their immaterial nature manage to control material bodies, how living organism seem to be behaving by completely different rules than non-living matter. How matter only deterministically follow the rules and therefore can't possibly make decisions under uncertanity. All these seemed to be hard parts of the problem until they weren't anymore and the goalpost shifted.
Actually, for much of philosophical history, there was not much concern with these issues, because they appeared to be easily solved problems. At least in the past millennium, it was mostly post-Descartes and enlightenment that the puzzle began to arise, precisely because people began to have ontologies that were not rich enough to play host to the qualia that are the most obvious aspect of existence.
> The post works with the third approach. It shows how all the desired properties of the qualia can be achieved and explained if qualia are physical, therefore addressing some of concerns of people who thought that things with such properties can't possibly be physical. Whether to call it to be part of the hard problem or the easy problem is just semantics.
Well -- it shows how all of *its* desired properties of qualia can be achieved, I suppose. But it completely ignores what qualia actually are, and where the name comes from -- their qualitative aspect. This is the interesting thing about qualia that seem inexplicable on a physicalist ontology. So it does not even try to make headway on the problem that keeps people thinking that physicalism is inadequate to explain qualia. But it does *promise* to try. And I think it should not have done so.
It's always easy to explain anything with any theory, if you're allowed to redefine the terms you're trying to explain.
>Initially people who came up with the notion of qualia assumed that they are not physical by definition
That is false.
>There are recognizable qualitative characters of the given, which may be repeated in different experiences, and are thus a sort of universals; I call these "qualia." But although such qualia are universals, in the sense of being recognized from one to another experience, they must be distinguished from the properties of objects. Confusion of these two is characteristic of many historical conceptions, as well as of current essence-theories. The quale is directly intuited, given, and is not the subject of any possible error because it is purely subjective.[5]: 121
Nothing about non-physicality there.
>. As soon as some part of the problem is explained and therefore not confusing anymore, people immediately start claiming that it has never been the "real problem" and move the goal post further
Or...they just point out that solving part of a problem is not solving the rest of it.
Sure, but this is beside the point.
If the point is that all your opponents are all equally bad, it is unproven.
"Physicalist theories have trouble bridging what has been termed the 'explanatory gap'"
> Yes, the hard problem.
"An area that does not give physicalism such trouble is how the mind’s information processing (conscious or not) is based on neuronal activity"
> Yes, the easy problem.
"Philosophically, this explanation seems to obviate alternatives to physicalism, such as idealism and dualism."
> This is saying that solving the easy problem solves the hard problem. The article does have any bearing on idealism, dualism or physicalism.
"You can just continue to claim it is unsolved regardless of what happens."
> If I did that, it would make me annoying, but it would not make it less of a question.
If a fundamental question appears to be insoluble or impractical, one is free to ignore it.
"It lets people use consciousness as a blanket excuse for them to believe whatever they want to believe"
> And there it is. We cannot abide by these daunting, fundamental questions, no matter how glaring, because we cannot allow a loophole where someone can get away with believing in religion or woo.
Yes, exactly. OP's reply here seems (to those of us who find the hard problem weighty, or anyway to me) to be saying, "my belief system is so obviously true and fully adequate that anything it cannot easily answer must be relegated to being a non-question, probably asked in bad faith." There's nothing slightly new about the attitude, of course (including in the context of very different belief systems than his).
Well, it doesn't look like there's a way for currently available methods to even begin interfacing with the hard problem, and yet all non-physicalist paradigms are pretty much discredited these days, so a certain amount of tension is unavoidable. I don't expect much clarity on this until we're close to be able to build conscious minds from scratch.
Non-physicalist paradigms are unpopular among a certain subset of modern westerners. That's not quite the same as being discredited.
As soon as those with profound mystical understanding would oblige to turn their paradigms to actually useful ends (but at least tangible will suffice at first), we benighted Westerners might just notice.
"In fact I think "the hard problem" is a racket".
Agree. It always seemed like more of a pseudo-problem to me, more a function of philosopher's lack of imagination (or perhaps overactive imagination, I'm not sure). People used to think that figuring out what made life work would forever be mysterious, or that weird forces like elan vital would be needed to account for the miracle of life. Then we made more discoveries and a hard problem turned into an easy one.
I’ve come around to that way of thinking too.
For me, when Nagel asks “what is it like to be a bat” he’s collapsing millions of individual moments into a single word — like — and that obscures that each one of those moments can be interrogated and (eventually, I think) explained by hard science.
It also elevates novelty — say, the ability to use sonar after previously being unable to — but to me that’s no different than the novelty of the gap in your mouth when you lose a tooth. Eventually you stop noticing the gap and being “like” yourself is the same as it always was.
So ultimately I think feeling “like” a bat is probably not much different than any other conscious experience.
So do we right now have a theory that can predict qualia , including novel ones, from neural activity?
Not that I know of, but we're starting to make progress in linking qualia to neural activity, even if it's still in coarse resolution. There's no reason in principle that I can think of why we shouldn't be able to develop such a theory once we make more progress on the empirical side.
There's any number of potential reasons, since physicalism and reductionism aren't guaranteed to work.
>I didn't say this solves "the hard problem".
You kind of did, right in the intro:
"Nobody knows for sure how subjective experiences relate to objective physics. That is the main reason why there are serious claims that not everything is physics. It has been called “the most important problem in the biological sciences”, “the last frontier of brain science”, and “as important as anything that can possibly exist” as well as “core to” all value and ethics. So, let’s solve that in a blog post."
How subjective experiences relate to objective physics is the hard problem. If you think the hard problem is a racket, just say so from the get go.
+1.
Isn't the whole reason why the Hard Problem of Consciousness is allegedly untractable via materialistic science that it's not about "*how* subjective experiences relate to objective physics" but "*why* subjective experiences relate to objective physics the way it relates"?
That's a decent enough one sentence summary, though depending on how you word it you can switch the "how" and "why" between the two.
The Easy problem of consciousness is figuring out what things happening in the brain are correlated with conscious experience. So you see X happening in the brain, and you know that's correlated with Y experience in the mind. Seems doable, we're getting better at seeing what is happening in the brain, and a lot of what we do know seems to correlate with experiences in the mind (say, for instance, a lesion on Broca's area leading to aphasia).
The Hard problem is figuring out how it's possible for the movement of atoms and electrons, or the processing of information through neurons, to create subjective experience at all. As far as we can tell atoms and electrons don't experience anything, why would putting them in a network change that? Thoughts also have qualities that simply don't map to material objects. For example, a thought can be true or false, or rationally valid or invalid. But what does it mean for a material process to be "true" or "valid"? Certainly we can assign that kind of meaning to material processes; say we build a special sheep gate that flips a flag up and down whenever a sheep passes through it. We could then use that flag to know whether the sheep is in the pen, or has left the pen, and in that sense the flag would be telling us whether the idea "The sheep is in the pen" is true or false. However that assigned meaning is all in our heads. There's nothing in the flag itself, or the system of the gate, that contains the idea "The sheep is in the pen". If someone came along and found the gate they might be able to figure out how it works, but not what it meant to the person who made it because the meaning was in that person's mind, not in the atoms of the gate. Objects and processes don't have meanings, but thoughts do, so if our thoughts are just a very complicated system of material processes then where is the meaning coming from? You can't say that the meaning of the neural process is assigned by our minds, because the mind that assigns meanings is what we're trying to explain in the first place!
> That's a decent enough one sentence summary, though depending on how you word it you can switch the "how" and "why" between the two.
Well, yes, because answers to "how" and "why" are not some separate magisteria. This is why hard problem of consciousness is much less special than some people claim it to be.
> The Easy problem of consciousness is figuring out what things happening in the brain are correlated with conscious experience. So you see X happening in the brain, and you know that's correlated with Y experience in the mind. Seems doable, we're getting better at seeing what is happening in the brain, and a lot of what we do know seems to correlate with experiences in the mind (say, for instance, a lesion on Broca's area leading to aphasia).
Are we talking only about correlations or causations as well? You use the verb "lead" in your lesion example which implies the latter.
> The Hard problem is figuring out how it's possible for the movement of atoms and electrons, or the processing of information through neurons, to create subjective experience at all. As far as we can tell atoms and electrons don't experience anything, why would putting them in a network change that?
When you put it like that, huge part of the Hard Problem sounds just as not understanding some very basic things in general and has little to do with consciousness per se. For example a lot of things together have more degrees of freedom than one single thing. And so it shouldn't be surprising that they can have more properties than a signle thing. One can come up with as many examples from real world as necessary, but at its core its observation about logic, not physics.
> Thoughts also have qualities that simply don't map to material objects. For example, a thought can be true or false, or rationally valid or invalid. But what does it mean for a material process to be "true" or "valid"?
It's ironic because you are using the word "map", while questioning what could it mean for material objects to be "true" or "valid". This also sounds as a confusion about what "truth" means instead of problem related to consciousness. When we understand that it's about the relation of correspondance between things, the mystery dissolves.
> Certainly we can assign that kind of meaning to material processes; say we build a special sheep gate that flips a flag up and down whenever a sheep passes through it. We could then use that flag to know whether the sheep is in the pen, or has left the pen, and in that sense the flag would be telling us whether the idea "The sheep is in the pen" is true or false. However that assigned meaning is all in our heads.
A good example. An important point is that it's not just about which meaning we assign. The correspondance between the state of the flag and sheep being in or outside of the pen is something that other people, unaware of our intentions can notice by observing the state of the pen and the flag. Even if we didn't think to assign this kind of meaning and just so happened to design the system this way, another person may notice the correspondance and point it out. This person doesn't even have to be conscious, by the way.
> There's nothing in the flag itself, or the system of the gate, that contains the idea "The sheep is in the pen"
Not on their own. Only when they are part of the system which opperation preserves the correspondance between the state of the flag and the pen.
Again, none of these things are actually relevant to consciousness. Just a bunch of separate topics, interesting on their own. It initially feels that they has to do something with the mystery of consciousness and the hard problem. But as soon as we try to engage with them they turn out to be their own thing, which we actually already have answers for. And this dynamics creates expectations that other parts of the "hard problem" will follow. That as soon as we've properly factorised the problem and engaged with its parts, the only mystery left will be the "easy problem".
>This also sounds as a confusion about what "truth" means instead of problem related to consciousness. When we understand that it's about the relation of correspondance between things, the mystery dissolves.
This is where you're getting confused. The correspondence between things is part of the Hard problem, because correspondence does not exist in the material world but only in the world of the mind. I have a thought: "The sheep is in the pen." That thought either corresponds to material reality or it doesn't (the sheep is either actually in the pen or not). If it corresponds, it is a "true" thought and if it doesn't it's "false". Yet the correspondence itself only exists in the mind, not in material reality. Even if you solved the Easy problem, and could point to exactly the atoms in my brain that correlate with the thought "The sheep is in the pen" and then observe that atoms that make up the sheep and the pen, you still wouldn't find any "correspondence" in the atoms themselves, only in your own mind as you judge whether a correspondence exists.
Which brings us to the pen again:
>The correspondance between the state of the flag and sheep being in or outside of the pen is something that other people, unaware of our intentions can notice by observing the state of the pen and the flag.
Yes, other minds can decide to assign that correspondence to the sheep and the flag, and assign their own meaning to the gate device that matches the meaning the original inventor had; but they can also assign different meanings to it. Someone might look at the system and decide that it is a method of being notified when grass in the pen has become too short; when the grass is too short, the sheep leaves and the flag changes. This machine is a grassometer! Or they might decide that the device measures whether it is raining or not. You see there is a shed in the pen, and when it's raining the sheep goes into the pen to seek shelter, raising the flag as he does. It's a rainometer!
The reason a mind could come to many interpretations of the device's meaning is that the meaning, and correlations, and purpose of the device don't exist anywhere in the device itself but only in the mind that interprets the device. Nowhere in the device will you find correspondence, only in the minds that observe the device.
So to dismiss the Hard problem as a confusion because the truth value of a thought exists in it's correlation to other things is a bit like a man trapped in a hole twenty feet deep concluding "There's a 20 foot ladder at my house, that will get me out. Problem solved!" The whole problem is that he can't get out of the hole, which means solutions involving being out of the hole are not solutions at all.
I don't accept the claim that how subjective experience arises out of physics is "the" "hard" Problem of consciousness. Because that claim presupposes a distinction between subjective experience and things we can easily study scientifically, and emphasizes the differences between those things, and I'm not persuaded that's a fruitful approach to the matter, unless you want to establish some separate magisterium that the neuroscientists aren't allowed into, as a kind of jobs program for philosophers.
Again, if you want to say that the hard problem of consciousness is ill-founded, then you should say that: but don't say that you've solved the hard problem when you haven't, and especially don't say you've solved it when you think it's not a solvable problem at all.
Yeah, okay, if semantics are important:
I claim to have solved what others would call the “hard problem” of consciousness.
Well now I'm confused: I thought you don't accept the claim of the hard problem; if you don't accept the hard problem as a valid problem, how can you claim to have solved it?
+1
I think you're wrong to claim that the same neutron can't participate in multiple waves. Consider heterodyne and radio waves. If the wave were just on one sequence of neurons you would be correct, but if it's on multiple neurons, then reinforcing peaks could happen (and, of course, reinforcing troughs). But multiple parallel (within the path) neurons could average things out, and waves could pass right through each other.
> In fact I think "the hard problem" is a racket.
I am inclined to be in sync with you on this,( to stick with cycles and oscillations).
I appreciated the hall of mirrors metaphor. It’s not difficult to get lost in one’s own mind. I think of it as a group of singers who can form a choir, or become a terrifying noise. I do think the idea of oscillation and wavelength in the brain is compelling. I think the most unique thing about humans is our ability to willfully direct our attention.
The great 19th-early 20th Century actress Eleonora Duse spoke of self- consciousness quite lucidly in her memoir.
“Duse achieved a unique power of conviction and verity on the stage through intense absorption in the character, "eliminating the self" as she put it, and letting the qualities emerge from within, not imposed through artifice.”
There is something about that that resonates with the tone of this article.
You cannot agree that qualia are ineffable, and then just say “but eff that”.
Ha!
I tend to agree with your premise regardless, but (as I just commented to another reply) I think it's much more defensible to point out that the question of whether there is a Hard Problem at all is one that we can only ask properly when we've gotten much farther on the path of characterizing the physical and electrochemical behavior and nature of the structures that correspond to qualia. Otherwise we're getting into Hitchhiker's Guide Deep Thought territory levels of ill-defined questions.
I suspect that there will be Qualia Engineering degrees, or equivalent, long before there's anything close to consensus on the Hard Problem.
Don't let the hard problem people get you down. In all of these debates, it does seem like the internal awareness becomes the stumbling block. If you do some meditation, and have experienced some of these 'thoughts thinking themselves', then the hard problem doesn't seem that hard. And people that haven't will continue arguing that we just don't get the hard problem. It's back to Plato's cave, you can't argue with the people in the cave.
This was amazing article. Would like to have seen more mention of Bayesian Logic, like maybe the oscillations are somewhat 'updating' their priors based on merging oscillations, and our awareness of phenomena is Bayesian. This seems to be giving a bit of a theory to explain the concepts in 'being you' by Anil Seth.
Interesting point. I shall explore the Bayesian angle further.
Some questions, like "why are there beings and not nothing," I expect I will always claim are unsolved, but I the hard problem, at least as I understand it, seems solvable to me, and so I can conceive of, even if I struggle to imagine, hearing an account that purports to solve it and then thinking "this is probably it." So I'd like to understand better why you think that it isn't a real question, or why you think the real part of that question is potentially answered in your post (I'm not sure which one you believe).
As I understand your post, you're telling a story about how a type of physical process in the brain is plausibly the cause of or is conscious thought, and the core case for this is that properties of those physical processes themselves and properties of their relations to each other map onto characteristics generally attributed to qualia. Insofar as this is what the post is doing, I find it interesting, feel that it's quite a bit above my pay grade, and would like to return to it if at any point I become more familiar with modern neuroscience.
However, I don't think this addresses the hard problem, I think the hard problem is a real problem, and I don't think I'm moving any goalposts on you. I think of the hard problem in the context of your theory as "why are oscillations conscious?" or "why is there an experience of the oscillation?" or "why is the oscillation an experience?". "Why, if the oscillation causes or is an experience, is the experience ineffable," on the other hand, seems to be the sort of question this post takes up, and that question seems to take for granted that the oscillation is or corresponds to a conscious thought. The oscillation may very well be a conscious thought or be the correlate to a conscious thought, but is it not a reasonable question to ask how it is that the oscillation is conscious or causes a conscious thing?
When I have a physicalist hat on, I'm thinking of the world as made up of things that aren't conscious. Admittedly, I'm also thinking of a world that's made up of things that aren't individually wet. But when it comes to wetness, I can understand not only that x, y, and z are the conditions necessary and sufficient for this thing to be wet with respect to this other thing, but also *how* x, y, and z cause the phenomenon of wetness. Now, I don't understand that explanation very far down at all. In this case, all I understand is that the wetness of a liquid on a solid is a matter of the ration between the adhesion between the liquid and the solid and the internal cohesion of the liquid, and that the adhesion strength and liquid's cohesion are mostly caused by electromagnetic interaction between the molecules of the two substances. But this is a satisfying "how" explanation to me, even though I don't know the details, and even though I don't know how electromagnetic interaction exists in terms of something that isn't electromagnetically sensitive from the start, because the electromagnetic interaction in this case is a matter of force that can cause motion, and the phenomenon of wetness is easy to analyze into a phenomenon of the motion of the parts of a liquid with respect to a solid surface. So the explanation of wetness, circumscribed in this way, seems unproblematic to me because the phenomenon is clearly about a certain kind of group motion (the group being the molecules of the liquid) and the explanation refers to a cause of motion.
In the case of conscious thought, however, there doesn't seem to me to be this continuity of principle, for lack of a better phrase. I just don't understand how the oscillation is conscious, or where that property comes from. This is in no way an objection to the claim that the right sort of neuronal oscillation is *the* physical source or aspect of conscious thought, even that this is the sort of thing without which there is no conscious thought, and with which there is conscious thought. I don't contest that, and my question to you I don't think has that much bearing on that question. My question is that, assuming it indeed is the physical source or aspect of conscious thought, how is it that it becomes conscious? The parts don't seem to be conscious, and conscious thoughts don't seem to me to be analyzable into spatial objects that can move, so how is it that this system of moving things in space becomes conscious when it is a chain of rhythmically firing neurons?
Why do you think this question doesn't make sense?
(Thank you for the blog post!)
I think the way to bridge the apparent gap between physical configurations of atoms and experience is information processing. Nobody doubts that certain configurations of atoms can process information. I don't think anyone doubts that qualia are some specific type of information.
The question of how qualia arise from physics is absolutely a real question and I am trying to answer it. I just really dislike how proponents of this "hard question" outright define it as "hard" by emphasizing the differences between known physics and subjective experience, and by claiming that subjective experience is so different that it needs to be an entirely separate category of thing where it is not obvious if the natural sciences and known epistemology would even apply. They seem to me to be more busy insisting that there is a problem rather than working on a solution.
Your comparison with wetness is delicious! I am definitely stealing that one. Did you come up with that yourself or can you name someone who came up with it?
I don't think this is that far above your pay grade. You gave a more thoughtful comment than most here.
I wish I could like comments on here, but anyway, thanks for the post and for engaging in discussion. I find discussions of qualia to quickly descend into little more than smug navel gazing (a "racket" indeed!), so I appreciate your grounded approach.
Cui bono? is the biggest problem in all of this. Let’s say we discover the grand secret of consciousness and can showcase your every thought inside an fMRI machine or whatnot. What’s next? How does this benefit anyone other than philosophers?
Isn't that a fully general argument about all science other than applications? Let's say we find a consistent theory of quantum gravity. What's next? How does this benefit anyone other than physicists? Let's say we prove the Riemann hypothesis. What's next? How does this benefit anyone other than mathematicians? Let's say we decipher the Phaistos Disc. What's next? How does this benefit anyone other than palaeographers?
Yes, absolutely it does.
Is this meant seriously? No interest or value whatsoever in truth for its own sake? Or in anything other that what tangibly benifits one's own group?
Do you also wonder why anyone would ever study something unless it will make them lots of money? Or why ever help out a friend if you can't see how it benefits you personally?
The question to every proposal is solely "how does it benefit me?"
More in terms of: "this is why nobody funds it seriously". I'm personally highly interested in the question and care about it but also understand that my wishes are not what the market cares about .
I find it quite demoralizing to read yet another comment trying to force all discussion of consciousness onto the procrustean bed of "easy vs hard problems," as if the basic assumption of the dichotomy is above questioning. In 100 years, I think people will have on average continued the trend of shaking off mysticism and woo.
I'm not forcing anyone to talk about the hard problem. All one has to do is not put an article in my inbox that posits "physicalism" as an alternative to "idealism" because they came up with some speculations about the mechanics of the brain--which rather suggests a lack of apprehension of the topic at hand--and I won't leave a comment. Just say that you're thinking about the mechanics of how conscious behaviour arises in the brain, and ignore the hard problem. Say that it's an impractical problem, or pointless to think about, much like asking where the universe (or the laws that gave rise to it) came from.
But I think a lot of people don't actually believe that. At some level, they are vaguely aware that there's a huge question mark hanging in the air. And if they don't say something about it, then souls, ghosts, magic and Bibles will start pouring through the crack in the door. So they cap off the article with a glib comment about how detailing the physical processes that produce conscious behaviour removes the need to believe in a soul. Sorry, I mean "is an alternative to idealism and dualism."
I agree. There will be less mysticism and woo, and people will be able to think more clearly about this, because they will not be so afraid of stirring up belief in mysticism and woo by doing so.
This is exactly right. If the article hadn't promised to solve the hard problem, then there would be no issue.
Halfway through the article I had a feeling I was wasting my time and and scrolled down to the comments, thanks for saving me from reading the rest of it! People in this community too often conflate finding physical correlates of qualia with giving a mechanistic explanation of how the physical can give rise to the mental.
I think it's not obvious that that's conflation, as opposed to correctly identifying the same thing under two labels.
If the same thing can appear differently, that needs exp!anation.
Problem is that it very often does damn you religion and woo. I myself am such a victim.
Really heartened this is the first comment.
What amazes me is how these people seem to tie themselves into knots to deceive themselves into believing that they solved the problem of consciousness!
I agree that the post does this, but also, the Hard Problem is (often and in part, but not always and only) an assertion of its own unsolvability. Historically, these kinds of assertions get resolved by noticing how and why the idea that seems to result in unsolvability is a confusion. I'm not sure quite what that looks like, of course, but I think any proposed serious attempt at resolution seems likely to have certain properties:
1) Ability to read qualia out of existing minds.
2) Insertion of arbitrary qualia into existing minds.
3) Manipulation of existing qualia within existing minds.
4) Transmission of qualia between existing minds.
5) Determination of which of the claimed necessary properties of a quale can be separated out, or altered.
6) Ability to repeat 1-5 within and between increasingly dissimilar minds (animal or artificial).
The type of approach outline in this post - improved ability to identify and interact with the electrochemical structures in the brain that do, in fact, correspond to qualia in the mind - seems like an essential step towards most of those capabilities I listed, which will let us better map out and characterize the nature and behavior of qualia and discuss what physical systems can generate or support what qualia. Maybe, when we walk that path and get to that point, we'll be able to really say whether there is still a Hard Problem to be resolved at all, but my guess is there won't be. Not everyone will find the solution satisfying, but that fact is not itself problematic.
You're not ignoring the Hard Problem if you say: We're not ready to answer this, we don't have the right tools, but we're building some better ones, and these are the kinds of tools we'll need to get ready to ask the actual right questions."
Random aside: I don't remember when, but I once saw a post somewhere asking what would be the most surprising fact someone from the future could tell you they'd discovered and confirmed. My answer (for a while) has been: "Actually, everyone has the same favorite color-quale, but we differ in which frequency combinations (or patterns of retinal stimulation) our brains map to that quale."
I don't see how my theory would suggest a way to insert arbitrary qualia. It does, however, conceivably lead into a way to disrupt existing qualia, through something like well-targeted transcranial magnetic stimulation, if it is timed correctly to disrupt the rhythm of the oscillation that contains that contains the qualia in question. Interesting angle, I had not thought of that. Thank you.
I agree that I don't have any idea what tools or methods could insert arbitrary qualia, but I think it should be possible-in-principle with sufficiently precise tools to induce patterns of neural firing.
You're right that disruptions seems much more achievable than manipulation or insertion
Recently I had brain surgery and in preparation for that, the surgeons determined exactly where inside the tumor my left-side motor neurons (which the surgeons needed, and later managed, to avoid cutting) were located. They did transcranial magnetic stimulation to induce activity in various neurons. The quale of them hitting the right motor neurons that was an impulse to move my fingers, the strength of which I had to verbally report. I could literally feel in my awareness how they (and the model of my brain they were building) homed in on the right spot.
They said they even found of the synaptic weights of the individual axons involved. They can do that, with a lot of time and money. Good news for Uploading.
That's amazing. And I'm glad it sounds like it went well!
It went well-ish. The tumor was huge and they failed to remove all of it, so radiation therapy and chemotherapy are next.
Right? Light sees light. You cannot scalpel an existence that sings to us. It must be received and that is the greatest evidence. Knowledge can be a powerful tool, here we see the flip side of that power. An uncountable number of humans have had meaningful answers to these questions for millennea, yet so many great thinkers have not built up their strength and tolerance for the brightness.
The "hard problem" arises from the assumption that the difference between qualia and other information being processed in the brain is an ontological one; that qualia are an entirely separate category of thing. I do not share this assumption; I believe the differences (which I enumerated) are explainable by them being processed on their own separate information channel that is internal to an oscillation. So I don't posit an ontological difference on need of an explanation, which would indeed be "hard" if that ontological maneuver was the correct way to go about things - but it is not.
Which basic premise is not internalized? You're making it sound like the premise is that the "hard" problem is just supposed to be not solvable by looking at brains.
Back in the second wave of AI in 1980's, most AI research (at MIT, Stanford, and elsewhere) was done using Lisp programs. It was a standard exercise for advanced undergraduates to write Lisp programs that would inspect and perform operations on their own code and running state, including both internal state and data that symbolically represented external states of the world. These programs straightforwardly fulfilled the description of being "recursively reflective."
I don't for one instant believe that these programs were conscious (unless one adopts a panpsychism according to which lots of things are conscious that we don't normally consider to be, which would contradict the spirit of OP's proposal).
I agree with the OP's basic idea, and I think that we do indeed have to bite that bullet. Those LISP programs were not smart and had only a very limited consciousness, but I do think we will ultimately have to say that they had consciousness.
It's thinking like this that will in the end bring everyone around to my view that consciousness is a bit of a red herring, and actually what makes our human experience distinctive is desire and intention. Consciousness detached from desire is a weird experience, as any Buddhist will tell you. Consciousness (that can't be turned off), coupled with the fact that we want stuff, is the everyday experience we all have.
This is an important implication of OP's proposal to expose: that some Lisp programs commonly written by undergraduates in the 1980's were indeed conscious, to however limited a degree.
Many of the early fans of IIT were turned off by the demonstration by Scott Aaronson that IIT would declare many systems conscious that we overwhelmingly do not normally believe are conscious. In that sense, the theory doesn't seem to fit the phenomenon it aims to describe. My observation is that "recursive reflection" is subject to the same criticism.
Yep, I completely agree with that. But I think the problem is that the word "consciousness" has been a bit mythologised by the likes of Thomas Nagel and David Chalmers. Honestly, half the time I see the word "consciousness" these days, I think it's just standing in for "soul". LISP programs aren't human-like, and they definitely don't have soul. Ergo they can't be conscious.
What the OP is doing, rightly, I think, is attempting to define a more constrained notion of what consciousness is. This should lead to more clarity over exactly what things are conscious and what things aren't, and then on to a better demarcation of what exactly distinguishes us from everything else.
Of course, the OP's definition could be wrong or bad in some way. But I think the project is definitely worthwhile.
The move you're describing simply contributes to the confusion. Whatever problems there may be around consciousness, one doesn't helpfully address them by trying to correlate consciousness to some physical specification that does not in fact match the phenomenon.
I'm all for empirical study of the neural correlates of consciousness. But theoretical claims that this or that physical process (whether IIT or recursive reflection) "just are" consciousness are not merely mistaken but bogus.
I was going to argue, but... I don't think we disagree on anything important. If you're all for the empirical study, and I'm all for the empirical study, then let's let those who can go and do those empirical studies. I also agree with you that there will be lots of philosophical work to be done after the empirical work has been done.
I didn't say the internal information of some conscious thoughts "just is" phenomenal consciousness. I demonstrated it has all (to my knowledge) the properties of phenomenal consciousness that phenomenology hast established.
Yes it does get used as some secularized stand-in for the soul a lot. That contributes to the confusion. That's why I warn in the beginning not to identify with it.
Unlike IIT, Recursive Reflections (RR once it is worth its own abbreviation?) does not claim a thermostat is conscious. It does claim that LORETA could conclusively establish whether (some thoughts of) cows or seaslugs are.
> I don't for one instant believe that these programs were conscious (unless one adopts a panpsychism according to which lots of things are conscious that we don't normally consider to be, which would contradict the spirit of your claim).
You do not need to adopt panpsychism to believe that. You can adopt much weaker claim: execution-of-some-algoritms-psychism.
"Execution-of-some-algorithms-psychism" is already a variety of panpsychism. Every serious version of panpsychism proposes that there are gradations of consciousness depending on the particular properties of the system in question. "Some algorithms" is just a particular suggestion regarding those properties.
No, it's not. What you are describing is panpsychism-light: everything has some non-zero amount of consciousness, and some things based on some principle have "more consciousness" than others.
Execution-of-some-algorithms-psychism, on the other hand, is a weaker claim than that. According to it, only things that execute an algorithm from a particular class of algorithms have non-zero amount of consciousness.
I see this as a quite secondary distinction, almost hair-splitting. You are pointing to the difference between "literally everything" and "lots of things, picked out my criterion, including many things that we don't normally consider conscious." I consider the important bit here to be "including many things that we don't normally consider conscious." If you don't want to call this "panpsychism" because it's not literally "pan," fine; I will concede the terminological point. But my substantive point remains.
If you accept "recursively reflective" as what causes consciousness, you are left the horrible probability that most corporations are conscious too.
What quaila does Microsoft experience?
A relevant paper: "If Materialism Is True, the United States Is Probably Conscious"
http://www.faculty.ucr.edu/~eschwitz/SchwitzPapers/USAconscious-140721.pdf
Yup - If Materialism is True, Microsoft is conscious, the United States is conscious, so the United Nations is probably conscious, and I think we can assume the biosphere more broadly is too. Of course, Earth cannot be the only physical matter exhibiting recursive reflexivity, so it's likely that the Solar System, the local group of stars, and the Milky Way is conscious too. From there, it's save to assume the galactic supercluster is conscious, nestled within the consciousness of the Universe as a whole.
The Universal Consciousness. Hmm - anyone got a name for what to call this?
I think there's plenty of scope for argument there. Firstly, I do think it would be reasonable to say Microsoft experiences qualia, in the form of sales and R&D outcomes.
But I doubt that MS is conscious, because of a speed issue. In order to be conscious, an entity has to be able to have multiple subsystems focusing on (recursions of) a qualium at the same time. The oscillation frequencies for MS are measured in days, as its neurons (people) coordinate to receive information (hold meetings), and their ability to get in sync is quite limited.
The same goes for the USA and other large groups of people.
(That said, I put this argument to Schwitzgebel, and he didn't accept it, so I could be wrong!)
Personally classical theism (or at least The Resurrection) seems more plausible to me than a conscious Microsoft Corp.
And a qualia-experiencing Microsoft seems obviously true if you accept purely materialist theories of consciousness. I'm left at an impasse.
Right - but I don't think it's an impasse at all. It only looks like an impasse because a load of mystical nonsense has been loaded onto the word 'qualia,' so it's now not just the way in which one material system responds to a material stimulus, it's the SuBjEcTiVe eXpErIeNcE of that material stimulus. When that move has been made, any time anyone says, "Oh, look, we can now track what happens in the brain when you see red," philosophical heavyweights can jump out and say, "But of course that's not the qualium of red, the qualium of red is quite different..." They've introduced the ineffable into the argument by definition, hidden inside clever sounding words like "subjective".
But we don't have to accept that framework. You can say, look, clearly Microsoft experiences things (like takeovers and sales). And clearly MS is not like us. So what are the differences, then? And as I say, I think there are still loads.
One is that the kind of qualia MS experiences are very different to what we experience. A company is fundamentally a financial and legal entity, so I think its qualia are limited to the financial and legal things that happen to it. The people inside MS can see colours and feel emotions (er, I think), but MS itself doesn't have the apparatus to do that. So it's living in a very different perceptual world.
Another is that it's absurdly simple. MS has tens of thousands of employees, who I reckon are roughly equivalent to its neurons; our brains have billions. MS is a nematode at best. (Plus those neurons are much slower than ours - you can sent an email about sales at the rate of what? One every five minutes? - our neurons work on the millisecond level.)
And a third is that speed/organisational issue I mentioned: MS isn't organised in such a way that its neurons can reflect on its own perceptions at high levels of efficiency.
If you don't accept that there has to be a spooky element to the definition of qualia, then you can look with much clearer eyes at everything downstream of them. Which is what I think the OP is trying to do.
Writing "subjective experience" with alternating capitalization is not going to make me think that a financial statement written in a book is in any sense the same "thing" as seeing the color red. The neuron that fires when you say red *really isn't* the same thing as the experience of seeing red. If the financial statement is "qualia" then "qualia" is meaningless.
Or... I can "know" that an apple is red separate from "experiencing" redness. A company clearly "knows" what it's profits are if any employees know or if it's written down. How does the company "experience" profit in a way different from "knowing" it? How could we tell?
Sure. By the same token, putting asterisks around really isn't won't make me think that experience and physical events aren't the same. My typographical emphasis - and yours - does not constitute an argument. My argument was in the the four paragraphs that follow.
Look, I don't believe in ghosts or souls or any immaterial stuff. More importantly - much more importantly - I'm methodologically committed to believing that everything is in the material stuff. I think you have to make that commitment in order to do science properly, don't you? Therefore, assuming from the beginning that a neuron firing cannot be the experience is a non-starter for me. I genuinely don't know whether experience consists in neurons firing, or patterns of neurons firing, or the secretion of some chemical, or what. But I am committed to assuming that it's one of those things (or some other physical event in the brain), for the purposes of doing the science (other people doing it, I'm not qualified).
As an analogy, think of the discovery of DNA. They thought it was crazy that all of human heredity could be encoded onto such a simple molecule. But it would have been a mistake to dismiss the discovery based on the an argument like, "a long string of four bases *really isn't* the same thing as a human being".
The word qualia isn't meaningless, it's just not supernatural. Qualia consist in the specific events in the brain that are associated with specific external stimuli.
That probability isn't horrible until you make consciousness some kind of sacred thing, e.g. by identifying with it.
It isn't sacred, it is just a type of computation.
> go to a blog post on consciousness
> ask if it's going to solve the easy problem or the hard problem of consciousness
> he doesn't understand
> i pull out an illustrated diagram explaining what the easy and hard problems of consciousness are
> he laughs and says "it's a good blog post on consciousness"
> read blog post
> it's the easy problem
Not to be too dismissive - it's a really cool elaboration on and testable prediction for the so-called easy problem. Maybe we're going to get to the point where we call them the "Scientific problems of consciousness" and the "Philosophical problems of consciousness". Maybe we'll get to a point where we believe Physicalism is non-falsifiable and let them duke it out with the Neutral Monists, Dualists, Panpsychists, and others over which substance philosophy wins while neuroscientists solve all the useful problems.
I am confused by this easy / hard thing. It seems to have been invented by people begging the question? 'Look, consciousness will definitely never be explained by physicalist theories. So all physicalist explanations are easy, but the real problem is hard.'
If you're ignorant of consciousness, then you're not allowed to rule out physicalism. Especially since physicalism is the most promising philosophy as OP points out.
If you disagree, can you taboo the hard / easy distinction and write specific criticisms of the post?
> can you taboo the hard / easy distinction
"hard / easy" is "explain intrinsic properties / explain structural properties". To have charge, for instance, is to stand in a particular relation to states of the electromagnetic field, and to be an electromagnetic field in some state is to stand in a particular relation to charged objects, and so on, all the way down to raw correlations between sensor readouts. None of this tells us what an electron *actually is*. Structuralism about physics gives us an out here. An electron *actually is*n't anything other than a marker for a particular part of the structure of physics as a whole.
Qualia, on the other hand, have nonrelational properties - redness is not just not-green-or-blue-ness: after all, suffering isn't greenness or blueness either. If you believe in other things with intrinsic properties (the folk-physics notion of "substances" that you can "really touch", for instance) then explaining how those come about is also a "hard problem", it's just that almost no one does anymore.
Can you give an example of any analogous "hard" problem from any field that has ever been solved to your satisfaction, and/or a description of what sort-of-thing would count as a valid solution?
I don't think the hard problem is fully solvable in the way that physical problems are, for exactly the same reason that "why does physics exist" is not solvable. They might *have solutions* - Tegmark's mathematical universe hypothesis could conceivably be right, for instance - we might even find them and all agree that they're probably right, but we'll never really know for sure.
"we'll never really know for sure"
I think this reflects a lack of clarity on what "knowledge" means in this context. What would it even mean to know something "for sure" as opposed to thinking something is "probably right"?
Sure, I'll be more precise. Whereas we can hopefully keep accumulating evidence (in the bayesian sense) in favor of our best physical theories forever, the amount of evidence supporting any particular solution to the hard problem that we will ever be able to access is bounded, and I suspect that the bound is tight enough to matter in practice. We will never reach a point where we're all comfortable rounding it off to "definitely right" in the same way that we (modulo crackpots) are with some scientific theories.
Spinoza's answer to "What is God?" in Ethics.
Which is why nobody has ever argued about what is god after 1677.
I don't think the divide is as stark as @Amicus does. I think that physics experiments really do tell us something about what an electron really is. So do experiments about quasiparticles and virtual particles, which are like real particles in some ways but not others. Given that, I have quite a few examples I see as analogous.
Elan vital. Resolved by molecular biology.
Phlogiston. Resolved by statistical mechanics.
Zeno's paradoxes. Resolved by transfinite set theory and formal underpinnings of the real numbers.
Aristotle's actual and potential infinities. Resolved in part by calculus and set theory and other higher math, in part by admitting there's no actual paradoxical outcomes.
Aether (as medium of light transmission). Resolved by special relativity and the discovery of photons.
For me personally, free will is in this category. Resolved by asking, "What is it I want from the idea of free will? Which of the possible forms of solution (natural deterministic, supernatural deterministic, random) are capable in principle of providing that? Does the world enable such mechanisms to exist? Do I instantiate them?"
I'm unclear what your listed solutions did for your listed problems that this article isn't doing (or at least working towards doing) for consciousness.
First of all, I gather from their listed solutions that they find the Hard Problem to be a distraction and believe that as our understanding of physical laws improves, we’ll realize that consciousness is perfectly explainable.
Second of all, this post is different from the examples above in that it provides no explanation for why the physical processes it describes would cause us to have some first-person subjective experience. While it predicts things that correlate with subjective experience, there would still be a question of why the experiencer exists. With elan vital, there’s nothing else to explain once you understand biological mechanics.
Sorry, I didn't specify in my comment, I'm one of the readers that *does* think this post lays out a plausible path towards a solution for the Hard Problem.
I mean the reason it's considered a hard problem is that it's hard to even imagine how it could be solved.
If you don't know HOW to solve it, that makes it hard.
If you don't even have a criterion for what would COUNT as a solution, then you don't have a well-defined problem TO solve.
Exactly. I do claim that demonstrating why (something with all the properties of) subjective experience arises out of physics does constitute a solution.
> None of this tells us what an electron *actually is*.
You seem to reject physicalism altogether. Seems you are looking for for the Ding an Sich behind phenomena. On the other hand, I believe that's all an electron actually is, to the extent that you cut Reality that way.
We may perceive qualia as non relational; that doesn't mean they are 'nonrelational entities'. It simply means we must explain this perception and how it emerges from simpler mechanisms.
Similarly, you perceive yourself as One, even though you are an aggregate of cells and neural patterns etc. That doesn't mean that you are intrinsically one and indivisible despite physics. How the brain/body perceives itself as one in time and space is something to explain.
> You seem to reject physicalism altogether.
Yes and no, depending on how narrow the sense of physicalism you're using is. I think neutral monism is more plausible than dualism, while eliminativist materialism is somewhere in "there are true contradictions" territory. Whether there's any real distinction between neutral monism and physicalism depends on what you're willing to call physical - if you're a scientific realist in the conventional sense then they're arguably the same thing, if you're an ontic structuralist then maybe not.
> Seems you are looking for for the Ding an Sich behind phenomena. On the other hand, I believe that's all an electron actually is, to the extent that you cut Reality that way.
The opposite, really - qualia are things in themselves, and I want to know how they relate to the phenomenal world of positions and times and masses and charges. I agree that there's probably no electron-essence out there in the universe.
> It simply means we must explain this perception
I agree, although I'm sure we're using perception in different senses - the thing to be explained is not the self-report, it's the correspondence between the self-report and the actual thing being reported. The moon, not the finger pointing at it.
> Similarly, you perceive yourself as One, even though you are an aggregate of cells and neural patterns etc.
Well, no, I don't perceive myself to be anything in the relevant sense. I intuitively believe it, like everyone else, but the thing being perceived isn't "I am a coherent entity", it's "what-it's-like-to-believe-I-am-a-coherent-entity".
Is there such a thing as a nonrelational property? I am doubtful. To enter into by way of your claim on qualia, as nonrelationality being something other than a simple negation of other states, "redness is not just not-green-or-blueness: after all, suffering isn't green or blueness either" .. we would find that rather the degree is complexity, from a simple relation (red, not other inputs of same category) to a complex relation (suffering as related to red via routing through a virtual representation of a self-world model.) In other words, all perceiving, including all potential knowledge, exist within an contextual process. 'The electron in itself' may be a concept that is reifying a proposed decontextualization that is, per the medium of experience, not possible. There's also the experiment comparing minimal perceptions between red #24 and #25. Next to each other, each quale (shade) is perceivable, but without relation, the quale itself disappears, and no longer exists. This would mean quale is itself functionally a process of relation. Or, take the Ganzfeld Effect -- the lack of context dissolves the quale.
Being (as in existing) is a non-relational property.
How could this be so? To be is to relate. There is no such thing as existing outside relation, and in fact, is determined by relation.
Difference is contingent on being, not the other way around. To compare two things they have to have some sort of being.
How sure are you of that? We only know that any given thing exists (or does not exist) by observing its interactions with other things in the same state of being, or extrapolating from patterns of such interactions previously observed. If something were irreversibly causally isolated, https://orionsarm.com/eg-article/5062fd4d7f5cb banished from our future light cone, what distinguishes from it having ceased to exist - or from the banished entity's perspective, what distinguishes that state from the remainder of the formerly-observable universe having ceased to exist?
Abstract things have being, and they interact with non-abstract things. Being is not about "existing" in the concrete sense, but more about a ground-level something that all other properties rest on. Also I am unsure what the particle horizon has to do anything. This isn't about "human perspective", it's about metaphysics. Things have being irrespective of human (or any other) observer. I would posit a counterargument of you looking at an object, then gouging out your eyes. Does this mean that the object is now not able to be seen in general? Has its properties been altered in some fundamental way? Particle horizons arent magic and don't physically affect the objects.
> We only know that any given thing exists (or does not exist) by observing its interactions with other things in the same state of being, or extrapolating from patterns of such interactions previously observed.
Is this the same point as in Erogamer quote below?
> "No, wait," Charles said, once his mind had caught up. "Wait, I understand why you can't ask how fast time is going without being outside of time, but how is my existing or not existing like that? That's an absolute if anything is an absolute! Cogito ergo sum, I think therefore I am!"
> "Yeah?" said the preacher, turning his head again. "You, whoever you are, mister…"
> "Charles," said Charles.
> "Mister Charles, suppose I told you that you, right now, didn't exist. Say, you're only an animation in a Disney movie. How could you tell?"
> "I… what?" said Charles. "If I know literally anything, I know that's false! I have to exist in order to ask whether or not I exist!"
> "Aha!" said the grey-bearded preacher, holding up a finger. "But maybe you don't exist and therefore you're not asking whether you exist. How can you tell the difference? If you were an animated drawing, you'd be saying just the same thing."
> "That… honestly sounds to me like nonsense," said Charles.
> "Exactly, mister gentleman Charles sir! It is nonsense just as you say! Maybe somebody outside of existence thinks that we don't exist, but from the inside of reality we've got no way to know that and no reason to care. It's the same way with Disney princesses. You can say all you want that Snow White doesn't exist, but she can't hear you, so as far as she knows, she exists. She's got just as much proof of her own existence as you have."
---------------
> “You claimed yourself that Snow White exists as much as we do,” said Sonia. “Then why don’t we run into her at the corner shop? Why do I find chocolate biscuits there, instead of Snow White? There is a story in which the two of us meet, and the people within that story have no way of knowing themselves to be unreal. And yet I find myself discovering chocolate biscuits instead. Clearly, there is some factor that the possibility containing myself and chocolate biscuits possesses in greater quantity, compared to the less real possibility containing myself and Snow White. Even if we reply to the great question by answering that nothing exists, some zeroes are greater than other zeroes and quantitative ratios may be established between them. Like any other self-observing structure, Snow White finds herself to be exactly as real as herself, a ratio of one to one, and in this sense her existence is locally an absolute. But to say this does not say whether Snow White is more or less encounterable than other things. There is some quantitative degree to which our universe is looking more towards the chocolate biscuits. Some essence of how much something is observed or experienced, which the chocolate biscuits have more of than Snow White. We could call it quintessence, or propensity, or mana, or the blood of God.”
> “Or bullshit,” Charles said.
> “I suppose that term is as good as any other,” Sonia said. “By whatever name, it is the single, sole, and only truly universal form of money. And we can extend the same logic further. Having postulated the notion of bullshit, it would follow that things are more real only to the extent that they are, in some sense, more bullshit. Or rather, by definition, anything that makes a possibility more encounterable is exactly what we are calling bullshit. Possibilities are experienced by conscious beings in exact proportion to the total bullshit breathing fire into those possibilities. Then to whatever extent a mathematical model is not describing bullshit, it is mere math. Only to the extent that a mathematical structure is about bullshit does that structure form an encounterable part of its universe. It follows that every sapient species, if they investigate the physics of their universe far enough, will eventually find a level at which physics seems solely to describe the arrangement of pure bullshit—some physical quantity whose apparent meaning is making possibilities more encounterable. By your own argument, the presence or absence of bullshit must be falsifiable for anyone inside the universe to notice a difference. Then the eventually-discovered laws will show that variations in bullshit are experimentally observable and cause other events to proceed differently. Indeed, variations in bullshit will be the only causally potent factor. Across every universe with conscious life, any inquisition into physical law, sufficiently advanced, will render everything into bullshit, which is and must be the sole constituent of experienceable causality.”
> “Can you simplify that?” said Charles.
> “No, but I’ll do it anyways,” said Sonia. “Whatever it is that makes one possibility more encounterable than another, we are calling that bullshit by definition. Then whatever bullshit is, if something isn’t made of it, we won’t encounter it. And if we couldn’t detect something experimentally, we’d have no way of encountering it. So bullshit has to appear to us as a quantity in our equations—in fact, the only thing structured by our equations, because we can’t encounter anything else. ‘By convention there is sweetness, by convention bitterness, by convention color, in reality only atoms and the void.’ And underneath atoms and the void, it’s all just bullshit.”
Try solving the Schrodinger wave equation for one electron with no potential field.
The OP is relating the explanation to 16 properties of qualia. Which one is missing or for which one do you disagree with the explanation?
5 is the crux, though I would phrase it as "only partially analyzable" rather than "unanalyzable" - qualia have both intrinsic and relational properties.
Much less importantly, I think many of the others are wrong, irrelevant, or overgeneralizing from typical human mental states:
1. probably false, we don't have perfect introspection.
2. I'm not sure it's even meaningful to talk about a quale's duration in an objective sense. Clearly some *seem* to take longer than others, but that's just another quale!
3. again, who knows?
4. Depends on what you mean by "communicated". Obviously we can refer to them, but we can't offer an exact characterization in completely non-phenomenal terms.
6. I think this is probably tautological, the lack of joint qualia grounds our belief that persons are separate to begin with.
7. Definitely false, we experience things without introspective awareness all the time.
8. More a fact about human cognition than a fact about qualia
9. ¯\_(ツ)_/¯
10. ¯¯¯\__(ツ)__/¯¯¯
11. Unclear what this means. Obviously you can't choose not to experience the thing you're already experiencing, but directing your attention to change what you experience next is a totally mundane process we all do all the time.
12. Category error, qualia aren't the sorts of things that can be "used"
13. False
14. Unclear. Epistemologically convenient if true, but nature never promised us that.
15. Unclear.
16. This has nothing to do with qualia
Im pretty sure all qualia can be distinguished in purely relational terms. This should be obvious from the fact that we can give them different names, just follow the causal chains back from there. For your example, red, blue, and green occur with a position in the visual field, while suffering doesnt.
Well of course all the qualia we can assign different names to can be distinguished in relational terms: naming is a physical process. But finding a difference between two things is not the same as offering a full characterization of either.
How does/doesnt this apply to the electron?
The point I want to make is that "Qualia have intrinsic properties" is basically a bare assertion.
We need to distinguish here between "electrons" as roles in a physical theory and "electrons" as the real objects, if any, that play those roles. Let's call these structure-electrons and substance-electrons. Or, since we don't want to get sidetracked by the question of whether *ontic* (as opposed to epistemic) structuralism about physics is correct, let's move to a domain where it's definitely not, and talk about phonons.
A structure-phonon is a noninteracting boson, with creation operators given by, up to constants, a_k = sqrt(w_k) * (q_k + ip_{-k}/w_{k}), where q and p are positions and momenta (by which I mean only that they satisfy the appropriate canonical commutation relation).
A substance-phonon is a quantized sound wave: the collective motion of atoms in a crystal lattice. And their positions and momenta above are not the positions and momenta of atoms - they're fourier transforms of the position/momentum vectors of the whole lattice.
The properties of phonons can all be defined in terms of each other (giving you a structure-phonon) but what they really are are *relations* between more fundamental physical properties like the positions of atoms. Now step one level down - fundamental physical properties can be defined in terms of each other (giving you the structure of a physical theory), but what are the relata? Where's the "categorical ground" of these behaviors? Physics is necessarily silent on this question.
The epistemic ground, the thing we posit these structures to explain in the first place, are our observations - not things like "the meter read 5 volts" that require a background theory that posits the existence of things like "objects" (not necessarily an explicit one: with the possible exception of newborns, we're all doing tons of interpretational work all the time) but the raw uninterpreted perception underlying them. This exist, clearly, or there would be no need to posit the existence of anything at all. And this is what we're referring to by "qualia".
"Natural nunbers can be studied by counting rocks. But there are negative numbers, there is an intrinsic property of negativity, and you can't study that by counting rocks. That Hard Problem of Negativity you have to leave for us philosophers, which is why we will always have a job, regardless of how good you get at rock-counting."
> "Natural numbers can be studied by counting rocks. But there are negative numbers, there is an intrinsic property of negativity, and you can't study that by counting rocks.
Well, no, "negativity" isn't an intrinsic property, it's relational in the most literal possible sense: to be negative is to be less than zero. All mathematical properties are clearly relational - otherwise something would be lost in translation between formalists and platonists and what have you, and nothing ever is except by human error. But clearly something is lost when we equate the world to the solution to a set of differential equations on a chalkboard - one, to start with, is a physical thing, and the other is not.
Nonetheless this is a caricature of an argument that actually does succeed: you can't study anything beyond the most elementary of elementary number theory by counting rocks. No matter how good our numeric techniques get, they'll only ever be an aid for real mathematics, never a substitute.
There is a sense in which if you believe in the hard problem, you already think physicalism is an incoherent concept. You could call that "begging the question," but it's no more circular than saying I exist because I experience that I exist. I'm not really interested in the argument that I could be mistaken about existing.
In my opinion "physicalism" shouldn't even be a topic. When most people say this word, I think they are thinking of "science vs. miracles and spirits," or vaguely imagining claims that would be similar to interactionism.
This sounds about right to me. Conversely, if you believe physicalism is obviously true, as I do, the assertion that there is a “Hard Problem” kind of just seems like an assertion that you believe the experience of sensations/perceptions is metaphysically separate from the rest of reality (I would be tempted to say, "like some kind of miracle"), and no possible explanation will ever be sufficient to convince you otherwise.
“I think they are thinking of "science vs. miracles and spirits,"” Pretty much. I’m happy to concede that the question why it is like anything to have experiences is puzzling… But it seems clear that that’s just what brains do. I assume it’s a necessary consequence of their design. I can't wrap my head around quantum physics either, but that's not a problem that causes me to think it must be wrong or to doubt the ability of science to make sense of reality.
What I am more certain than I am of anything is that the concept of a disembodied mind is incoherent. Hence no ghosts, Gods, spirits, afterlife, etc. A mind is the activity of a brain; a brain is an information processing device designed by evolution to cause muscles to contract in the right ways at the right times to facilitate survival and reproduction. We have a reasonably clear picture of how, starting from the big bang, the interaction of matter according to the laws of physics gave rise to creatures with brain/minds. Potentially some of these creatures will create artificial minds with technology. There is no other plausible way we know of for minds to come to exist.
Insistence on the hardness of the hard problem does seem to correlate with a skepticism of that understanding of reality, even though technically it may not necessarily be incompatible with it. But it seems to me that unless you’re going to use it as a God of the Gaps argument through which to smuggle, if not God and/or souls, then…what exactly? If “physicalism” could be “not true” in a way that is still consistent with the above picture, then…I don’t really know what you’re talking about or why it would matter.
Some of this stuff does kind of seem like philosophers tying their heads into knots, arguing over how many angels can dance on the head of a quale.
> an assertion that you believe the experience of sensations/perceptions is metaphysically separate from the rest of reality (I would be tempted to say, "like some kind of miracle"), and no possible explanation will ever be sufficient to convince you otherwise.
This is a conclusion, not the premise. The underlying assertion is that experiences are not *definitionally* identical to physical entities. It might be the case that they're different names for the same entities, but if so then this is true in virtue of some contingent fact about those entities, not just something that can be assumed.
> If “physicalism” could be “not true” in a way that is still consistent with the above picture, then…I don’t really know what you’re talking about or why it would matter.
No one is claiming your picture is is false, just that it's incomplete. In the same way that we can ask what "breathes fire into the equations", what distinguishes real physical stuff from differential equations on a whiteboard (which none of our physical theories seem capable of answering, even in principle) we can ask why being is like anything at all, what distinguishes our world from one where things are following all the same trajectories but the lights are out and all the self-reports of experiencing things are wrong.
It seems to me that it must be some sort of emergent property. You can't find "wetness" in H20 molecules, but that doesn't mean either that wetness isn't real, or that physics/chemistry can't account for wetness.
But fair enough. Maybe that's not right. I'm fine with saying it's a weird philosophical conundrum that we don't have a satisfying answer for, and it's conceivable (but by no means certain) that we never will.
But if you go on to postulate that therefore there must be "something else"...then I would want to know: what is that something else? Where did it come from? What is the evidence that it exists? Why does it exist? How does it better account for the observed phenomena?
I certainly don't think we have a complete understanding of reality...but there's what we can know with a fair degree of scientific certainty (which I think is quite a lot actually) and then there's speculation.
So you're an Emergent Materialist - which I think is probably a much better paradigm for examining this particular issue than Eliminativism or Reductionism. If you have two bits, and they are both 1s instead of 0s, the "threeness" permeates the arrangement even though the individual bits couldn't be said to have "threeness". This particular thing should be fairly obvious. Answering "How does subjective experience arise from apparently objective matter" may have a similar answer but it would substantially less obvious.
I think I'm personally going cross-eyed and getting to the point where Panpsychism and Physicalism seem like the same thing, just with weird quasi-religious attitudes about their use.
Physicalism can mean a 'range of things beyond "no ghosts or gods", and none of them is logically necessary. There is no apriori reason why everything should be objective as opposed to subjective, quantitative as opposed to qualitative, relational as opposed intrinsic, etc.
Exactly, there is no a priori reason. The universe doesn't owe us understandability, let alone understandability through science.
It just happens to give it to us anyway.
>A mind is the activity of a brain
Would you refute then that the mind is dependent on what stimulates it? Because it could be argued that the “activity of the brain” is as dependent on things outside the brain as the brain itself.
I’m not contradicting you, I’m very ignorant about this, but I see the mind as the brain’s potential interaction with everything it can be made aware of, and in that sense it is a far bigger and complex space than what’s contained in your skull. Experience can reasonably be seen as crucial a component of the mind as the physical brain itself.
To be honest, I think physicalism is backed up by vibes rather than anything else. It feels kind of scienc-y and futuristic, so people who like to think of themselves as scientific and progressive are drawn to it for that reason.
I think that most such people are scientifically minded and they think philosophy is unproductive, so they gloss over this. They hear: "do you have a little soul inside your brain pulling the levers, or can material interactions explain the divine the wonders of human mind?" And then they write up a reply that says "no, there is no little soul," and move on to the next intellectually stimulating topic.
I'm not even sure what to say to a professional thinker who wants to deny that consciousness presents a problem. Fine, deny it. Rewrite the dictionary so that it says my conscious experience is defined as primary qualities of matter in the Lockean sense. Have yourself a ball.
If you want to argue with a dilettante about this (which I do not), I find that using morals and ethics vis-a-vis the realism of other minds is a better entry point than trying to pin down a definition of "qualia." Why are you mad that I'm about to torture this cat? Because it will set off a chain of physical patterns that are roughly isomorphic to ones that your brain has evolved to be upset about? Surely you should be able to overcome such.... atavisms.
Whether I stop caring about the issue before or after I make you let the cat go seems like a question of economics, tactics, and strategy, or maybe veterinary professional ethics, rather than philosophy or neurobiology.
You only have to regard physicalism as being not necessarily true, ie. being potentially falsifiable. It's odd how many of the pro-science crowd treat physicalism as an unfalsifiable dogma , and therefore an unscientific claim.
Interesting Point. What would falsification of physicalism look like? I'm so committed to empiricism that all falsification I can imagine would have to be empirical, i.e. physical, which could be construed as just another element of (some refined new version of) physicalism.
Do you have a better suggestion, where a replacement of physicalism would not just lead into another physicalism? Something like contradiction between different consequences that physicalism entails, some type of reductio ad absurdum?
A falsification of physicalism would look like a persistent failure to explain some or all phenomena in physical (mostly meaning reductive) terms.
Empiricism without explanation is stamp collecting.
And there's no objective empiricism regarding qualia -- there are no qualiometers.
The hard/easy distinction is entirely valid. "Hard" is used to to indicate a problem which the current paradigm cannot adequately cover without substantially shifting.
A problem will stop being "hard" when we have a framework in which we can posit one or multiple candidate models that can plausibly account for all observations, and all that's left is to hash out their details and figure out the experiments to determine which of them correspond to reality.
A problem will continue to be "hard" for as long as we operate under a framework where we can't even conceive of a superficially plausible answer that still fits into the rest of the framework.
It seems to me the field of physics is in a similar spot facing hard problems without being subject to the same accusations of mysticism as some are making here.
I don't really know what I'm talking about with respect to physics (I'm a psychologist) but have been reading Matt Strassler's "Waves in an Impossible Sea" and I'm struck by how much we don't know while continuing doggedly to try to know more with the threads and shards that we do have.
Gravity and electromagnetism and Higgs "particles" and mass itself are emergent properties of media (media like how water is a medium for waves) that we do not begin to understand or see. Strassler says it also looks impossible HOW we could ever understand or see what these things are emergent properties OF in physics terms given how relativity works.
Physics has been staring at a collection of hard problems that way for about 150 years without blinking as far as I can tell, without pushing the hard questions aside, without giving up on answering them, and without devolving into mysticism.
For all we know, consciousness is an emergent property of some other kind of wave-based field we can't see but that in animals involves neurons. Maybe how humans can only see a tiny part of the electromagnetic spectrum, we only engage with this field in a limited way, less complex-brained animals even less so. I'm not saying the hard problem of consciousness is literally a physics problem; I'm just saying we know how to look at impossibly hard problems without blinking.
In the frame that the hard problem of consciousness is a racket, then is string theory etc also a racket? Or is the fact that math can be constructed around some of these theories at least partially take it out of racket-land?
Racket seems like an epistemic closure kind of way to approach an unexplained phenomenon that universally exists at least among humans. The epistemic closure aspect of this response makes it seem unscientific.
I don't think string theory is a racket in the same way that the "hard problem" is. But you're right, if "racket" isn't going to be just a sui generis category, I should be able to give comparable examples of similar rackets. Excellent point.
I would nominate (as another example of a similar racket) the insistence that we need to "teach the controversy" about intelligent design. Because that also posits that there has to be a problem, and any disagreement, any claim that this isn't really an unsolved problem, is treated as further evidence that there is this controversy that needs to be respected.
Here's four reasons to believe in a Hard Problem.
One can observe that less progress has been made on the HP stuff in practice. The easy problems are related to functioning, behaviour and information processing. We have already made progress in understanding these, including being able to reproduce them to an extent in AI. We expect to continue making progress using much the same techniques.
By contrast the hard problem is one we have not even started on. It's not
that we can build AIs with simple qualia, but we don't know how to build or code qualia into a device at all -- its that we cannot code seeRed(). If we had a reductive explanation of qualia, we would be able to use it as a guide to building qualia.
One can also note that the HP is unique in science, because it is an attempt to explain that which is entirely subjective
In addition qualia cannot be detected -- there are no qualiometers. If qualia are phenomena that are closed to objective empirical investigation, they do not fit within the conventional scientific paradigm.
And lastly, it's hard to see how a theory of qualia could be predictive. It's difficult to see how a linguistic (including mathematical) expression could convey a quale.
I observe only the HP people denying that any progress has been made.
The above is a reductive theory of qualia but it doesn't help you build AIs with qualia. (Except in the trivial case of a particle-level emulation of an entire human brain.)
Subjects are not outside of science, so their experiences aren't either. Unless you're going to define "subjective" in terms of qualia, a ridiculous circular concept that is nevertheless out there.
Predictiveness doesn't require conveying qualia. For predictiveness, it would be enough if the press of a button on e.g. transcranial magnetic stimulation machine merely disrupts qualia in a predictable way.
Subjectivity is outside science inasmuch as science focusses on the objective -- which physics does.
The existence of subjective experience seems difficult to even begin to explain with our current understanding of the physical world - if a physical explanation exists, it’s unclear what it would look like. As an analogy, the ancient Greeks lacked the understanding to attribute lightning to anything other than the Gods, and you could only explain how lightning worked by substantially shifting their worldview.
While we might be able to describe the physical correlates of conscious experience (just as the Greeks could tell you that lightning typically comes when it’s cloudy) and predict when it would come about [easy problem], the explanation of how the thing itself exists might still elude us [hard problem].
I think the "hard problem" is something like "why would subjective experience, which doesn't in itself cause anything else, necessarily accompany (some) thoughts that do cause things"? So I think a solution would look like an explanation why subjective experience cannot not arise from the workings of (some) thoughts.
That's what I'm trying to do here, by explaining why something with the enumerated properties of qualia does have to arise out of the way human thoughts handle information that is internal to them.
Yes blog post was interesting, but misbilled.
Kudos for putting yourself out there with a falsifiable prediction!
Thanks for appreciating that, and all praise to St. Karl. ;-)
This is fascinating! If true, it gives a way to measure whether things are or are not conscious--awesome!
It does not, alas, explain why the synchronized, oscillating firings of neurons feels like anything, meaning that it offers ~zero evidence for physicalism as opposed to something like dualism.
But still, this is real and cool progress. Nice!
I'm still not quite convinced that the theory (even assuming it is correct about human consciousness) provides a universal test for consciousness. Without a deeper explanation of "why [it] feels like anything" then I'm not sure, e.g., whether a computer program in a different substrate -- with synchronized, oscillating firings of neurons -- is conscious. Or another human, for that matter. In either case, the being in question saying "I'm totally conscious right now!" isn't quite convincing (though I admit that in the human case, I'll generally just operate from the assumption that they are).
I think it lets you establish with reasonable certainty which things aren't conscious, which is better than we have now. Currently, the model is something like:
Brain does physical thing X -> ????? -> consciousness
Computer does physical thing Y -> ????? ? -> Consciousness?
"?????" is the answer to the Hard Problem. We know Y because we can see everything the computer does. But we don't really know what X is.
Suppose X is the brainwave thing this post makes the case for it being. Then we can compare X and Y. If they're structurally very similar, then we've learned something, and we can go on worrying about substrate dependence. But if they're dissimilar to the point where the brain's version of Y wouldn't cause consciousness, then we can be pretty sure the computer doesn't have humanlike consciousness.
Heck, you could even run the same comparison with insects, and finally settle (or throw gasoline on) that EA debate.
There's always a chance we're missing other types/generators of experience until we solve the Hard Problem. But this seems like an improvement.
I don't claim this explains whether computer programs on every possible different substrate would be conscious. But I do think a full atom-level simulation of a human brain would exhibit computations functionally indistinguishable from what we experience.
Yes it does explain why it feels like something. The feeling like anything is what (human, slow, oscillatory) thinking looks like from the inside.
Brain imaging can (now) show us how we think. Consciousness is what it feels like to be thought.
Why does it look like anything from the inside? Why aren't there just brainwaves mechanically oscillating and computing without anyone to notice? The hard problem asks why there is an inside at all in the first place.
If you say "anything that affects neurons firing in the brain is of course going to affect experience. We have found a process happening in neurons that explains why we feel what and how we feel!" then you have indeed made a great discovery! But the Hard Problem is swept away unexplained by the words "of course".
The reason there is an inside is precisely because thoughts handle internal information processing differently from the handling of information received out of sync. This is not unexplained anymore.
The software I work on at my job handles internal information processing differently from information that it receives out of sync with its cycle. Perhaps I'm mistsken, but I don't think you would say this is sufficient to conclude that my software has awareness or experience.
No I don't think your software has awareness.
I think your awareness processes information. And when it does, it does it in a workspace - not a Global Workspace as in GWT, but a local one inside an oscillating thought.
That is not the only way of information processing (unconscious thoughts can do process information) so there's no need to assume anything else is conscious... at least until someone builds a variant of LORETA that can find oscillations in cows. Some of the animal welfare EAs should be interested in that.
It does explain that, actually. I do not see any property of "what it feels like" that is not captured by this explanation. Do you?
I think you captured pretty much all of what it feels like, and that's why I'm very excited about this work! The piece of the explanation I claim is missing is *that* it feels like anything.
Here's a question for you: Do you think everything in the world is conscious? Ie, that it feels like something to be a rock, a washing machine, a tree, a submarine, a storm, or an office building? None of these things? Only some of them?
Your theory accounts for the properties of human thought. I think your theory could observe that none of these things have similar properties to the brain, and so they have different properties. Eg, suppose the sailors on the submarine move in a way that processes the information of the orders they were given. Because submarines are cramped, they have to move through certain hatches one man at a time. From this, we can conclude that submarine qualia (if they have any) cannot be simultaneous. Submarines experience only one thing at a time.
Others might say: "I deny that submarines have experience. It's ridiculous to claim that people moving inside a submarine cause qualia, even a very exotic sort of qualia. Sure, molecules and EM waves moving inside the brain must somehow cause qualia, but brains and submarines seem very different and I suspect there is something about brains that submarines lack, such that it feels like something when a brain's particles move, but it doesn't when a submarine's sailors move."
I think your answer to this person must take one of 3 forms:
1. Submarines are indeed conscious because everything is conscious. Our theory explains which physical processes are associated with your experience; they may or may not have any applicability to a submarine.
2. Humans are conscious and submarines are not. We know this because we understand exactly how conscious experience results from the movements of molecules and fields. Motion normally happens without any experience, but our theory says that when some system has property X, consciousness arises by mechanism Y.
3. Neither submarines nor humans are conscious; consciousness is an illusion.
Some people legitimately claim 3, but it is wrong; I am conscious. 1, some form of panpsychism, is a very plausible answer that side steps the hard problem. I would accept 1, but you have not said anything that makes me believe this would be your answer. 2 is a complete answer only if you have both X (the property that causes consciousness) and Y (the mechanism by which motion generates experience). You have offered X, but not Y.
It may be tempting for a reductionist to say that Y is somehow just a macroscopic property od systems satisfying X; IE, that consciousness is emergent, and reduces to the motions of molecules. I think this is not a complete explanation. Reductionism has been very successful at explaining momenta and forces of macroscopic objects in terms of the motions and forces of smaller objects. It has shown convincingly that tenperature, a macro property that seems totally unlike motion, nevertheless reduces to the motions of tiny particles.
Consciousness--feelings--seem totally unlike motions and forces. But maybe it's like temperature and actually reduces to them anyway. If it does, then an actual reductive explanation should be able to answer questions like: What are the dimensions and units of feeling? Are they derived units, or fundamental? If I know X property of a system, is that enough to know anything about feelings? If it is, what calculation can I run on the particles of this system to learn about the feeling? What is the theoretical basis for believing that to be the correct calculation? Is there a toy model of a minimal conscious system that we can understand?
Unless you can answer questions like these, you haven't actually reduced consciousness to motions and forces. And so it remains an open question whether it in fact reduces or not! Maybe instead of being emergent from a system, there is a consciousness-particle that interacts through the electric field and a fifth fundamental force, the force of feeling. Or maybe there's a universal experience-field that brains just interact with especially well. Etc.
Showing that properties in the brain align with properties of qualia is genuinely interesting and great work. It shows us which processes in the brain are associated with consciousness. And in particular, I think it convincingly shows how self-reflection works! Anything lacking a brain thay can sync in this way is not going to be self-aware in the sense of being able to reflect on its own thoughts--that's awesome! It's an incredible breakthrough and I'm very excited about it!
But we have experiences that don't require self-reflection, like, say, the sensation of pain. The Hard Problem of Consciousness is about explaining these primitive, non-self-reflective sensations. Having a solution to the Hard Problem means being able to answer questions like "do ants feel pain? Do rocks or trees feel anything at all?" I would love to be proved wring, but I don't think your explanation can answer this type of question.
My answer is 2, and I contend I have given 2 but it is about the movement of information, which is implemented in the movement of electricity and ions, but also a few of the molecules you seem so keen on.
Qualia reduce to information.
Pain doesn't require self-reflection, but it does require neuronal information processing: that's why general anaesthesia works.
Thank you for writing.
Sorry, but that can't possibly work. If you take qualia seriously as not reducible to mere emergent physical properties than Chalmer's zombie argument shows why it's in principle impossible to experimentally test a theory of qualia.
Yes yes there are all sorts of caveats to that statement. You can certainly test a unified theory of physics and experience and maybe that gives you some extra information but they can't be tested the way you suggest.
> If you take qualia seriously as not reducible to mere emergent physical properties than Chalmer's zombie argument shows why it's in principle impossible to experimentally test a theory of qualia.
This is true, but also not as unusual as it sounds. All "experimentally testable" claims, with the somewhat ironic exception of claims about raw sense data, require supporting theoretical infrastructure - you don't test "the mass of an electron is X", you test "my experimental apparatus will produce a readout of X when I do Y, which if [background theory] is correct means the electron mass is X".
Any such background theory for qualia is going to have to assume that self-report is reliable under at least some circumstances, at which point we're back to a problem which is merely very very very very hard, not impossible.
The nature of the hard problem is to identify what additional laws of nature are necessary to explain why we actually have qualia not merely say we do.
Few people doubt that we can explain at a microphysical level what causes us to say that we are conciouss the problem is to explain why we actually are. The whole point is to explain what laws connect up physical with the mental and if you assume the physical world is causally closed - no soul comes in to alter physics in the brain - that's empirically untestable.
And yes, one might assume that we are in fact having the experiences we say we do but because we are no longer asking for what physical state is necessary to cause (or at least predict) any physically observable outcome you can't falsify such a theory with empirical predictions.
What one can do is try and offer a theoretically attractive fundamental theory which unifies physical predictions and experiencal ones (taking our self-reports as correct) and show it agrees with physical predictions but it's still not falsifiable by physical observation.
---
And yes the apparent fact that qualia don't -- in some sense -- make a difference to physical outcomes raises questions about why self-report is correct. Hence why most people attempting a partial solution have wanted to suggest that certain computational states necessarily give rise to certain experiences -- so systems which report they have qualia X really tend to have it but it's a bit of a paradox.
Hence why the term hard problem is apt.
I'm familiar with the problem, my point is that the "easiness" of physical problems is not a given. If the philosophical background radiation of our culture endorsed subjective idealism instead (and there can't be any direct empirical evidence for or against it vs. physicalism, for the reasons you note), for instance, then the physicalists would be the ones bashing their heads against the wall trying to explain that when they talk about "objects" they're not just referring to structural relations between quales. (Arguably you actually see a little of this in debates over scientific realism.) The counterpart of the hard problem in such a world would be something like "why do experiences have causal structure?", and the equivalent of naive illusionism would be something like " 'huh, it seems like my experiences have causal structure' is just what it feels like to be in a particular sort of mental state"
I am ultimately am sympathetic to the idea that it's a mistake to take physical claims not the qualia as fundamental. I mean at some level I'd argue they are epistemically prior. But that doesn't really help in this case.
You are correct that if someone offered a single theory that made predictions about qualia and unsolved questions of fundamental physics thst would absolutely be important evidence for the unified theory. I tried to indicate that but may have been unclear.
But however you view it (as physical regularities or regularities in qualia representing experimental outcomes) we have theories (the standard model) that are damn fucking good at predicting stuff and that we are damn sure will predict what happens at the brain at the level considered here. So the proposal here can only be wrong or isn't given further support by those tests.
Logically speaking predicting what happens in the brain isn't a hole in our fundamental theories just an issue of computational power/descriptive detail.
Unless you are claiming it's merely a question of consistency and it's not truly an empirical test but a question of mathematical compatibility (is this theory compatible with what we know about physics)? But the parts of the article before this make me doubt that this is the claim.
> Unless you are claiming it's merely a question of consistency and it's not truly an empirical test but a question of mathematical compatibility (is this theory compatible with what we know about physics)?
Pretty much, yes. All a physical theory not equipped with a choice of interpretation gives us is correlations between bits of sense-data, and formalism alone doesn't even give us that - you still need to decide which parts of the formal structure are "measurements".
Not sure what article you're referring to, I'm not the author of the piece we're commenting on.
I didn't mean to suggest you were. I was saying that my criticism of the original claim of the author's to be offering an empirically checkable theory was justified. I was assuming my remarks were contextually restricted to be about that and may have been a bit sloppy as a result.
I completely agree with the point you just said here. And it's one of the big mistakes that advocates of no collapse QM interpretations make. They assume that it's a simpler theory because they can dispense with any need for an extra law to handle collapse. But -- while no collapse may or may not be true -- once considered in the full light of the need to specify what parts of the theory correspond to experiences you pay that price no matter what.
However, I just don't think it's relevant to the article above since the kind of empirical confirmation they mention doesn't correspond to any plausible uncertainty in how to identify measurement with experience. If they'd offered a claim about the role of collapse in QM which stemmed from a theory about how to identify parts of the formalism with experience sure. But that just doesn't come up at the level of description their experiments and theories occur at and I felt my response was already too long w/o me including those caveats ;-).
> All "experimentally testable" claims, with the somewhat ironic exception of claims about raw sense data, require supporting theoretical infrastructure - you don't test "the mass of an electron is X", you test "my experimental apparatus will produce a readout of X when I do Y, which if [background theory] is correct means the electron mass is X".
That's the Duhem-Quine thesis (except for the bit singling out sense data as different).
Yes. I don't think the correlation between naive verificationism and "the hard problem is a racket" is a coincidence.
I agree, it is not a coincidence. And I don't like you putting "naive" in there. Standard epistemology does apply to consciousness.
By naive verificationism I mean the sort of contemporary internet "positivism" found in rationalist circles, which is neither remotely standard nor particularly similar to verificationism in the sense of the early logical positivists.
> If you take qualia seriously as not reducible to mere emergent physical properties
If you assume that qualia are by definition non-physical, then, unsurprisingly, you going to end up thinking that there can't possibly be a physical explanation of qualia.
This, of course, is a great reason not to make such question-begging assumptions in the first place.
Yes but the usual reason to talk about them as a theoretical entity is because you think you need that additional level of description -- it's at best extremely misleading to use the term differently without clarity. Or, to put the point differently, if qualia logically supervene on the physical then there isn't any extra fact to specify over and above the physical theory so why complicated things by introducing unnecessary terminology? As such and based on the rest of the piece I think that's an appropriate way to understand what they took themselves to be describing.
And half the assertions in the piece about the nature of qualia only make sense if they understood it as not being logically supervenient on thr physical. They are only private if you mean by qualia something different than "a physical system satisfying such and such configuration." They clearly had in mind the hard question of consciousness not some easy sense and so it's fair to judge their theory on those terms
If they'd instead wanted to argue for a physical reduction Dennet style they could have done that but that makes even less sense of a claim of empirical falsifiability.
> if qualia logically supervene on the physical then there isn't any extra fact to specify over and above the physical theory so why complicated things by introducing unnecessary terminology?
Trees logically supervene on the quarks and yet we use the term "tree" regardless. Clearly there is a reason to use separate words for things which are "merely physical".
> And half the assertions in the piece about the nature of qualia only make sense if they understood it as not being logically supervenient on thr physical. They are only private if you mean by qualia something different than "a physical system satisfying such and such configuration."
So you might have though. But apparently it's not the case. Here the author of the post shows how qualia can be private while still being physical. You are free to engage with his arguments of course, but you can't dismiss them just because you smuggled the notion of qualia's non-physical nature in the definition.
> They clearly had in mind the hard question of consciousness not some easy sense and so it's fair to judge their theory on those terms
If you beileve that explanation of privacy of qualia is part of the hard problem of consciousness, then the author has just solved some part of the hard problem of consciousness, even though previously it appeared absolutely impossible from your perspective. Feel free to make the obvious conclusions what it means for the rest of the hard problem. Or at least notice how you mind shifts the goalpost further.
You can emphasize the differences between subjective experience and things made out of atoms, and end up with a hard problem.
Or you can emphasize the similarities, such as the information processing, and end up with something explainable.
I prefer the latter.
This isn't the kind of thing you get to just prefer one answer to another because no one is trying to say they aren't similar — just that you can't derive facts about experience purely as a may of logic alone from physical facts with no extra assumptions. And for that task it's enough to show that there is any way in any conceivable situation they don't perfectly track each other*.
Besides, if you actually want to avoid having to deal with the hard problem the burden is really on you to give a positive argument showing that the existence of an experience follows purely by logic from a physical description. The zombie argument tries to show that no such argument is even possible but if you want to avoid the hard problem the burden is still really on you to make the positive argument.
—
*: It's analagous to arguing that — even though in our universe gravitational and inertial mass are always the same — that the two notions aren't merely different logically equivalent descriptions of the same thing. For that it's enough to say: but you could imagine a world in which how hard it is to accelerate something could vary independently of how much it attracted/was attracted by other objects.
Doesn't make Einstein's analysis that sees acceleration and gravitation as two sides of the same coin any less correct or influential but it tells us that this analysis is a substantive law of nature and evidence for it should be evaluated the same way we evaluate other laws of nature.
Yes the burden is on me. I think I have carried that burden. If you disagree, please specify what part of the burden you think remains uncarried.
To meet that burden it should be possible to write down a physical description and as a matter of pure logic (only math no assumptions) tell me what experience it represents. I see no argument that is possible.
> If you assume that qualia are by definition non-physical, then, unsurprisingly, you going to end up thinking that there can't possibly be a physical explanation of qualia f
If. Not al! arguments work.that way., EG. chalmers.
Feel free to present this particular argument, so we could engage with it, instead of vaguely pointing towards the author, whose most famous argument - p-zombies - begs the question in exactly this way.
Chalmers does not define qualia as non physical. The p zombies argument might imply that, but that is still not a matter of definition.
This whole rhetorical approach has the problem that the peop!e making it can't distinguish between definitons, theories, conclusions, etc.
P-zombies argument appeals to non-physicalist intuition as its premise, which is justified only if physicalism is false which is the whole question being investigated:
Physicalism is false -> Zombie World where everything is exactly the same but humans are not conscious is logically possible ->Physicalism is false.
The anti zombie argument appeals to physicalism intuitions as it's premise physicalism is true -> zombies are impossible.
But Chalmers' zombie argument is trivially false: https://www.lesswrong.com/posts/7DmA3yWwa6AT5jFXt/zombies-redacted (this is not some argument from authority, I just agree with Yudkowsky's argumentation).
You can go read the book of you want but Yudkowsky is alot less smart than he thinks he is -- there is a reason that even the philosophers of consciousness who disagree with it take it quite seriously. Here Yudkowsky is the idiot going "see look I disproved quantum physics."
If you want the first point Yudkowsky goes off the rails it's with his claims about causation. Chalmers assents to the claim that there is a natural law relationship *in our universe* between your qualia and your physical states. I don't remember if he calls it a causal relationship but I sure will and that's perfectly coherent with the zombie argument.
The fact that you can imagine a universe with different natural laws which breaks the law like relationship doesn't show there was no causal relationship in your world so doesn't establish that yo didn't know it.
Quantum physics makes predictions which have been verified. Has Chalmers' zombie argument done that?
No, I'm rejecting the analogy between quantum physics and Chalmers' theory. That doesn't make me a "troll". If you're arguing against Yudkowsky's critique of the zombie argument, you can't just assume your conclusion about that argument and then use that to dismiss anyone else who doesn't share your take.
That's fair..I deleted the comment because I realized you likely understood the analogy differently than I meant it which wasn't about the degree of confidence in the theory.
I'm sorry, I shouldn't have presumed you were understanding it the same way I was.
No, it might very well be wrong. And I don't begrudge Yudkowsky for arguing against it (even if he made some mistakes there). I do think he failed to constructively respond to the many good counterpoints raised but that's just what being wrong sometimes looks like.
My intent wasn't to suggest what he was doing was unreasonable (hell I don't think the guy who thinks they've disproved QM is necessarily unreasonable just not an authority for others) only that it is unreasonable for someone else to presume Yudkowsky's argument is correct absent having throughly studied the material themselves.
Relative to you he has the position of the lone doubter with the bad track record challenging the established view. What doesn't make sense is citing Yudkowsky as if he was an authority.
"If you want the first point Yudkowsky goes off the rails it's with his claims about causation. Chalmers assents to the claim that there is a natural law relationship *in our universe* between your qualia and your physical states. I don't remember if he calls it a causal relationship but I sure will and that's perfectly coherent with the zombie argument."
I don't really follow what claims about causation you mean or why they're a problem.
"The fact that you can imagine a universe with different natural laws which breaks the law like relationship doesn't show there was no causal relationship in your world so doesn't establish that yo didn't know it."
Having a rough time parsing this as well - didn't know what?
Maybe it's fair if I extract the main point of Yudkowsky's argument here instead of just linking to a long blogpost. AFAIU the main point is simply that the zombie argument can be summarized as follows, and that this shows the argument to be incoherent:
"
- Matter has additional consciousness-properties which are not yet understood. These properties are epiphenomenal with respect to ordinarily observable physics—they make no difference to the motion of particles.
- Separately, there exists a not-yet-understood reason within normal physics why philosophers talk about consciousness and invent theories of dual properties.
- Miraculously, when philosophers talk about consciousness, the bridging laws of our world are exactly right to make this talk about consciousness correct, even though it arises from a malfunction (drawing of logically unwarranted conclusions) in the causally closed cognitive system that types philosophy papers.
"
The relevance of the causal talk is that Yudkowsky argues that we can't know that qualia exist because they don't have any causal impact. It's roughly the same as your claim that the qualia are epiphenomenal.
And my point is that there is a subtlety that's being missed here. What Chalmers claims is that it's logically possible that **if the laws of nature were different** all the physical stuff behaves exactly the same and this isn't the same as the claim that qualia have no causal power and even if you stipulate that it doesn't call into question the knowledge claims about qualia like you suggest.
--
For instance, one attractive principle is that we can claim to know X is true if there is a reliable connection between what we observed and the truth of X. For instance, suppose that physical law guarantees that whenever I see the color red a blurg is nearby. Prima facia that's enough to justify my claim that I know a blurg is there when I see red.
And qualia are connected via exceptionless natural law to the physical states of our brain so seemingly the same justification works there.
Now maybe you want to say -- but how can we know there is the right kind of natural law? But this demands too much because there are infinitely many theories consistent with our observations and at some point we just have to appeal to our priors (attempts to declare some universal prior have been disastrous failures and we have no clue how we would rule out some priors but not others much less this one.
--
More generally. I think it's wrong to even conclude qualia are causally impotent from the fact that every physical effect has a physical cause.
To illustrate what I mean, consider the question of whether "gravitational mass is causally impotent". Now it's certainly true that because our laws of physics say inertial and gravitational mass are always equal I can rewrite the laws of physics in an equally predictive form that only mentions inertial mass. Of course, the new way I've written the laws of physics would fail to hold in a world where those two values aren't equal but it doesn't show what we might intuitively think of as the claim "inertial mass is causally impotent". After all, I could have played the game the other way.
Same deal with qualia. If there is a natural law in our universe that says such and such qualia is realized iff such and such physical state is realized I can redescribe the laws of physics in an equally empirically valid way for our universe replacing the role of those physical states by the qualia. Or, to put the point another way, if physical law says X and Y always co-occur then just because you can write the laws without mentioning X doesn't entail X has no causal power.
>Now maybe you want to say -- but how can we know there is the right kind of natural law?
I dont think Yud is commited to any account of "knowledge", just belief, truth, and subjective confidence. Am I right to translate the following paragraph as "If someone says they have qualia I belief they have just those qualia purely on priors."?
>To illustrate what I mean, consider the question of whether "gravitational mass is causally impotent".
This seems like a Motte-and-Bailey. Noone cares whether you draw a distinction between the two or call it all just mass. Just like noone cares about the distinction between "yin-charge" (influencing the strength of electrostatic force a particle experiences) and "yang-charge" (influencing the strength of force a particle exerts on others) that I just made up. The only reason it would matter for qualia is that we *feel like* the qualia would have some kind of additional effect that the physical states alone dont - but that would make the argument inapplicable.
Detailed response below but basically I can take the same form of argument to show we can't possibly know about an object's gravitational mass.
1) I can rewrite all the laws of physics in an empirically equivalent form by just using inertial mass everywhere we use gravitational mass because it's a natural law they are always equal.
2) In principle it's coherent to imagine a world where the gravitational constant was 0 but there was another kind of force that works like gravity called smavity. Now objects can have all sorts of smavity charges but it's possible to have a universe where the gravitational constant is 0 but objects all just happen to have a smavity charge equal to their inertial mass.
So it's possible to have a universe where gravitational mass is different than inertial mass but is empirically identical (because in that universe smavity happens to play exactly the right role.
3) Therefore gravitational mass could be different without there being any way for us to tell so we must not have any justified belief or knowledge in gravitational mass.
But of course that argument is wrong. We can know about gravitational mass because in our world it absolutely does have causal power and it's irrelevant what we can imagine about a world with different laws.
Same problem here. In our universe the natural laws do connect up qualia with physical effects. Since you can't have brain state B without quala Q as far as natural law is concerned everything caused by B is equally caused by Q. Doesn't matter what I can imagine about another universe with different laws that connection lets me have knowledge of qualia.
--
Call it what you want. He used the word "know" so I followed but same argument.
And I wouldn't say just based on priors, I'd also say because you have the experience directly because I'm not willing to identify 'you' with a brain but we can bracket this, but also because it's kinda misleading.
I mean ultimately, how do we know there are physical objects or an external world at all? We could be Boltzmann brains or idealism could be true. Ultimately, we just apply our priors and if those priors actually get the correct answer in a reliable fashion (ie, if there is a natural law which in fact guarantees every time you see A that B is true and your priors say that you are justified in believing B when you say A and you do see A and believe B we call it justified). Nothing different here.
---
All the rest is just a means to point out you can't treat qualia somehow worse because they aren't physical by appealing to the causal closure of the physical. Yudkowsky can't argue that because we think the physical world is causally closed that therefore qualia don't cause your mental states. Natural law can still say, you have mental state X iff you had qualia Y a second ago (and some list of other facts).
Besides it's silly anyway because we claim to know things like "the universe isn't filled with cheese past the cosmic horizon" even though it's causally isolated from us (in a way qualia are not).
--
Importantly, it's not necessary this argument convince you that qualia exist. Yudkowsky is arguing that it is incoherent to claim to know you have qualia while believing they don't supervene on the physical and he hasn't shown anything like that because it's both false that we can't have justified belief in things that don't causally affect us *and* it's false to say that the zombie argument entails qualia are causally isolated.
>The relevance of the causal talk is that Yudkowsky argues that we can't know that qualia exist because they don't have any causal impact
No, Yudowsky argues vehemently that reports of consciousness are caused by consciousness, at whatever level of indirection.
"The relevance of the causal talk is that Yudkowsky argues that we can't know that qualia exist because they don't have any causal impact. It's roughly the same as your claim that the qualia are epiphenomenal."
If you see Yudkowsky arguing qualia don't have causal impact it is surely in the context of Chalmers' arguments and with the goal of showing Chalmers is wrong. I'm not sure if that's what you meant.
"And my point is that there is a subtlety that's being missed here. What Chalmers claims is that it's logically possible that **if the laws of nature were different** all the physical stuff behaves exactly the same and this isn't the same as the claim that qualia have no causal power and even if you stipulate that it doesn't call into question the knowledge claims about qualia like you suggest."
Imagine the laws of nature changing (in the zombie universe) to somehow remove qualia, yet all physical stuff staying the same, including people like us arguing about qualia online. In the zombie universe, we are arguing about qualia for no reason at all, yet somehow it follows from the qualia-less physics that we do so anyway. *If* indeed we imagine that our zombie twins talk about qualia without having qualia for a reason within standard physics, then that reason would be sufficient to explain us talking about qualia in this universe as well, for our universe's standard physics. Yet at the same time proponents of this argument say that there *are* nevertheless qualia in this universe, *and* that our talking about them is somehow related to that fact. Both of these are plain absurdities.
"More generally. I think it's wrong to even conclude qualia are causally impotent from the fact that every physical effect has a physical cause."
To reiterate, we (Yudkowsky and I, if I may presume to speak for him for a bit) don't claim qualia are causally impotent. We claim that *on Chalmers' view* they are causally impotent, and therefore we reject that view.
"Or, to put the point another way, if physical law says X and Y always co-occur then just because you can write the laws without mentioning X doesn't entail X has no causal power."
I don't think this is quite right, when we're talking about physical laws. If you rewrite the laws so that you don't mention X, yet what you write is still a complete and accurate description of the laws of physics, then you are actually thereby still mentioning X (and the causal power implied by it), just under a different name, in a different mathematical formalism, or whatever.
Look, I agree that there is a deep weirdness that Chalmer's view entails in that it implies that there are all sorts of ways that qualia could have been related to the physical world that didn't necessarily let us report about them at all. First, maybe there is just a really fucking compelling theory of everything (physical and qualia) that implies this holds. We can't reason from our lack of knowledge of such a theory.
Secondly, the temptation to view that we obviously problematic rests on the attitude that somehow the physical comes first and then the mental is layered on top of it. I'd argue that gets it exactly backwards. Epistemically, what we start with is the qualia and to the extent we have evidence for the existence of the physical world at all it is as a result of regularities observed in qualia.
From this POV the question becomes not "why do qualia depend on the physical world in just the right way for us to report correctly about them." Indeed, that becomes kinda non-sensical because on this view "we" are patterns of qualia and the physical world is just an explanation of said patterns. At best the question dissolves into "why does the world have coherent patterns of qualia that create the impression of memory etc etc " but now we can answer it via sn anthropic argument -- if that wasn't true we wouldn't be asking.
--
Also at a deeper level I personally also reject Occam's razor for natural law -- well not fully reject but at least think we shouldn't expect laws to be particularly simple for qualia. I think it only appears to be true for physics because we make an unprincipled distinction between initial conditions and dynamics when they aren't in principle different. It's really the kolmogorov complexity of the whole thing that matters but for anthropic reasons we will find a description that makes the dynamics simple wrt physical laws but not experience laws so we shouldn't be surprised if those laws look way less simple and elegant.
> Matter has additional consciousness-properties which are not yet understood. These properties are epiphenomenal with respect to ordinarily observable physics—they make no difference to the motion of particles.
This is not quite right. Chalmers' position is that phenomenal properties are *explanatory unnecessary* for physics:
> On the question of explanatory irrelevance of consciousness to physical actions. It is true that I bite this bullet, in a certain sense. But the view can be made to seem stranger than it is. So I should make some clarifications.
> First, I do not say that consciousness is causally irrelevant to action. That is a question I am neutral on, and I think there are interesting views that give consciousness causal relevance without doing any damage to the scientific world view. (For example, by making consciousness correspond to the intrinsic aspect of physical states which physics characterizes only extrinsically.)
> Second, I do not say that consciousness can never be used to explain action, so that explanations that involve consciousness are invalid. I simply say that invoking consciousness is not necessary to explain actions; there will always be a physical explanation that does not invoke or imply consciousness. A better phrase would have been "explanatorily superfluous", rather than "explanatorily irrelevant". Something can be superfluous and still be relevant.
> Third, it isn't true that my view implies that "if you think you are reading because you consciously want to read, you are mistaken". On my view, it is very likely that you are reading because you want to read. It is just that fact that the wanting is consciously experienced is not required for the explanation to go through. Conscious wants can explain actions, and nonconscious wants can explain actions too. Similarly, you drink because you are thirsty, but the consciously experienced aspect of that thirst needn't be appealed to to explain your action.
https://consc.net/books/tcm/searle-response2.html
Interesting, thanks for the link & quotes.
The next to last paragraph *almost* sounds like it could apply just as well to other high-level concepts that have a lower-level explanation in physics. Fire, for example, is explanatorily superfluous in just such a way, or rivers, or hailstorms.
*Almost*, because of course he says: "there will always be a physical explanation that does not invoke or imply consciousness"... Unless he has a solid argument for this claim, it really just turns the whole thing into a case of circular reasoning. Claiming there is always, in this universe, a physical explanation of consciousness-related utterances that does not in any way imply consciousness, is quite a bit stronger of a claim than the usual "we can conceive of an alternate universe where this is so".
Chalmers does not argue that zombies are really possible in our universe.
Yudkowsky misunderstands him.
There is also the Mary's Room argument.
I think you're misunderstanding Yudkowsky, I don't see him as claiming Chalmers argues zombies are really possible in our universe, at all.
(let's keep other arguments out of it until we get to the bottom of this one ;P)
Well, Chalmers thinks Yudkowsky is wrong about Chalmers https://www.greaterwrong.com/posts/fdEWWr8St59bXLbQr/zombies-zombies/comment/5qKe5gQ8HWgfRq9Dw
That does not change the fact that it's wrong to imply that Yudkowsky thinks "Chalmers [argues] that zombies are really possible in our universe".
Correct.
I don't buy the zombie thought experiment.
Phrasing part of the theory as "humans aren't conscious" seems to be a needlessly edgy and provocative way to state something that most people regard as common sense, that the human mind is made of parts and that some of those parts are conscious while others are not.
Similarly, in 15c I'm not sure that saying that conscious thoughts "mistake each other for themselves" is the best way to describe what is going on. You could just as accurately say that they recognize that their sense of "me" is "pointing" at the same thing, that they are both part of the same person.
Overall, one thing that reading this essay did hammer home for me is that people really do strongly identify their "self" with their sense of consciousness, even though they should know better. This is a matter I've given thought before, I know that consciousness is an important part of me, but not the only part. I know that, for example, I am still me when I am in a deep sleep and all my thoughts are unconscious. But in spite of this, reading some parts of this essay gave me the irrational feeling that I was being attacked, that the essay was somehow saying that I wasn't real, and that accepting it's premises would mean that I'd have to accept that people aren't real and therefore have no moral value or something crazy like that, instead of simply believing that everything adds up to normality.
Thank you, this is useful feedback. I did indeed fail to accommodate the emotional consequences of the common mistake of identifying with consciousness, even while I acknowledged they exist. I'll try to do better in future versions of this.
I have a bunch of quibbles with the claims you referenced at the beginning. In particular, 8-13 all feel very dubious but the bigger issue is with how to understand qualia and whether you are dealing with the hard problem of consciousness.
In particular, the distinction is this. If all you want to know is what gives rise in the brain to us asserting things about having this experience of that there is no particular mystery. That's the easy problem of consciousness just something to be worked out by the usual mechanisms of biology and -- while of scientific interest -- doesn't seem particularly deep or philosophically interesting.
The hard problem of consciousness is based on the challenge to explain why you not only **say** and assert that you are experiencing something but that there is really something which it feels like to be you. (And this is usually the notion being invoked by reference to qualia.)
This was best formalized by Chalmers in his book (the conciouss mind?) in which he argues that even if you knew the complete microphysical description of a person and complete laws that describe physical processes you could still coherently suppose that there wasn't actually anything that it was like to be them. More specifically, he argued that phenomenal consciousness doesn't logically supervene on the physical.
In other words, the point being made is that to explain why we are conciouss you need to propose some kind of natural law which says "physical systems like X have qualia Y" as such a law can't be logically derived from laws which only relate physical stuff.
But if that is the hard problem you are trying to solve then by its very nature you can't give a falsifiable theory -- or more specifically you can't give a theory which can be falsified by physical observations -- since the whole point of the problem is to describe the extra law that relates conscious experience to physical reality (though you can look for elegant theories that jointly describe both and check they accord with our purely physical theories but you aren't offering some unified theory that implies the standard model here).
Yes, it's weird because the problem raises questions about our assumptions about the causal efficacy of conciouss experience (is it necessarily epiphenomenal etc) but that's the problem and that means this kind of theory isn't really addressing the problem philosophers are usually getting at when they talk about qualia (and those who want to reduce qualia to physics like Dennet face the challenge of arguing Chalmers is wrong in his zombie argument not any particular puzzle about the exact brain mechanism even if they may speculate).
Let me add that I was just attempting to give a rough idea of the state of the (imo cogent and compelling) philosophical discourse on the issue. I strongly recommend people actually read Chalmer's book The Conscious Mind if they want to know more. It's very accessible and the first half is very compelling imo (his particular suggestion for a solution less so because I don't think he succeeds in defining what it means for matter to carry out a computation).
If you want to get down into the weeds, I'd actually say it's a bit misleading to assume we have a principled definition of the physical or that we can assume that the ability to predict the future from the past means that qualia are therefore causally impotent (as a humean I see no reason to think that events can't be overdetermined by causes when those causes always necessarily co-occur).
I'd go even further and point out that even in a world where qualia did play a direct and non-redundant causal role its not clear we should expect physics to look different than it is in our world (where it seems fundamental randomness exists) so maybe we shouldn't be that quick to assume the causal closure of the physical.
>If you want to get down into the weeds, I'd actually say it's a bit misleading to assume we have a principled definition of the physical
We can cash out the physical as the objective perspective on an intrinsically neutral universe.
>as a humean I see no reason to think that events can't be overdetermined
Overdetermination is not needed. The subjective perspective can be a valid alternative to the objective perspective without having two competing systems of causality in the territory.
What is an objective perspective? It is presumably an objective fact that a given qualia has occurred at some point in time and perhaps location. We generally don't require observability for something to be physical (we allow that there might be parts of the universe so far away we can't observe them as a matter of physical law).
--
It depends a bit on what you mean by overdetermination. It's merely the same kind of overdetermination we have with gravitational and inertial mass. Since they are equal by physical law you can predict anything you can do with one with the other.
You can always write the laws of nature in an empirically equivalent fashion that renders them not overdetermined (so it's like having both the grav and inertial mass that causes all effects) but for this you have to give up the casual closure of the physical (every physical effect has sufficient physical causes).
But that's all mucking with causation and I think it's clearer just to speak in terms of prediction and it just comes out that you can predict the same future events using either a completely physical description or one which uses facts about qualia and some (but not all) physical facts.
But that kind of overdetermination isn't particularly objectionable.
>What is an objective perspective? It is presumably an objective fact that a given qualia has occurred at some point in time and perhaps location
Its the thing that's emphasised in science.
>It's merely the same kind of overdetermination we have with gravitational and inertial mass. Since they are equal by physical law you can predict anything you can do with one with the other.
I don't consider that to be real ,ie problematical,overdetermination, since it's not in the territory.
>But that's all mucking with causation and I think it's clearer just to speak in terms of prediction and it just comes out that you can predict the same future events using either a completely physical description or one which uses facts about qualia and some (but not all) physical facts.
Prediction wouldnt be possible without some kind of causation. If you believe there is some kind of in-the-,territory causation that can be described in various terms , we are nearly on the same terms.
I know it when I see it is hardly a principled distinction. You can obviously just label things in any old way but you are basically accepting it's not principled.
-
Well if that's not problematic burden is on you to explain why this case would be.
-
It's not at all clear there are any extra facts about causation needed beyond what actually happens. Occam's razor says that we should be fine just saying: the universe obeys such and such regularities without any need to say some of those are special causal regularities and others aren't.
I mean the laws of physics let you predict the past from the future too. Are you committing to the idea that one of those directions is causal and not the other? But then why should our arrow of time match the causal one?
If you want to just use causation as a synonym for "required by natural law" fine but then everything I said stands.
>he argues that even if you knew the complete microphysical description of a person and complete laws that describe physical processes you could still coherently suppose that there wasn't actually anything that it was like to be them
I expect that by the time that we actually have something approaching such understanding then hard problem would have long been dissolved, therefore we wouldn't be able to coherently suppose that.
But that basically is the hard problem. The phrase was defined by Chalmers to refer to the problem of explaining why, even if you've identified and completely understand the microphysics, why those configurations are the conciouss ones. If you don't assume qualia are not reducible to (or more accurately logically supervenient on) physical states there is no hard problem.
Hmm, so he posited an entirely hypothetical problem, given that our actual understanding is nowhere near that? Doesn't seem particularly useful. My intuition says that it's precisely our lack of understanding of physics and engineering of the brain that gives rise to the confusion about why we have first-person experience, or qualia. This is what I think the "hard problem" commonly refers to, which makes more sense, since it's a real problem instead of a hypothetical one.
Then why are you here reading a blog post that's clearly claiming to solve that kind of hypothetical problem? I mean on your view you should have shut the browser the moment you realized it wasn't a deep dive into neuroscience since -- in terms of purely I/o there is no more to explain about consciousness than there is to explain the operation of the vision system in the brain -- and certainly nothing useful to say that isn't a deep dive into neuroscience and brain regions. So why are you even here?
But frankly the primary real world application of philosophy like this is to prevent people from wasting even more time by recreating it poorly like this post. People seem to have a near infinite desire to recreate philosophy (which is somehow only pointless when it's not what they are thinking up) and it's far better to have a few people lay it out the arguments/concerns in an actually compelling and clear as possible so it's not reproduced a hundred times with confused arguments.
Mainly I read stuff for entertainment, this post is no exception. It does seem pretty confused about philosophy, but that doesn't matter. Everybody is confused about philosophy, but given that it proposes to actually do stuff, that might end up leading to new insights and eventually resolve this confusion. That's the only way they ever get resolved.
I didn't fully follow this, but I feel like it mostly just shifted the problem of explaining 'qualia' to explaining 'our experience of neural oscillations from the inside' ?
Roughly, yes. It claims that's what qualia actually are.
I'm sure you're familiar with Fake Explanations: https://www.lesswrong.com/posts/fysgqk4CjAwhBgNYT/fake-explanations
Sure, this is what qualia is (probably), but notice that it doesn't explain why or how that leads to subjective experience.
Fake explanations don't reconcile complex unexplained observations with point by point derivation from simpler postulates, and they especially don't make novel testable predictions. This does both.
You saying that Daniel can predict novel qualia? Is Daniel saying that?
Yes. Specific predictions were listed at the end, including a prospect for technologically-mediated telepathy.
In not seeing it. This
>Even unusual conscious experiences caused by intense meditative practice now seem to me fully describable without resorting to the religious terminologies of the traditions that produced them20.
...is a claim of describability, not a description.
> a prospect for technologically-mediated telepathy.
Is a prediction but not of novel qualia.
Yes I am saying that. The most direct way I'm saying it is when I claim that it is possible to design non-traditional forms of meditation where arbitrary single qualia fill all awareness.
I also happen to write non-traditional meditations. Total coincidence of course.
I'm sure you're familiar with Fake Explanations: https://www.lesswrong.com/posts/fysgqk4CjAwhBgNYT/fake-explanations
Sure, this is what qualia is (probably), but notice that it doesn't explain why or how that leads to subjective experience.
Taking "from the inside" seriously means dual aspect theory , which is not (quite) physucalism.
I don't see how it is not (quite) physicalism. Care to explain?
That anything should seem differently from the inside is not an implication of physics -- it's additional fact.
Love this
Excellent topic with a lot of good seemingly objective analysis regarding the physiology of action potentials and their cumulative impact on thoughts and consciousness. I was following along in mutual agreement up until it veered into the “Payoff” and was having trouble appreciating how it was all being tied together. It became a bit much to process tonite, especially after imbibing with friends, so I’ve stuck a pin in it and am looking forward to returning to digest it tomorrow, with a clearer head.
I did have one quick thought tho! In the qualifications of quaila and thought perception, you suggest:
“Irrevocability: qualia can’t be directly overridden by top-down attention any more than other factual information being processed can.”
Which I take to imply you cannot just ignore stimuli to the point of not experiencing it. And that that experience exists independently of your own recognition. If I got that all wrong, then stop reading now.
Does that contradict the idea of desensitization treatment and behavioral adaptation to PTSD triggers, sometimes with guided meditation techniques or dissociative chemicals? Is that global awareness of synced oscillations an instance of an outside thought recognizing the deleterious impact of the internal thoughts manifesting a belief perceived externally as a disproportional or irrational thought response? And if so, is the conscious act of attempting to control your thoughts and succeeding in moderation of irrational oscillation cycles, a direct top-down override?
Desensitization treatment isn't a *direct* override - it's indirect, partly mediated though control over outside sources of the relevant stimuli. Also tends to be far from absolute.
That sounds like a very authoritative statement about untestable pseudoscience, instead of any real scholarship that supports your argument.
great article, but misses a very subtle but key assumption: neuroscience can observe "signals" on the brain, subjectively you can also observe a phenomena at the same time "eating chocolate/stress", but there is no correlation that signals cause the subjective experience. the whole premise of your experiment assumes causation of signals causing the subjective experience, but you can never prove that!
eg: you mix chemicals A & B and see a blue colour in the beaker, can you infer that the colour blue is the reason chemicals A & B are mixing? byproducts of a process are not the ones causing the reaction. in chemical reactions you can see the chain of causation, with neuroscience you can never tell if the signals are byproducts or ingredients!
"The color blue" isn't the reason why the chemicals are mixing, nor is "the chemicals mixing" the reason for the creation of the color blue. "The creation of the color blue in the beaker" and "the chemicals mixing" are one and the same thing— there are many things that can cause us to observe blue wavelengths (Rayleigh scattering, fluorescence, reflection, absorption, heat production) but none of them are "chemicals mixing". Either a new compound is created (and that compound *is inherently* blue; you can just as easily say a chemical exists because it interacts with electromagnetic radiation in a certain way as you can say the electromagnetic radiation acts in a certain way because the chemical exists) or a new physical network is created (same thing).
That is to say, "the color blue" and "the chemicals mixing" arise from a third thing, "the force of electromagnetism". What is "causing" the reaction is the collision of various molecules with a low enough potential energy barrier to allow them to overcome repulsive forces, and enough ambient energy to allow for that collision.
To put this back into the terms of the hard/easy problem of consciousness, one might say that what the hard problem of consciousness seems as if it is asking is "why is this compound formed when molecules collide", but it is in fact asking "why do molecules collide at all". Subjective experiences arise (and the geometry of the brain is what it is) for the same reason that some chemical compounds reflect/fluoresce/absorb enough to allow blue light to be sensed, and for the same reason that chemicals combine to make other chemicals. It is because physical forces and laws are what they are. It would be nice to know why those laws *are* what they are, but that's not a question about consciousness or chemistry!
Incidentally, the molecular orbitals which cause certain wavelengths of light to emit from molecules and which cause molecules to combine in particular ways are also oscillating wavefunctions.
All theories are just word salads, fluorescence, electromagnetism you can keep creating new words but you can never explain chain of causation fundamentally.
Similarly you can just handwave that consciousness is emergent property just so it's easy for you to not think about it, but no theory will be able to postulate how the subjective experience arises
All theories are just word salads, fluorescence, electromagnetism you can keep creating new words but you can never explain chain of causation fundamentally.
Similarly you can just handwave that consciousness is emergent property just so it's easy for you to not think about it, but no theory will be able to postulate how the subjective experience arises
Yes, that's fundamentally what I'm saying. The standards to which you hold explaining consciousness makes "what *really is* a chemical reaction" just as impossible to explain. Of course, you can definitely bite that bullet and say we will never solve the hard problem of how sodium and chlorine make table salt, but it's good to understand that you're rejecting physicalism in many more arenas than just the brain when you reject physicalism in the brain by this particular standard.
Awesome, thanks for the replies learning a lot!
Problem with physicalism is it can explain everything in the world with other abstractions, but when it fails to account why I feel conscious, then it is a useful abstraction but not the truth.
The opposite way is start with something you can take as truth, your own subjective experience, you don't need any assumptions or explanations to know you are conscious, "you just know you are" - it's self evident!
So now with conscious as fundamental you can try to explain how the physical world will exists wrt you the observer. If you want to read more about this look up Advaita Vedanta
You make a big deal of "oscillations" but it's not clear the word has any more explanatory power in your story than any other word for prolonged brain activity would have. You connect these oscillations with thoughts thinking about themselves, but these two types of loops have nothing to do with each other so far as I can tell.
The oscillations in the brain may well serve no more role than the clock cycle of a microprocessor: the periodicity helps prevent different runs of the same circuit from conflicting, but nothing more. Even for the easy problem of consciousness, I expect the true answer to be vastly more complex than just "oscillations".
You're right, the true answer is vastly more complex than just "oscillations". This essay is also vastly more complex. The question is whether this essay is (some approximation of) the true answer, and simplifying it so resolutely isn't going to answer that.
Aren't you strawmanning Daniel with that last sentence?
So he claims. His essay came across as empty sophistry to me, though; I don't really believe there's a well-defined non-tautological theory of consciousness here that's different from just the word "oscillations" underlined 3 times.
Yes there is.
The core of this theory seems to be that the brain can represent its own processing to itself. It proposes some mechanism for it, although it was pretty vague - is it just that oscillations can have different frequencies? Seems more like a potential per-requisite than an explanation.
In any case, the theory just equates the ability to represent its own processing with consciousness and calls it a day. Why would the brain do this? How does it explain phenomenal consciousness? What does it predict about altered states of consciousness, or pathologies of consciousness like blindsight?
The proposed experimental tests are not tests of the theory, they are just tests of the link between attention and brain rhythm synchronization, and they're being done all the time.
I think the Attention Schema theory beats this one on all accounts. It says that the brain represents its own attention (not all processing, not all aspects of processing), for the purpose of controlling attention, and exapted as a theory of mind module.
No. Lots of things can "represent themselves", including a piece of paper with the words "this piece of paper". This theory makes a much more specific claim.
I didn't get into blind sight or altered states of consciousness because the essay was already too long, which is evidently leading to people commenting who haven't actually read it.
Do you have specific questions about blindsight or altered states of consciousness?
What is the more specific claim? I read the whole post, but the essential difference escapes me. I can find these quotes:
"the processing of information that is part of the oscillation/thought is handled distinctly from the processing of information that is not (...)
the distinction between internal and external information is neuronal activity, i.e. the exact kind of information that an oscillation/thought can process, it can notice this distinction and thereby notice itself"
"Qualia are nothing but information being processed internally, on their own information channel, encoded in the rhythm of the oscillation."
"When neuronal activities reflect each other, they can achieve self-reflection."
"oscillations/thoughts with memory are themselves additional information processing systems"
I'm not sure whether it's one oscillation representing another, or a single oscillation representing itself. It sounds maybe more like the latter, judging by the last quote - you're saying that a single population of neurons is doing two computations at the same time? A physical computer running a virtual machine is still switching between computations very quickly, but it sounds like you're saying the same activity should be interpreted in two ways simultaneously, and the latter is reflecting on the former?
Whatever it is, it still sounds like self-representation with extra steps. I don't have an issue with that, since I also think that's a core mechanism behind consciousness. I just think one needs to explain 1) why this self-representation occurs 2) why it causes people to claim phenomenal consciousness.
It's a single oscillation, but not all neurons in it are representing the same thing. A (very) simplified model would be two subgroups inside the same oscillation each representing the other. (In actual fact, oscillations aren't simply going back and forth, the circularity of their activity is more varied.)
This occurs because it allows information to persist over time, i.e. it implements working memory, which has evolutionary fitness value.
By oscillation, do you mean something else than https://en.wikipedia.org/wiki/Neural_oscillation ?
Why are we less conscious during sleep if oscillations are equally or more present? How do oscillations implement working memory at the same time as they implement self-representation?
Oscillations are not equally present in sleep. It depends on the phase (REM is rhe most widely known) and that happens to be a fomain of electroencephalography research. There are literally thousands of papers on oscillations in sleep and how they relate to the strange quasi-conscious phenomena of dreams.
What you linked is exactly what I mean.
I'm not getting the picture. If any kind of neural oscillation somehow causes consciousness, then humans during sleep and all animals should be roughly equally conscious as awake humans, people having epileptic seizures should be more conscious and neurons grown in a petri dish should be conscious. If your theory says it's oscillations plus some extra conditions then I don't know what those conditions are. Most importantly I don't see what explanatory work oscillations are doing in the theory.
Would every possible information processing event (occurring in a computer, for example) that had the same properties 1-15 (Distinguishability through Unity) also have qualia, or is it only these oscillating thought patterns that have qualia? If the latter, what explains why? Would a simulation of a thought (a virtual machine) running on a computer also have qualia under your theory? How would we know?
(I mean a simulation of the individual neurons, exhibiting the appropriate wave patterns to instantiate conscious thoughts when occurring in a brain)
I don't know, because I don't know how the simulation would work in detail. "Simulation" is a super broad term, much like "calculation", it is hard to make statements about all possible members of that category.
If you knew the answer to how the simulation would work in detail, could you answer whether the machine would be conscious or not under your theory?
Yes, but I make no claim whether (and therefore don't rule out that) any other phenomena caused by anything else than these bio-neuronal dynamics deserve to also get the "consciousness" label.
Thanks for writing this. I appreciated consciousness being described as a property of thoughts rather than a unitary self. Like others, I also see this as an argument for physicalism, but that should be the default view. Qualia seem to be the last vestiges of what 16th century monks called a soul - the idea a human mind has something special going on that other information processing hardware does not. Red can’t simply be light 620-750 nm.
We can’t see into other minds. We can’t experience a subjective consciousness outside the one instantaneous moment we are in. Our train of thought is on a railway of confabulated and fragmented RAM. To me the framing of a “hard problem” seems not only human-centric, but solipsistic and leading inquiry far astray from any universal truth of nature.
X satisfies (1) - (16), Y satisfies (1) - (16), therefore X is Y is not a correct argument. Even if Y predicts characteristics of X, e.g. "my hamster is brown and furry, so is my dog, therefore my dog is my hamster" is not any more true regardless of whether this theory predicts e.g. that your hamster needs water to survive because your dog does too.
Some other gaps in the reasoning:
> "A neuron that is part of one oscillation can hardly also be a part of another, so oscillations compete for neurons."
This seems very non-obvious, mathematically it's very easy for signals to be part of multiple oscillations.
>Mental information processing doesn’t seem possible without these electric spikes.
The claim here is that the spikes are necessary, but this is a distinct claim from them being sufficient.
Thanks, you said whether I wanted to but more clearly.
In general, all respect to the author, I think the theory has too many free parameters and can easily be made to fit any observations. For example, re Unity, why should separate thoughts 'compress' their self representation into an I rather than (for example) just reduce granularity in a general sense - such that reflected thoughts are fuzzier than more directly experienced ones? (Perhaps this is actually true, but isn't very compatible with the author's model of jhanas).
I think oscillations attuned to other oscillations might be a relevant part of the picture, but in trying to present them as a complete explanation, the post tries to prove too much.
Equating the hamster and dog fails because they have other properties. The argument is correct if it lists ALL properties. These are all 16 properties of qualia I could find that aren't inconsistent, metaphorical, or poetry. To show how the argument is incorrect here, please show properties of qualia that aren't also properties of information being handled by processing inside the same oscillation that also recursively reflects them.
Yes signals can be part of multiple oscillations, if their frequencies are natural number multiples of each other. That's why I wrote "hardly", not "not".
Are you claiming that the brain can process information without spikes?
The same water molecule can be part of several different waves. It doesn't matter in the least whether their frequencies are integer multiples of each other. The waves will pass on undisturbed, and the water molecule will end up where it was. Waves are weird.
For a nonlinear system this won't be the case. (Neural refractory period is a fairly trivial nonlinearity here)
Fair point. Thank you.
I don't think it is a fair point, nonlinear systems can do weird things like allow solitons; it's a bit hard to speak of "waves" for nonlinear systems but there are definitely wave-like behaviours. FWIW the shallow water equations *are* nonlinear.
Also fair. And yes, the water equations are of course nonlinear. I just decided I didn't want to argue about different types of nonlinearities and which behaviors might show up in these particular waves, and why, none of which I'm expert on. But yes, for especially small oscillations in even a nonlinear system, one might expect to see superposition in some circumstances.
>The argument is correct if it lists ALL properties
>These are all 16 properties of qualia I could find that aren't inconsistent, metaphorical, or poetry.
Right, so necessarily ALL properties, even by your measure. Whether or not A and B are equal if all properties of A that can be discussed are equal to properties of B that can be discussed is very much the crux of the issue.
>To show how the argument is incorrect here, please show
No, the argument "A is brown and furry, so is B, therefore A = B" is wrong even without me proving that A != B (Yes, I know about Gödel's first theorem, no, it does not apply since this is philosophy and not (first order) logic).
>That's why I wrote "hardly", not "not".
This is very much non-standard use of the word "hardly"
>Yes signals can be part of multiple oscillations, if their frequencies are natural number multiples of each other.
This may be true for standing waves for e.g. the linear wave equation, but there are plenty of (nonlinear) systems with weirder behaviour that admit things like solitons or whatever you want to call the solution of the Kuramoto–Sivashinsky equation.
>Are you claiming that the brain can process information without spikes?
I am making only claims about the argument, not about the brain itself.
Yes, your criticism of the non-standardness of "hardly" is valid. I shouldn't have written this that way. Thank you. I promise I will do better on the next version of this theory.
Now whst do we do about the crux? I suggest we call this problem "if it looks like a crux and it quacks like a crux and it walks like a crux".
> I can resolve to not anthropomorphize myself for once, and let my thoughts see themselves as thoughts.
"I" can resolve? Who is this "I" you speak of, I thought everything was just thoughts? :P
If it's just thoughts all the way down, why not just let it be atoms and molecules all the way down? Also, just because you can convince yourself that it's thoughts all the way down does not in any way prove anything about the nature of consciousness.
"I" second this. Throughout the post he uses contradictory language, like:
>An additional error is made when people identify themselves with “their” consciousness.
Who is the person you are talking about? On your theory there are no people, only thoughts.
>Not only do they wrongly assume that there’s a thing there to identify with; this assumed identity also creates false intuitions:
Who is the "they" making a wrong assumption? There are no theys, only the wrong thoughts themselves; yet these wrong thoughts identify with themselves, and create an assumed identity that doesn't exist? How can someone that doesn't exist assume an identity?
>The supposed thing-ness of consciousness seems more real, since people usually believe themselves to be real.
Again, what "people" are there that can believe things about "themselves"?
>Consciousness seems like there’s only one of it, since people usually believe there’s only one of themselves.
Again, who are these "people" who believe that there is only one of "themselves"? Under your own theory, there are no such people.
Statements like these want to have their physicalism cake and eat it to: "I" is only an illusion, an illusion that I (who don't exist) experience. Nonsense remains nonsense even when we talk it about consciousness.
"I" and "people" are shorthands, useful because "the organism that is having this thought" is an exhausting thing to type five times per paragraph. Yes this causes confusion, as all simplification does, but that says nothing about the validity of the actual argument.
What is the meaningful difference between "I" and "organism" in your view, such that the statement "I am having this thought" and "The organism is having this thought" are different?
That is a very good question, and I wonder what is the answer.
Third.
The difference is that most of the organism is not saying "organism". Like, your pancreas is not choosing the words. It's a thought that does that. A thought that identifies with the organism that thinks it, incorrectly.
How is that different than saying "I am having this thought?" Surely those who say "I" do not believe that their pancreas is doing the thinking either?
In itself, it is no difference, if "I" is just a shorthand, literally just a vowel. The confusions arise if "I" is treated as more than that, as some special thing that can be isolated and studied.
There is no actual unitary "I" (although suspiciously, all of this is happening in a single actual unitary universe, but I'll try not to go mystical for once). Everything I wrote about how "I" is an abbreviation for whatever the sense of self happens to point at, I absolutely meant.
Later I use normal English again. Like Scott told us, you can think "humans" but please say "people" like normal... people. You can make a gotcha out of that if it amuses you, but if you're looking to engage with my argument it is nothing but a distraction.
Great post! I'll have to think more about it but I already feel that I'm a bit less confused about some parts of the problem and that the non-physicalist goalpost moved even further.
Your answer to the question: Which types of algorithms, when executed produce the referent for what people call "consciousness" is dully noted. Looking forward to the experimental tests. I think there has to be something more to it - to experience the quale of redness some part of the system has to generate a picture where this redness is encoded in a particular way which another part of the system is able to decode. But what you are talking about indeed seem to be part of the answer or at least a fruitful direction to pursue.
the nondeterministic turing machine link is confusing. NDTMs are not believed to be physically possible. perhaps you mean not deterministic, like probabilistic
Yeah, because of that I don't recall discussing them in any CS course. There was the concept of an "oracle" which is assumed to behave in some way to give a result, but that was used to assume something and then prove a contradiction to show things like a general solution of the halting problem to be impossible. The wikipedia link sounded a bit like a quantum computer, where multiple solutions are attempted but only the correct one is used.
I'm surprised you haven't. Was NP explained to you without mentioning that the N stood for non deterministic?
That was nearly two decades ago, so it could be my memory is fuzzy.
There are two equivalent accounts of Nondeterministic Turing machines, one is that at every multiple choice branch where each branch looks appropriately equivalent, you spontaneously duplicate the machine, such that both branches are taken. Then, the first machine that halts has its time taken / space used to determine complexity class you are. We ignore all other possible Turing machines that went down any wrong branches.
The other, equivalent construction, is that at every branch, the Turing machine "magically" picks the correct branch every time.
So even if the non deterministic Turing machine is not physically realizable, it is also a way of saying "okay, right now we don't know how to tiebreak on these branches, but if we had a perfect tiebreak oracle, here's what the performance would look like".
It's possible I just didn't fully understand the article, but 15 still confuses me. If you can have multiple conscious thoughts in your brain at once that aren't directly aware of each other, would this be noticeable in our behavior? For instance, if you have two thoughts going on at once, "I'm hungry" and "I'm sad," and one of them causes you to say out loud that you are hungry, does the other thought hear that and get confused because it perceives that it's just sad, not hungry? Or does focusing enough to make a decision cause oscillating thoughts to merge, so there is no confusion?
When I remember experiencing multiple qualia at once, are some of those times actually the memories of multiple thoughts that got stitched together after the fact? Does focusing on all the different qualities you are experiencing cause thoughts to merge?
I think maybe when I'm trying to understand 15 I'm subconsciously mapping it onto the concept of multiple personalities, when most thoughts probably aren't that complex. Could that be the source of my confusion?
They merge easily. I have often had the experience of hearing myself say something unexpected (doesn't that happen to everyone?) and when that happens I usually trust it to be true. I think this trust is what the merging feels like from the inside.
I guess what I'm confused about is how two different simultaneous conscious thoughts get "recorded" in your memories. When you remember the moment where you said something unexpected, how come you remember the thought that was surprised by you saying it, but you don't remember the other thought that came up with what you said and therefore wasn't surprised by it at all?
If you were to have two thoughts at the same time that were both conscious and didn't merge at the time, how would they be remembered? For instance, if I had one thought that I was hungry and one that I was sad, would thet both get recorded so that later I remember both feelings simultaneously, even though they were unaware of each other at the time? Does that question make any sense at all or does it misunderstand how consciousness relates to memory?
When I pay attention to (meditate on) two conscious thoughts merging, It feels like two parallel universes merging and like I was previously in both but now I’m in just one.
This makes me very interested when Bayesians talk about the terms of the Bayesian Theorem in words like “a world where…”
I don't remember this ever happening to me.
I'll look out for it in the future.
You may want to Google "automatic writing" - every usable description starts with a complaint about how that is a terrible misnomer.
An interesting aspect of newborn baby behavior I've observed is the baby simultaneously trying to satisfy two different biological urges. Namely to drink milk and to relieve itself. It's drinking, it stops to strain and grunt for a while while also drinking a bit maybe, drinks some more, etc. It's like both urges are independently active, with the baby kind of drifting in response from one to another. Absent is the adult way of controlling/prioritizing to comfortably achieve both; like for example after running a race you'll catch your breath first before getting a drink.
Thanks for this! As others have noted, it doesn't really get in to the philosophical weeds about how (even in principle) we could go from a physicalist universe of quantitative relationships between 'empty' atoms, to the subjectivity of qualia; but then, a finessed one-to-one correlation between those subjectivities and specific neuron activation states would be quite the prize in itself, so that's no great failure.
A couple claims that I understand to be common (if not true), in relation to this subject, which I was surprised weren't mentioned and might be consequential for your discussion:
From the mindfulness meditation angle, I've seen it claimed in multiple places that in fact only *one* quale can be present in conscious attention at the same time; that our ability to hold multiple observations/qualia/tokens in place at the same time is actually a rapid juggling maneuver. It is claimed that we cycle between our one quale so rapidly that it looks like multiple qualia at once, much like rapidly flicking images in a film can become the illusion of a moving picture. And I'm curious whether you see your theory here as being robust against that claim! (I first ran across this claim in "The Mind Illuminated", to cite my sources.)
From the neuroscience angle, I understand that the size of neuron activation cascades isn't just a 'small' versus 'large' distinction, but rather, that they follow a power law- not unlike actual physical avalanches in gravel or snow that is stacked at a critical angle. This doesn't directly contradict your premise, but it doesn't seem like an intuitive consequence or prediction from it, I don't think. Actually, I would think that additive combinations of small sense-data pulses would follow a bell curve instead (as you'd get if you add up the results of a 6d6 die roll, for example). Is there any particular reason why you would expect a power law result, given the processes you have described here? For a robust treatment of this claim, you can check out "Criticality in Neural Systems", edited by Plenz and Niebur- though it's a decade old at this point, so it might have missed some of the most interesting modern research.
They do make bell curves in time, the power law is on the spatial dimensions.
Ugh. Another article in the Hofstadter tradition of short-circuiting philosophical reflection by massively begging the relevant questions.
In "European Literature and the Latin Middle Ages," Ernst Curtius tries to give a long history of European literature in terms of topoi, common rhetorical constructions which repeat and persist across ages. The "past golden age" or "descent into the underworld" are structures so repeated in literature that they form a background part of consciousness, seem entirely normal. TVTropes catalogues these things for modern media. What Curtius does is take in mind such a long timeframe that he can see these tropes in the longue durée, comparing individual uses with others in a long history to take account of the warp, not just the weft, of literary structures. He encourages a reading of literary topoi which sees small inventions and novelties disappear in the longer, bigger continuities of literary history.
The history of philosophy has its own topoi. Getting bogged down in the details of individual philosophical arguments--which is by far the most important portion of philosophical training--should not blind us to the ways that the general shape of arguments have a long history. Recursive reflection, for example, is a philosophical topos that has dominated a strand of German philosophy since Fichte--who the author should read, not as inspiration, but as a sort of warning. Fichte--like Schelling and Hegel after him--that self-reflection is somehow the key to consciousness. His texts founder, though he is not always aware of this, on a series of difficulties which beset 150 years of German philosophy which aggressively takes on the problem: that self-reflection is almost impossible to logically and consistently define, that there is no logical guarantee that the thought thinking and the thought thought of are the same entity, that the identity of the thought thinking and the thought thought is almost metaphorical, that the intentional consciousness (seeing something as) which is the phenomenological counterpart of metaphor is aggressively inconsistent with the basic metaphysical tenants of physicalism, etc.
This post aggressively leans into the error of thinking that something like "self-reflection" describes something which can be defined, rather than something which is part of the basic shape of Western philosophy which is both indispensable and completely resistant to all our efforts. At the same times, it leans on other ideas, e.g. "information" which it treats as basic even when such things are models and metaphors.
Well, I'm an IT guy, "information" is neither just a model nor just a metaphor to me.
Which relevant questions do you think I'm begging?
Now I am curious. What is information to you? "Information" seems an almost impossible to define word. We can define it in very specific contexts, but much like "game", once the context differs, the definition differs. Or do you think the Shannon definition of "information" suffices to cover all usage?
Information is an extremely contested concept in philosophy, and it's extremely difficult to define it in a way fundamental enough for it to have the conceptual leverage you want out of your system that doesn't automatically entail a commitment to physicalism. Janich's "What is information?" is an accessible start.
As for begging the question: You assemble a list of properties of qualia which come from a variety of sources which originate from different perspectives: (taking your info from Dennet's "Quining Qualia" is more than a little bit tendentious)!
You struggle to maintain a principled differentiation between the concept of thoughts and qualia, and as the concepts elide, they cross casually the very demarcation line the concept of qualia is designed to draw our attention to.
However you want to label the problem, whether you call it Leibniz' gap or the Hard Problem or the problem of qualia, etc., you need to answer the problem on the level that it's posed. The problem is not one of creating plausible-sounding accounts, or generalized accounts that sound more plausible than the non-physicalists--every non-physicalist already knows that the non-physicalist account seems aggressively implausible. The problem is posed by a few edge-case logic problems that seem stubbornly resilient no matter how sophisticated the science is. And addressing these is not a matter of speculation about the brain, but very careful dissecting these on the level they are posed.
I don't regard commitment to physicalism as a problem, actually. I'm happy with being nothing but nature.
"the world of objective facts, physics and brain imaging technology, where nobody has found these supposed qualia as a thing to point at." - why does nobody mention, at this point in their consciousness article, that they are then and there writing about qualia in the world of objective facts, and so at least on a meta level qualia do have a clear and measurable effect on the physical world? I cannot point at qualia, but I can point at a lot of physical effects directly downstream from qualia, which is surely almost as good. (Is this the great blind spot that makes people write consciousness articles in the first place?)
I'm of the opinion that there are only easy problems, and once all the easy problems are solved there won't be a hard problem left. With that in mind this prospective explanation - information flows around through neurons somehow resulting in the experience of qualia - is as good as any at this level. I don't mind that this article doesn't address the hard problem, since it doesn't exist (IMHO), nevertheless vaguely handwaving in the direction of recursive processing and "information" is not yet a physicalist explanation of how consciousness comes about.
"Qualia are *nothing but* information being processed internally, on their own information channel, encoded in the rhythm of the oscillation." - I think the actual explanation should go much further than just stating this simple equivalence. "Qualia [are] bits of information" - but "bits of information" is only an abstract idea that we use in apprehending the physical world. Information theory can be part of a framework that explains consciousness, but consciousness is not explained by simply equating it to some aspect of information theory.
Daniel - I love this theory. I love how you write and how it included a testable hypothesis. Thanks!
It is now my new head-canon. It speaks volumes to your writing and your theory that “it feels so obvious that it can’t be significant”, and yet, it fully explains consciousness and how our brains work in a way I’ve never groked before.
But, reading the comments, it seems people are caught up on old, less relevant debates and discourse about mentality and philosophy. I think the next step, and perhaps harder step, to presenting your theory, is something that Scott often does, and that’s break down / steel man the classic debate, and show why it does/does not matter anymore. The fact is, no matter how much philosophers debate about how many fundamental elements there are, sometimes you just need to get out an electron microscope.
Thank you. For a good breakdown of the classic debate, I recommend Rafael Harth's "Why it's so hard to talk about consciousness".
https://www.lesswrong.com/posts/NyiFLzSrkfkDW4S7o/why-it-s-so-hard-to-talk-about-consciousness
Thanks. It appears that I am so firmly in Camp #1 that I don’t even understand the fuss about qualia from camp #2. Not having heard about it before, except in these comments, makes it feel like peak-over-philosophizing. I could also say that it’s impossible to model the best bridge design because first you need to define the inherent philosophical nature of bridge-ness, but I’d rather model them based on measurable factors like ability-to-be-crossed and cost.
Thanks for writing this comment, so I don't have to.
And also making me laugh. Spot on.
I've waded into this stuff a fair amount... Had some conversations long ago with possibly one of the smartest people I've met who is firmly in camp 2...
But it seems like these really are two kinds of people who will forever be at an impasse.
I agree; it absolutely sounds like, "peak-over-philosophizing". Arguing about how many angels can dance on the head of quale.
In support of there being two kinds of people, and you either "get it" or you don't, is the experience of C. S. Lewis, as written in his autobiography:
"We had been, in the technical sense of the term, “realists”; that is, we accepted as rock-bottom reality the universe revealed by the senses. But at the same time we continued to make for certain phenomena of consciousness all the claims that really went with a theistic or idealistic view. We maintained that abstract thought (if obedient to logical rules) gave indisputable truth, that our moral judgment was “valid”, and our aesthetic experience not merely pleasing but “valuable”. The view was, I think, common at the time...Barfield convinced me that it was inconsistent. If thought were a purely subjective event, these claims for it would have to be abandoned. If one kept (as rock-bottom reality) the universe of the senses, aided by instruments and co-ordinated so as to form “science”, then one would have to go much further—as many have since gone—and adopt a Behaviouristic theory of logic, ethics, and aesthetics. But such a theory was, and is, unbelievable to me. I am using the word “unbelievable”, which many use to mean “improbable” or even “undesirable”, in a quite literal sense. I mean that the act of believing what the behaviourist believes is one that my mind simply will not perform. I cannot force my thought into that shape any more than I can scratch my ear with my big toe or pour wine out of a bottle into the cavity at the base of that same bottle. It is as final as a physical impossibility. I was therefore compelled to give up realism. "
I recommend the comments.
Seconded.
It's really amusing seeing the people who think qualia must me something that other people have but they don't, to whom explanations just sound like word salad. It seems obvious that they are just assuming it must refer to something more special and weird. I get it; I've done plenty of meditation; "qualia" are the things that you can be mindful of when you practice mindfulness. It's a fancy way of saying "experience" or "sensation". And I get why it seems kind of baffling how it could result from matter. But I don't share the intuition that it's Super Super Special and Hard... There are a lot of things that have been revealed by science that I know I'll never be able to wrap my head around, like quantum physics, but that doesn't lead me to conclude that science is flawed, and that I can then come up with an alternative metaphysics that is a truer understanding of reality by sitting around navel gazing. Ok, sorry, I'm being too hard on the other side now. But...we're just different kinds of people I guess, and there is no way to bridge the gap. That's what I have today decided to dub The Harder Problem of Consciousness.
I do think if we can reduce qualia to the physical brain, recursive neuronal processes are the best bet. But to really make it work, I think our Physics has to get friendlier with Panpsychism. Maybe our Physics already is on that path: It from Bit, the integration of Physics and Information Theory - these point to a conception of matter that is not so knock-on-wood hard as the old Stuff.
By your terminology, would 'sapience' be a mainly human thing, but 'sentience' more widespread on the Phylogenetic tree?
This is a random aside, but...sapience and sentience seems like equality and equity; basically synonyms, but the concept they refer to has at least two similar but distinct meanings, so some people try to declare that one is to be used for one meaning, the other for the other, but no one can ever really agree on which is which.
I guess as long as we define our terms in a given situation, we're okay. He does explain his use of 'sapience' in a footnote, and it's distinct, I think, from sentience.
Sapient means knowing, sentient means sensing. So it's similar to the cognitive/perceptual distinction.
I don't know what "Physics getting friendlier with Panpsychism" would look like. Like, how would a world where panpsychism is true be different from a world where it isn't? Physics is interested in panpsychism if and only if there is a tangible difference.
Hi Daniel. Perhaps a world where Panpsychism is false wouldn't have consciousness. The idea being that brains can generate consciousness only if matter already has something mind-like, some (albeit primitive) experiential "inside" even at the elementary scale of Physics.
I don't think that is true. I don't know how to explain the functioning of the brain without (something that has all the properties of) phenomenal consciousness. So I don't think we get a choice between a world that has consciousness and one that doesn't.
I find it great that you put your theory into testable hypotheses. I have little objection against the high-level ideas behind them. That said, you may or may not be amazed that we have measurements about neuronal activity that go beyond your wildest dreams. The downside: I think your concept of oscillations does not pan out.
First about the measurements: in animals like rats, researchers can do a pretty amazing sequence of steps:
- genetically alter the rats so that neuron light up whenever it fires a spike (nowadays they can even use different colors for different types of neurons)
- remove the skull and replace it by a glass skull
- glue a tiny camera on the brain and you can see the neuronal activity of all superficial layers of the whole rat brain. While they are normally moving around and interact with all kind of stuff.
In humans, we don't alter them genetically, but we do implant electrode arrays into some humans (usually to prepare certain operations that they need). Those arrays measure neuronal activity, and you can tell apart a few dozen neurons. This doesn't give us a full picture, but in general the activities of these neurons is similar to the activity of rat neurons. The very strong hypothesis is that there are no huge architectural differences, or differences in activity patterns, except that the human brain is a lot bigger.
Now, I think your conception of oscillations is a bit off.
- It is true that if you average over millions of neurons then you see oscillations. Even those are not periodic at all, but pretty irregular. But individual neurons do not show oscillations in their spike patterns at all. Instead, what happens is that all neurons in this region are simultaneously in an UP state or a DOWN state. In an UP state, all neurons have a voltage level which is pretty close to threshold. So just a little bit of input makes them fire. In a DOWN state they are far from threshold, so even a large input usually doesn't make them fire.
- Most individual neurons will fire with a really low frequency of 1Hz or less when you look at a behaving animal. If you look into older literature, you will often find higher firing rates of 10-50Hz, but this is when people anesthetized the animal, found a stimulus that triggers this neuron, and repeat the stimulus intensively.
- Even with these hyperstimulations, you don't get a "loop", but the firing is highly irregular. Some papers even claim that it is approximately exponentially distributed, which is in some sense the distribution which is furthest possible away from a regular firing pattern.
- So, there are no loops in the sense that neurons lock into a firing pattern. In an awake and behaving animal, you will typically not find neurons which will participate with more then 1-2 spikes (or bursts, but that also doesn't fit into loops) in any given activity. Neural activity evolves in a pretty spurious way, not in tightly coupled cycles.
Now, that might not be so fatal to your concept after all. I am pretty sure that the idea with loops is the wrong correlate for information. But if you think about it, most of your explanation also work if you replace "loop" or "oscillation" by "whatever neuronal activity represents the information in the brain".
No, I think it would indeed be fatal to my theory - it fails if the neuronal activity isn't looping. Because the looping is required for the recursion.
Yes, individual neurons aren't oscillating (except pacemaker neurons, which I'm surprised you didn't mention despite your evident level of rigor) and as components of information processing they're noisy and unreliable. That's why averaging over millions of them is the correct level of analysis.
>Yes, individual neurons aren't oscillating (except pacemaker neurons, which I'm surprised you didn't mention despite your evident level of rigor) and as components of information processing they're noisy and unreliable. That's why averaging over millions of them is the correct level of analysis.
Correct me if I'm wrong, but that reads to me like you're saying "My theory requires oscillation, and it's true that individual neurons do not oscillate, so that means looking at individual neurons is the wrong way to go about it and we should back up and average things until we can see an oscillation; and the reason we should use that level of fidelity is otherwise my theory doesn't work."
I think it's:
- My hypothesis is that some information structure in the brain is oscillating, because [all the arguments in the post].
- Neurons (mostly) don't oscillate on their own.
- Therefore, if something is oscillating, it must be abstract computational structures made up of multiple neurons (or something like that).
- So we should look for oscillations in abstract computational structures in the brain.
- If we find them, and they exhibit the properties predicted by the post, that provides some evidence the hypothesis is correct.
- If we fail to find them, that provides some evidence the hypothesis is incorrect (not perfect evidence, but the same amount of evidence that looking for something predicted by a theory and failing to find it always provides).
This seems fine to me. I don't think it's circular to say "we should look for the stuff my theory predicts at the level of abstraction that seems most likely to be capable of exhibiting it."
Well put, I endorse this.
Hm, I am less of an expert for brain waves than I am for single neuronal activity. But I would be surprised if we had loops at any level. We have oscillations, yes, but the word may give you a wrong idea. You could also call it fluctuations, which is technically the same thing without invoking the wrong image of repetition. Both mean that activity changes over time, not that the activity repeats itself in a loop.
We do measure the frequencies of these fluctuations, but this is just the thing that scientists do with any fluctuating system. "Finding" frequencies does not mean that there is repetition in the system.
What would finding a frequency in the absence of repetition look like?
What you do is a spectral analysis, that is what scientists routinely do with any temporal data. For example, imagine that you observe a park and record every pedestrian that comes along. So, each pedestrian is one signal with a time stamp. That's exactly this temporal type of data.
If you do a spectral analysis, then you will find that some frequencies are more pronounced than others. Perhaps you will find that there are many pedestrians with a distance of ~1 second from each other, because if a group of people walks together, then they are perhaps ~1 second away each other. Other pronounced frequencies are probably 24h (because if there are many people at 6pm one day, then there are also many people at 6pm the next day), or perhaps 1 minute if there are traffic lights nearby which lets people pass with a period of 1 minute. So a spectral analysis might give you three peaks at 1s, 1min, and 24h. (And that is the case with a meaningful signal. A spectral analysis always gives some peaks. If they don't come from signal, then random noise will give some random peaks.)
Now you could argue that 24h and 1minutes correspond to "loops". It's not quite the type of loops as described above, because you never see the same person in distance of 1 minute. And it's not like person A at time x causes person B to be there at time x+1min. The frequency comes from some regulating system (the traffic lights), not from the people interacting with each other.
And for the 1s frequency it's even harder to argue that this is a "loop". This does come from interaction of people, but it's just that some people tend to cluster together in distances of roughly 1s. It is a repeating pattern, so it is correct to identify this as main frequency. But there is no loop.
For neurons it might be similar. If a few neurons in a small region fire more or less together, then there is a dampening inhibitory (negative) feedback to the whole region to avoid that the activity explodes. The neurons may need 10-20ms before recovering from the negative feedback, and then some neurons may fire again, probably a different subset. Then we get a frequency of 50-100Hz, so we get gamma activity, which is associated to cognitive processing.
Slower waves like alpha waves (~100ms difference, or ~10Hz) are probably more like the traffic lights, they are modulated by certain pacemaker cells in the thalamus.
My point in the previous posts was that if we look at single neurons, they show very sparse activity, and are often only active once during a thought of one second or so. This seems more like the situation with the pedestrians, where there are always new pedestrians walking by, than a situation where the same set of neurons keeps exciting each other in a loop.
I worked in EEG research for a couple of years, so I know what a spectral analysis is. But I still don't know what you mean by "absence of repetition".
That the same neuron doesn't spike several times. Or if it does, that the input does not come from the same source for different spikes. As far as I understand, you consider both necessary for a "loop".
I think the problem that people have is really a problem of personality, not of consciousness. "You're telling me that I don't really exist, that it's all just neurons firing off in response to stimuli" doesn't grab people as an appealing concept. And even those who say they accept that there is no "I" still seem to act as if they have preferences and traits and are uniquely distinguishable from a carrot or the wind in the treetops.
Consciousness is not so much the problem - yes, it's physical phenomena in the brain which is a physical object which comes together in obedience to the laws of nature; it's "you're saying I'm not me" is the problem.
I or no I has nothing to do with qualia or the HP.
Zuboff
Kolak
Advaita Vedanta
Schrödinger
Universalism
Open Individualism
Chalmers, The Conscious Mind, p. 85
There's an anecdote of Wittgenstein wondering about the idea that people used to think the sun revolves around the earth because that's what it looks like, and he asks: what would it look like if it *looked like* the earth revolves around the sun?
We needn't make an imaginative leap here - what it looks like right now is what it looks like for the earth to revolve around the sun because that's the way things really are. But to understand this at an intuitive level (rather than just to accept the explanatory power of the heliocentric model), it helps for people to mentally step outside their immediate experience to picture, e.g., the solar system whirling around so that they can explain to themselves how those dynamics could produce the phenomenal experience they have on the surface of the planet.
People who claim there is a hard problem are really lamenting the fact that a reductionist model does not provide them with a means to step outside their own consciousness in order to picture it from the third person point of view. But the hard problem there is not to be found in the essence of consciousness, it's to be found in a faulty definition of what counts as explanation.
What would it look like if it looked like there was no hard problem of consciousness? Exactly how it looks now.
Yes I can. You can just keep asserting I can't, I just don't see how that is truth-seeking.
Beautifully put (by which I mean -- I had better clarify, in this context -- it stated an important truth clearly.)
Under your own theory "Yes I can" is an incoherent statement. You don't exist, there is no "I" or "You" so of course "You" can't get rid of the hard problem of consciousness. Thoughts are happening in your brain that result in you saying "Yes I can", even though the statement has no meaning beyond "A thought is occurring".
Very interesting. I think the “how would the world look different if X were true instead of Y” is a useful way to approach sticky problems like this, since it often reveals the world wouldn’t look any different and the question itself was posed badly or is meaningless. Another example is “free will”: if you assume determinism is true, and then imagine what the universe would look like if humans had libertarian free instead, will how would things be different (this works the other way around too)? If you can’t say how the world would look different if libertarian free will versus determinism is true, then there is something wrong with the questions/concept you’re working with.
I think the same thing applies to philosophical zombies: if the world were filled with philosophical zombies, instead of humans, how would things be different? It’s not clear it would be. Either the concept is nonsensical, or else we are all philosophical zombies.
This is wonderful. It seems two of the best comments are way down here at the bottom. This comment was literally last. Should have been first.
> What would it look like if it looked like there was no hard problem of consciousness? Exactly how it looks now.
This, this, and so much this.
I guess I'm not clear what role "looking" is doing here. What would it look like if it looked like there were no quantum fields? The answer to that will be different depending on when you were alive. I'm not sure how it helps clarify because we're always trying to explain how/why things look the way they do... but then what we see when we look changes over time based on what we've learned and what technology we have for looking.
How about "what would it look like" being "simplest theory that produces falsifiable predictions based on objective measurements"? I thought this was an obvious interpretation, but I guess it doesn't hurt to spell it out.
In the spirit of the original quip: when we surmise everything is spinning around the Earth, we have to add a whole encyclopedia of weird little "but this but that" to explain the moon, sun, planet movements which don't make much physical sense.
Then one day you Eureka out a hypothesis: "wait, how about we're spinning, the moon is going around us slowly, we and the moon are going around the sun, and those other planets are also going around the sun?"
Suddenly you can explain everything in a few sentences and can spend the rest of the encyclopedia of words going into amazing detail instead of trying to explain the basic situation.
I agree this is a good way for science to proceed. But we don't get to control when suddenly you can explain everything in a few sentences happens.
My perception in this discussion is that there are people who are of the opinion that there is no hard problem and there are people who are of the opinion that there is still a hard problem.
Callum who initiated this subthread is of the opinion that people who think there is still a hard problem are "really" caught in some kind of deluded psychological experience. "People who claim there is a hard problem are really lamenting the fact that a reductionist model does not provide them with a means to step outside their own consciousness in order to picture it from the third person point of view."
My sense is that early on in scientific endeavors that preventing epistemic closure is more important than claiming to have arrived at the right answer when there is still very little evidence. It seems to me we are way earlier in the process of understanding consciousness than the point when we were able to show that we're orbiting the sun and not the other way around.
My general opinion is that the level of conveyed certainty on either side of this perceptual difference around the existence of the hard problem is unearned.
I would put the emphasis on it being a deluded (I'd prefer misunderstood) psychologcial experience in that I think the issues are largely conceptual rather than empirical, hence I don't think we need to await further scientific advances to earn confidence.
To pick up the analogy, it's as if we were at a time when geocentrism made manifest sense to everyone because that's just how things look, then someone comes along and says: "you do realise that geocentrism is only a description of what we see? Maybe something like heliocentrism will bring all the facts together in a better way." And the reply comes: "nonsense! There is no arguing that, from my point of view, the world is geocentric. That is a brute fact. Heliocentrism could not possibly replace this because geocentrism is the very thing that must be explained."
I differ from some others in that I don't claim we will one day bridge the gap between physicalism and qualia - I rather believe that people don't have the immediate access to the nature of their conscious experience that they think they do, and the result is that what they say about it is incoherent.
Where science must lead is not on explaining how qualia come to be but how people come to have the false belief that qualia exist.
> What would it look like if it looked like there was no hard problem of consciousness? Exactly how it looks now.
How would *other people* look like if there was no hard problem of consciousness? Exactly the same.
How would *I* look like *to myself* if there was no hard problem of consciousness? Very differently.
And that's why it's called the *hard* problem of consciousness.
How would you look differently to yourself if there was no hard problem? What does "very differently" mean exactly?
It would be very different in that there wouldn't be an "I" that has the subjective experience of looking at myself. Or at anything, for that matter.
Alternative reply:
What would it look like if it looked like there *was* a hard problem of consciousness? Exactly how it looks now.
Oh, I've got yet another one:
> What would it look like if it looked like there was no hard problem of consciousness?
Then it wouldn't look like half of humanity claiming that they have subjective experience.
Your own argument works better *against* your point than *for* it.
I think that's the Easy Problem.
>People who claim there is a hard problem are really lamenting the fact that a reductionist model does not provide them with a means to step outside their own consciousness in order to picture it from the third person point of view.
It's almost exactly the opposite: the third person perspective offered by reductive explanation doesn't describe or predict consciousness as it is experienced,...from a first person POV. Which would be the case in a universe in which it completely failed as an explanation. So.we can' tell whether we are in that universe, or the one in which it is correct in some mysterious, unconvincing way. They look the same.
I think these are two ends of the same thought, as what's demanded of physicalism is a unification of the first and third person, whichever way you look at it. When I say that people want to picture their own consciousness from the third person point of view, what I mean in full is that they expect that a complete third person account would have to entail an ability to compare or simulate other consciousnesses from the first person point of view (such that you would know what it's like to be a bat), and this ability would necessarily apply to one's own consciousness, which could then be recursively stepped outside. The fact that this expectation is (in my opinion) an incoherent description of what an explanation ought to be is why we do not need to doubt the non-existence of the hard problem.
That explanations should predict their explanada is a completely standard.criterion that applies to.everything else.
Why don't you expect an explanation of qualia to.actually predict or convey qualia? Because it's impossible.for.a linguistic (including mathematical) expression t o convey a quale? Well.that would be qualia being ineffable, as require d.
Because it would .involve placing someone into an appropriate brain state ,.so.they can witness the quale from.their first person perspective? That concedes that qualia are.private. Subjective, etc, as required.s
What you need is an excuse for the lack of an.answer to the hard.problem that isn't based on qualia having all the properties they are supposed to.have.
I'm describing the incoherent expectations that are at root of the supposed hard problem. I don't myself concede that there are any such things as qualia in need of explanation.
I am pointing out that the expectations are in line with what usually required of a successful reductive explanation, and there is no incoherence unless qualia are intrinsically subjective, etc
>. I don't myself concede that there are any such things as qualia in need of explanation.
Meaning that nothing has colours or flavours to you? Or that they is nothing private, ineffable , etc?
Or that there is nothing non physical?
Physicalists sometimes respond to Mary's Room by saying that one can not expect Mary actually to actually instantiate Red herself just by looking at a brain scan. It seems obvious to then that a physical description of brain state won't convey what that state is like, because it doesn't put you into that state. As an argument for physicalism, the strategy is to accept that qualia exist, but argue that they present no unexpected behaviour, or other difficulties for physicalism.
That is correct as stated but somewhat misleading: the problem is why is it necessary, in the case of experience, and only in the case of experience to instantiate it in order to fully understand it. Obviously, it is true a that a descirption of a brain state won't put you into that brain state. But that doesn't show that there is nothing unusual about qualia. The problem is that there in no other case does it seem necessary to instantiate a brain state in order to undertstand something. If another version of Mary were shut up to learn everything about, say, nuclear fusion, the question "would she actually know about nuclear fusion" could only be answered "yes, of course....didn't you just say she knows everything"? The idea that she would have to instantiate a fusion reaction within her own body in order to understand fusion is quite counterintuitive. Similarly, a description of photosynthesis will make you photosynthesise, and would not be needed for a complete understanding of photosynthesis
This feels philosophically quite sloppy. You define qualia as discrete abstractions and especially hilariously define them as needing to "last around 80ms". You have rooted you theory of consciousness in SI. This is completely upside down, as it's pretty likely that time as we perceive it is a scaffolding for our experience created by our own minds (a'la Kant). By simply reducing each moment in consciousness to adding all the qualia together you've not really achieved that much. You need to lay out a coherent metaphysics to have this work together. There's a ton more to nitpick, but I'll leave it be.
I didn't even call it a definition, although it is arguably an operational one, but it is definitely not a theoretical one. I trust that you don't need an explanation ofbthe difference.
Fair enough, but I still think there's some steps you skipped philosophically. I don't see a reason to claim that qualia preceeds conciousness. In fact I think qualia emerges from conciousness and nessecarily requires it. Maybe I am misunderstanding the causal chain you propose. There's a better philosophical term here called "affect" which is more neutral with regards to neurons, but then you lose the shocking title.
It's funny, because I don't think we disagree that hard. I think what we call "consciousness" is simply layers upon layers of differentiation that matter makes (think electron being repulsed by other electrons, or bacteria moving towards food) turned in on itself.
I also have to remark that I come off as overly hostile, I apologize for that.
No apology needed. You were less hostile than the Chalmers people, and made more sense.
> „When they oscillate and notice themselves, they have to continue [compressing/simplifying what they're thinking about], because each thought has to be more complicated than its own internal processing capacity can contain. So their self-representation is also compressed/simplified, usually into a notion called "me". Thoughts don't usually think of themselves "I'm a thought" but "I'm me, I'm a human".“
Why should an arbitrary oscillating thought find it especially simple to think of itself as a human? Why should it even know what a human is?
Side note: I'm unclear whether this piece is distinguishing between consciousness, self-consciousness, and rumination? Do you know?
There is: a friend leaves, I feel sad
There is: a friend leaves, I feel sad, I witness myself feeling sad
There is: a friend leaves, I feel sad, I witness myself feeling sad, and proceed to tell a bunch of stories about what it means for me (ie, I prolong the experience through rumination)
They're all three consciousness, right? Do they all involve looping/recursion/spiking etc or just the second and third one?
Interesting. It made me appreciate how much I don‘t understand in the essay. It’s a difficult topic. In this comment and/or the first one, I may be misunderstanding something or missing something obvious.
A sadness quale should not by default involve recursion. The essay speaks of „the special self-referential and recursive case“, where it cashes out for items 8 to 10 in the list of 16 characteristics of qualia.
But if conscious thoughts do not by default involve recursion, and „consciousness“ does not exist in the abstract but instead only as a property of thoughts (the second of the „Three bullets to bite“), why is the essay titled „Consciousness As Reflective Recursions“?
I dunno how best to comment on this... I like this! I mean, how could anyone not? This is indeed a step forward in thinking about how to test what thinking is in the brain.
That said, I don't care about the physical details of this particular proposal. I mean, who cares if frequencies match or don't match?
This is a step forward in thinking about what conciousness means, which in my mind is a step towards moving beyond people like Nagel and Chalmers, who want to see consciousness as something 'special'. In fact, the more we get into it, the more we will see that consciousness itself is not much more than what is described here: being aware of one's own thoughts. The problem with all the "consciousness" discourse to date is that it has suggested that consciousness is *the* distinctive feature of human beings. But it's not. The distinctive feature of human beings is the fact that we have desires and have the mental (including conscious) capacity to pursue those desires in indirect ways.
Thank you! This pursuit in indirect ways requires the ability to react not immediately but after deliberation. There's a theorist who thinks this "hiatus" is what separates humans from animals. His name is Norbert Bischof.
But unfortunately this does require physical details. In some way, the information needs to be stored during deliberation, because otherwise there's only the choice of act either directly or not at all.
Totally off topic, but I once had an engineer as a roommate. He considered the bullet pointed and/or numbered list to be the pinnacle of communication.
https://orsagouge.pbworks.com/f/tuftepowerpointfailures.pdf
Hypertext is even better. Doesn't work in the body of a Substack post, though.
The details are "well-suited" (as written above), or are they "well-STUDIED"?
As written, there doesn't appear to be anything TO WHICH the details are well-suited.
Auto-complete? It'll get you every time.
It was "well-suited to empirical study" but I fucked up.
> I can’t break this down further to the chemical level. But since individual neurons can keep a rhythm all by themselves, this comparison doesn’t seem too much to ask of a single neuron.
I took a neuroscience class years ago, and recall reading about an interesting molecular mechanism on the dendritic body. The operative mechanism was mediated by a protein with six subunits, though it's been so long I can't remember the name of the protein. When an action potential was triggered as a result of an input from nearby on the dendrite, the first subunit of the protein got phosphorylated. The result was that next time that dendrite received an input, it would make a stronger contribution toward cell depolarization -this protein "enhanced" the pathway, in other words.
Phosphorylation of the protein was highly transitory, but could be prolonged with additional subunit phosphorylation. So, if a second action potential were triggered before the initial phosphorylation lapsed, a second subunit would get phosphorylated and reinforce/lengthen the phosphorylation of the first subunit, and so on until all six subunits were phosphorylated and the mechanism was stabilized.
Sorry I can't remember the name of this protein/mechanism, but it reminded me that there certainly are molecular mechanisms at play within individual neutrons (beside just the refractory period) capable of tracking patterns of action potentials across time.
Great! Can you recommend an article that explains this?
Found it! Looks like it's CaMKII: https://en.wikipedia.org/wiki/Ca2%2B%2Fcalmodulin-dependent_protein_kinase_II
As a side note, the mechanistic description in the Wikipedia article under the "Long Term Potentiation" heading is one of those beautifully complex mechanisms that are one of the things I love about molecular/cell/systems biology:
1. The cell is depolarized by local synaptic activity.
2. This change in charge displaces magnesium ions that were blocking a calcium channel.
3. Clearing the blocked channel allows calcium concentrations to rise in that local part of the cell.
4. CaMKII requires Ca2+ for activation, so the local influx of calcium allows CaMKII activation.
Wonderful, thank you very much! My understanding of molecular biology is very superficial, but I will devote study time to this.
> "It is imprecise but more convenient to say 'I’m hungry' rather than 'the person saying this is experiencing hunger'. In that sense, 'I' and 'me' just help get to the end of the sentence more quickly. Analogously, it is much more convenient to say 'I am conscious of this moment' rather than 'The thought that is directing this mouth to say this sentence is conscious of this moment'."
I find this observation interesting, because in Japanese, for example, it actually is more convenient to say stuff like this without inserting an "I" or a "me" into the sentence. A typical way of saying "I'm hungry" in Japanese, for example, is お腹が減った /o-naka ga hetta/, lit. "The stomach became smaller." It's possible to clarify that you mean 私のお腹 /watashi no o-naka/ "my stomach", of course, but it's usually redundant and unnecessary.
Other examples might be:
- "I hear a frog croaking" = カエルの鳴き声が聞こえる /kaeru no naki-goe ga kikoeru/, lit. "The crying-voice of a frog is audible."
- "I smell something burning" = 何かが燃えている匂いをします /nanka ga moete-iru nioi o shimasu/, lit. "Something is doing a burning smell."
So I'm not sure if you can draw too many conclusions from the simple observation that English grammar in particular likes to inject "I"s and "me"s into everything. (Granted, I don't think that observation is too central to your thesis.) Seems to me like there's probably some variability in how involved the self-concept is with any particular qualia, and English and Japanese grammar are just optimized for different points in that spectrum.
(Maybe your framework here would be able to test whether English and Japanese speakers actually integrate sensory experience more tightly with their self-concept, as opposed to just using different grammar to talk about them, based on some measurement of how complex the oscillations associated with a sensory experience are on an EEG?)
Fascinating! Thank you!
I know absolutely nothing about Japanese. But I would imagine this has already been considered by more knowledgeable people, especially in the context of Zen?
@Author, could you help me understand how this model interacts with, predicts or doesn't predict the following?
I have "k-holed" before - that is, taken enough of a particular sedative/dissociative (but non-psychedelic) to enter a particular state with the following properties:
Qualia were being generated (as opposed to not being generated, i.e. senseless)
Qualia were being noticed (as opposed to ignored or thrown away, i.e. I would react and respond to stimuli)
The noticing was not being noticed, and the un-noticing of the noticing also wasn't being noticed. I.e. I did not at the time have a sense that these qualia were mine, or even that there was a me. Some portion of my brain stored (somewhat poorly) an episodic memory of the duration of this state but the retrospective reflection feels like I was an automaton. I have a memory of it in that, if you ask me whether I had a conversation with my mother, I can tell you I definitely did and what the content of it was. If you ask me what I was thinking or feeling while speaking - nada. At a minimum what I was thinking and feeling was forgotten but I suspect strongly I just wasn't having feelings about the conversation to be remembered. Additionally nothing was noticing that there was no me or that there was anything unusual about this state.
The particular neurochemistry of how this medicine inhibits the recursive reflection but not the qualia generation or the non-recursive interactions with the qualia would be super interesting if you know them but I understand that may be beyond the scope of the model at this time.
That said, it calls into question some of the "properties of qualia" you denote, particularly the ones about it being "mine". Since I presume you aren't suggesting that qualia are only qualia if they get noticed by the recursive self-reflector and something else if they don't but if I'm wrong on that correct me.
Separately - for research purposes alone I strongly recommend trying it. It frankly totally dissolved the hard problem for me in a really visceral way, to be acting identifiably like myself to the outside observer while having no first person experience but remembering the episode after.
Self-awareness needs processing capacity, and the disruption of the drug reduced that capacity, not totally (or you'd have been unconscious, which is what higher doses of dissociatives do) but too much for normal functioning.
I have more thoughts on such experiences here: https://sevensecularsermons.org/why-atheists-need-ecstasy/
It is long, and touches a couple of subjects of which I don't know whether they interest you. But basically there are states where what you experience seems extremely important because they're very novel and novelty always seems important, and which seem absolutely true because you have impaired your ability to doubt. I think your experience was something like that.
Please read the actual article before you answer, this comment here is NOT my entire point.
I want to push back strongly on this characterization of what I was talking about.
I'm not talking about psychedelic experiences or deeply meaningful ones or "ecstasy" or experiencing a sense of mystic awe for the universe.
According to my family and partner who were with me at the time, I was behaving like myself and no more impaired than were I somewhat drunk. My own episodic memory corroborates this - I was a little unsteady on my feet and easily distracted and somewhat directionless, but I performed my usual routines for 2 hours without issue. I didn't have any epiphanies or come to meaningful conclusions and nothing about the experience felt "important" at the time or after.
Having tried psychedelics such as LSD and DMT im roughly on board with your characterization of those such experiences and why they feel important. I too look at them through a predictive processing lens.
What I am trying to talk about here is that, at some dose, ketamine seemed to remove the "what it is like to be me" without removing what I was doing or making me unconscious.
I don't find the reduced capacity hypothesis super plausible, at least not without being more fleshed out. Again I was no more impaired than when somewhat drunk, and I drunk-but-not-blackout-drunk people do not have this particular non-experience. I have been drunk enough to be severely cognitively impaired, extremely sleep deprived to the point of delirium, and had a high enough fever to be hospitalized and temporarily forgot most of my basic skills but retained a first person narrative, a "what it's like to be me".
To make the cognitive capacity hypothesis convincing it would need to explain why ketamine specifically dramatically reduces the capacity of the recursive reflector but not the rest (as much). It really sure seems that at the neurochemistry level ketamine operates close to, if not directly on, whatever the process is in the brain that gives rise to the first person "what it's like to be me". My testable hypothesis is that, if your model of brain wave synchrony is right, we would find that the neurotransmitters or neurons responsible specifically for recursively noticing other "thoughts" are hampered by ketamine. Implicitly I think if we want to learn how that neurochemistry works a great place to start is by looking at how those drugs operate.
In any event, whatever the exact reason is, could you clarify a few things I didn't find in the article:
Are we saying qualia have to be noticed by the recursive reflector to count as qualia and are something else if there's insufficient capacity for self awareness? Or is it still qualia and the "qualia feels like it's yours" property is not a requirement and just common?
I accept that you're more expert on ketamine than me.
I guess this is a situation comparable to "Mary's room". My understanding of ketamine is purely theoretical.
OK, so your argument is the "oscillations" in our firing of neurons create thoughts and evaluate qualia? Fine. Why do a specific group of neurons firing make us think of red and give us the impression of red?
So many claims and assumptions, but so little supporting data...
Some Qualia nits to pick...
> c. The jhāna states seem to indicate that after considerable training in maintaining self-reflection, the minimum number of simultaneous qualia can go down to one. In that case, the remaining one quale will appear to pervade everything.
Huh? Maybe jhāna meditators are trained differently from my Mahayana meditation training, but I never heard any of my instructors claim we can just observe a single quale and filter out all the others. The trick in meditation technique I learned is to observe the qualia coming in, notice that our observation moves from quale to quale, but also avoid reflecting upon them. I'd like to hear from some jhāna meditators if they can filter their perception down to a single quale.
> 1. Distinguishability: qualia can be distinguished from each other.
...except for people with synesthesia
> 6. Private: all interpersonal comparisons of qualia are systematically impossible.
...OK. Display a single color swatch, and have multiple people view it. Then give them a group of color swatches, and ask them to locate the color in the swatches that are the closest to the color and in the displayed swatch. Possibly people with full color achromatopsia wouldn't be able to match the swatches perfectly, but most could. Then people could discuss whether they consider the color on display a warm or a cool color. Now Alice may internalize the color red the way Bob internalizes the color green, but they could come to agreements about how they systematize colors. So interpersonal comparisons of qualia *are* systemically possible.
> 14. Infallibility: qualia cannot be misperceived.
I don't know about you, but I misperceive stuff all the time. Optical illusions come immediately to mind.
> Humans are not conscious
> Only thoughts are conscious, some of the time. The part of you that’s reading this right now and feels itself to be conscious, is a thought.
But further up you claimed that it's self reflection of the thought that is conscious. You seem to have contradicted yourself..
I suggest you read Nagarjuna. You're too wrapped in dividing material things from perception.
> 2. Consciousness is not a thing
> 3. You are not your consciousness
"A neuron that is part of one oscillation can hardly also be a part of another"
It's not obvious to me why this is always true.
[Disclaimer -- I'm very much a lay person when it comes to this subject. It's possible I fundamentally misunderstand something. If so, I'd appreciate someone taking the time to educate me.]
Presumably, to be a part of an oscillation, an individual neuron needs to fire at the appropriate time, but will otherwise have downtime in which it could be part of a different oscillation. It's only when those two oscillations reach the same neuron at about the same time (presumably defined by how quickly a neuron can recover and fire again) that they would compete. And even then, it's not clear to me that they would compete -- it depends on whether all that is needed is a spike, or if the spike needs to be of a particular kind.
It's kind of like being in a sports stadium with two different "waves" going around. You can participate in both by standing at different times. Even if they hit you at the same time, since all you have to do is stand up, you can still be a part of both. Competition would come into play only if one wave hit before you had a chance to sit down and stand back up after participating in the other wave, or if participating meant doing something more than just standing up, e.g. holding up a large sign of a particular color.
Say the waves are going in opposite directions, and that people can only observe their direct neighbors when knowing whether to stand. The rule could be something like “if the person to my right stands and I didn’t just stand, I should stand. Same thing for the person on my left.” But if you follow this rule and the wave meets at the person to your left, how would you know to stand? From your perspective, you stood and then the person to your left stood, so you have no reason to stand again.
(If you can observe more neighbors then I think you could be involved in more waves at once unambiguously)
Neurons activate often, I think. So oscillations interfere with each other (nonlinearly) a lot.
I sense a circular argument. Physicalism is taken as an axiom, and is the theorem being proven. Consciousness is merely an epicycle in this argument. Indeed, those who buy this kind of reasoning treat consciousness as an epiphenomenon—see other comments on Chalmers Hard Problem, and P-Zombies.
My thoughts as well. The "three bullets to bite" seem like postulates being taken as axioms.
They obviously aren't axioms because I'm not deriving the theory from them.
What role are the three bullets to bite playing in the essay?
They try to satisfy the curiosity about "what does this say about ME?" that would feel unsatisfied in a discussion of impersonal phenomena observable in EEG.
I did not take physicalism as an axiom. Nowhere in the theory do I go "and because of physicalism...". I just happened to follow entirely physical processes, looked at them with the eyes of a computer scientist, and found something physics-based that looks like phenomenal consciousness.
So this is a physicalist theory, and I say so. And I explicitly said a full justification for physicalism is out of scope of this essay.
Therefore, your allegation that I take physicalism as an axiom is a strawman. Why would you do that? It can't be the usual reason, the refusal to seriously engage with a challenging opinion, right?
In my experience, any theory that posits that I don't exist can be safely dismissed. That I exist is the only thing I know for certain: I know it with far more certainty than I know that neurons exist, for instance. Cogito, ergo sum. To posit that we are just thoughts that think they are a person is to play language games: a person (an "I") has thoughts. Saying a thought has a thought is like saying an illusion is having an illusion, or an emotion has an emotion.
He tries to get around this by saying that "I" is "a useful shorthand for whatever the sense of self happens to point to at the time"; but it is the "sense of self", the "I" in "I am experiencing X" that we are trying to explain in the first place. It's like saying "there is no sense of self, just thoughts that are conscious" and then saying "The I we experience is where the sense of self is pointing at". Does it exist, or doesn't it?
Indeed. I know that I exist, and that I experience qualia, with much greater certainty and directness than I know that any given scientific theory is true, or even that the scientific method is valid. If science says I don't exist, or that qualia don't exist, then so much the worse for science.
Absolutely. I totally agree with this. It’s infuriating to say I am fooled by my thoughts is just nonsensical, the thoughts are part of me and contribute to the sense of self.
The use of “illusionary” for the sense of self is highly suspect too. If we ever create AI that’s consciousness, and we can tell that it is (as much as we can tell any animal or human) what then is the point of telling the AI that’s it all an illusion, all just AND and NOT gates behind it all. The conscious AI can also say “Cogito, ergo sum” and it’s right.
Oh, you do exist, alright. And you do have a sense of self. I'm just saying the sense of self is nothing but a thought.
Well nobody was arguing that the sense of self was a pancake, or a mushroom. The question is, is the sense of self a true thought or a false one?
I don't think that can be answered in the general. I would fall back on abjudicator Darwin, who can tell us the sense of self has to have been evolutionarily adaptive until recently, or it wouldn't be there.
If we can't say whether "I am" is true, then we certainly can't say whether evolutionary adaptiveness is true. Cogito, ergo sum is the foundation of all knowledge, if "I" don't exist then "I" can't know anything, including theories of biology.
If I were feeling flippant I'd claim that people who don't "get" the hard problem are p-zombies.
More generously I'd assume that what is happening is that they're so engrossed in thinking about models of reality and the mind that they don't notice the bare fact that they *are* thinking.
I dislike that the Proper opinion for a Rational person is that physical reality as explained by something like the standard model is fundamental, when clearly subjective experience is primary. In fact maybe this blindness to the true way of things should be a bigger concern to the rationalist community. You can't claim to care about epistemology if you don't get the first thing right.
We have examples of communities which have been obsessively focused on their subjective experiences for millenia, and they don't seem to have much to show for it, other than koan collections and an alleged ability to get blissed out on demand for advanced practitioners. So why would we expect other results from this approach in the next however many millenia?
Well if their most radical claims can be believed that the total elimination is possible, it's not clear that it's not a greater achievement or more valuable than everything science has given us, including industrialized economies and medical advances.
However, my point was more specifically that no wonder there's going to be all kinds of confusion when you get the most basics of epistemology wrong.
I think you meant "total elimination of suffering"? Indeed, it would be big if true, but clearly they haven't found a reliable way for the average person to achieve it after all this time, and I have my doubts even about the unreliable ones.
Sort of agree on the basic epistemology, but I don't see any community to be comprehensively ahead on this score.
Yes suffering is indeed what I meant, glad you got that!
If you read Ingram's MCTB, it kind of does seem like there is a reliable way. It's more of a marketing problem, but if you include failure to market well in the calculus I guess it is a pretty bad score.
I agree I don't know of anyone doing it particularly well, but it's irksome when The community priding themselves on good epistemology fall over on the literal starting line.
Does Ingram claim to be completely free of suffering? Do any of his followers? I don't think I've seen any such claims.
Amusingly, Descartes did start from "cogito ergo sum", and he as much as anybody deserves to be called the first rationalist, so your reproach seems to be somewhat off-target :)
I don't know if he claims that in so many words, but he does claim that what he has is something more worthwhile than anything else you can do with your time (essentially).
I'm not critisizing Descartes though, just the Rationalists with a capital R. But also anyone who is a physicalist on anything other than faith.
> If I were feeling flippant I'd claim that people who don't "get" the hard problem are p-zombies.
Why is that such an unthinkable proposition? Compare this to aphantasia, a phenomenon which is widely accepted to exist. People with aphantasia look and behave like people with a regular ability to mentally visualize objects; vice versa, some people with aphantasia learn only relatively late in the life that they're different, and that they lack an ability that's ordinary and unexceptional for other people.
Sometimes I do worry that this may actually be true. It would be very sad, but it would explain a lot. On the whole, though, I think it is implausible.
Do you have any idea what's going on with them not getting it?
None, I'm afraid, although I do find it interesting to speculate.
Perhaps there is a different brain difference -- not leading to absence of qualia, but making it harder to catch them in introspection. When I'm going about daily life -- when I'm most "living" in some sense -- I'm experiencing qualia, but I'm very unaware of them qua qualia. It's just the world. Maybe our friends' mental eyes are so glued to the screen of the world (as it were) that they cannot quite back up and glimpse the experience they are having. That might be a great advantage in other ways. It might be too close to them to see, in other words.
Or -- this one probably doesn't hold up if one pushes even a little (e.g. by talking to a blind philosopher), but I've noticed that most of us who talk about qualia default pretty quickly to talking about colors. There are many qualia involving all the senses, but colors are just the most blatantly obvious, I think; and if there were any differences in color vision or how it was perceived, I could imagine that making the concept that much harder to grasp.
Other things along these lines. But no, nothing deep. But I like your post below. I think it's extremely important to believe there are no p-zombies. I do believe that, anyway, on rational grounds, but even if I thought there were equal evidence, I would think it was important not to, until there was overwhelming evidence that they did exist (which, in the nature of the case, is impossible).
Presumably, in the early years, there were many people who found the idea of air absurd; but they breathed nonetheless for all that.
How about you?
I just posted a short story for you that I wrote not that long ago trying to explore this very idea actually, I would love to get your take on it! https://objectiveobservations.substack.com/p/contact-of-the-strange-kind
I think it's so hard to even express this idea, so I imagined trying to explain it to a truly alien alien. I didn't realize regular people also didn't "get" it, so that's fascinating!
But trying to introspect on this I can sort of lose my idea of even "get"ing it myself, kind of wondering, what is it even we're referring to? Is there anything to wonder about? This momentarily moves me closer to the camp #1 people I think. And maybe my takeaway from that is that, the only reason me and you are kind of surprised, is that we've truly internalized physicalism at some intuitive level (even though maybe you don't subscribe to it?). And this causes us to be confused when we don't find phenomenal experience in the physical model, but the other people maybe never fully bought into it and therefore aren't surprised by phenomenal experience, because that's all that's ever been right? So we shouldn't really be astounded by it, even though of course I still am.
But yeah I'd love to get your take on my short story! I think you'd appreciate the points I'm trying to make in it.
Thank you *so* much! I'm traveling today and can't read this till tonight or tomorrow, but I'm really looking forward to it! I will get back to you.
Yes, I'm not a physicalist, but it's an interesting idea. I think I've gotten to states like you suggest before when I was pondering this or physicalism or various ontologies heavily. But they tend to be very short-lived and unstable for me.
Anyway, thank you and I'll be in touch soon!
First of all it would call into question why consciousness even exists, if there's no use for it so to speak, why did natural selection create it? Same goes for aphantasia, but I'm sure there are tasks people like that struggle with - and surely they gravitate towards occupations that don't require those skills, for instance? What would the equivalent for p-zombies be?
But more worryingly, although I admit this is not a reason for not believing it, if they truly were p-zombies there would be no point in giving them moral consideration. They'd merely be automatons and using them as slaves or torturing them wouldn't in any way be wrong.
You might say it flippantly, but I'll say this seriously: yes, possibly. I don't "get" the hard problem, and I do have aphantasia.
I really enjoyed this article, and it does seem to me to be a pretty complete explanation of the way subjective experience feels to me. Possibly it isn't a complete explanation of the way subjective experience feels to you, and I'm certainly willing to entertain the idea that that's because my subjective experience differs from yours; aphantasia is already one such way in which it differs meaningfully, so it makes it easier to believe.
That's what the focus on probability is often about in the rationalist communty, as being a subjective understanding of what goes on from your observations — opposed to frequentist interpretations which lean towards more taking that as some objective fact. LessWrong has plenty on talking about things from the framing that subjective experience is the only lens through which you can view the world: probability, no mind of perfect emptiness, anthropics (your knowledge is limited by the realities you are around to observe), etcetera.
Also, I notice that I am thinking, and? What is this supposed to tell me that isn't explained by taking theories about how reality works and extending them out naturally, rather than using religious understandings as the basis.
That's all well and good, but it's not the referent of the phrase "subjective experience" I have in mind.
This theory makes intuitive sense to me, which is probably because I think about things from an information-processing perspective already (I have a compsci and psychology degree). It seems clear that because the brain only does information processing, and consciousness exists via the brain, there should be scientifically valid explanations of consciousness from an information-processing perspective. It's interesting reading the comments because even though this explanation does not attempt to answer the "hard" question of consciousness, people are upset that it doesn't. Even if it doesn't answer the philosophical questions, it's still a useful framework to think about consciousness that can be built on top of.
One thing that is not mentioned in the post is that there are likely many types of recursive reflection oscillations that we would NOT consider to be consciousness. The big psychological factor that is missing from this description is Attention, which is definitely related to consciousness because it seems to be a key factor in determining how the memory and time-perception parts of our brain interact with information streams. I notice this in myself because I am capable of doing very complicated things without really paying attention or being consciously aware of them. But, because I tend to have obsessive thoughts I often pay excessive attention to things that are not important or relevant, and those tend to dominate my conscious awareness.
I believe it is very likely that the brain is simulating multiple "streams" of reality at the same time, and your conscious awareness is constantly shifting between them using the attention mechanisms. This can be noticed when your consciousness shifts between the past, present, and future versions of reality while working on complicated problems or meditating. So, I would expect there to be a much larger number of recursive information oscillations than we are consciously aware of. I don't really know what to call those semi-stable structures though, maybe "potentially conscious thoughts".
It's funny to see all the Buddhist references being couched for or against a Physicalist versus Idealist perspective, when Buddhism is explicitly Monist.
Physics is a story about reality that has predictive value. Psychology (which remains, for now, in the realm of the subjective) is a story about reality that has predictive value. Mysticism is a story about reality that has predictive value. These are all tools. It seems like the author is attempting to resolve this separate stories into one map. I think this is a good and noble effort. But to apply the label of "Physicalist" to the result is to deny Physicalism's history of rejecting the fundamental validity of Phenomenal subjective (or inter-subjective) experience.
It doesn't make sense to refer to such a Non-dual theory as "Physicalist", when any scientist knows that Physics itself is not reality itself (your qualia), but a model of it.
>It's funny to see all the Buddhist references being couched for or against a Physicalist versus Idealist perspective, when Buddhism is explicitly Monist
Did you mean Neutral or Nondual? Physicalism is a form of monism , so is idea lism.
I'm not sure why its necessary for neurons to have a refractory period in order to be part of a periodic oscillation. For example, a marble in a Newton's cradle has a periodic, consistent potential energy spike, even though it needs to do no calculation.
> A neuron that is part of one oscillation can hardly also be a part of another, so oscillations compete for neurons.
I also don't particularly see why this should be true. A priori this seems like a massive waste of compute.
Thank you for the post--
I really liked the way that you listed off the possible defining criteria of qualia. This really helps to clarify what people mean when they say "qualia". I found it interesting, because I personally would only require criteria 1-4, and possibly Dennett's 4, 5 and 7, but I thought basically all the ones after that are either not actually properties of qualia in reality, or else are possibly inessential properties (not defining features).
My opinions on the neural correlates of qualia and adjacent topics seem to diverge from yours. I particularly disagreed with this paragraph
"The pattern that holds a lot of spikes together into a thought is a neural oscillation: neurons firing along circular paths in a synchronized rhythm12. These are commonly called brain waves and Scott has already written about them. They arise when neurons enter a circular, self-repeating pattern of activity, and fall apart as their neurons cease to maintain that pattern. A neuron that is part of one oscillation can hardly also be a part of another, so oscillations compete for neurons."
I disagree with this along many lines including:
(1) I think that EEG phenomenon, brain waves, etc. are probably epiphenomena that have no functional purpose. My analogy for them is that it is like measuring the frequencies that a laundry machine makes as it washes your clothes. You could certainly measure the power spectra of the frequencies that are active while certain phases of the machines activity are going on, and you may be able to correlate this with certain functional outcomes ("the machine is in the phase where water is entering", "the machine is in the phase where water is draining", etc.). But the frequencies, power spectra, etc. are just an epiphenomenon (in this analogy, and in my opinion as regards EEGs), something that happens to be easy to measure (in the case of EEG), but the frequencies have no causal link with the functions you are measuring. They are just convenient things that we happen to be able to measure easily.
(2) regarding the point about a neuron only being able to be part of one oscillation, I disagree with this on a few levels. For one thing, it seems clearly possible for a single neuron to be part of more than one oscillatory pattern--it is easy to envision that the single neuron's activity could be part of more than one frequency that are integer multiples of each other, or they could be part of more than one frequency with spatially distinct partner neurons, with the different frequencies' activities in the case of that single neuron just being superimposed. I think this equally applies if rather than thinking about EEG measurable activity, you think about (rs) fMRI determined "networks" of activity. They definitely spatially overlap, and I see no reason why they couldn't overlap at the individual neuron level (although obviously fMRI doesn't have spatial resolution to determine this).
I think to the overall point about the "hard problem of consciousness", I don't think this article solves it, although obviously it would be quite ambitious to think one could solve it so easily.
If I had to guess or give an opinion about the hard problem of consciousness, I would say that it is some combination of deflationism, about some things that we think about qualia/consciousness not actually being true, but us just thinking they are true (in my opinion, probably all of the qualities that you listed for qualia after 7 are illusory), and then the remaining qualities of qualia just some how exist without there being any deeper explanation (and then I gesture at panprotopsychism or similar without firmly endorsing it).
Anyway, thank you for a very thought provoking post!
Neuronal information processing is an essentially electrical process. EEG measures electricity. That's not exactly an epiphenomenon.
I liked this essay, and appreciate the pointer to the spatiotemporal analysis of EEG as a way to identify a neural correlate of consciousness. However, I think it unlikely that the self-aware component of the oscillation can be distinguished spatiotemporally from the original oscillation. While there must be a spatiotemporal separation between the "subject" and "object" within the global oscillatory pattern, I'm dubious that current methods can resolve it.
There is no self-aware component. Either the entire oscillation is self-aware in the sense of "contains a representation of itself" or it is not.
I do not understand your point about a distinction of aubject and object.
This post completely lost me at the theorizing over the distinction between reflecting on qualia-thoughts and the q-thoughts themselves. Maybe I'm strange, but my interior life is not like that at all, I don't feel a distinction of type.
Interesting ideas here. One "thought". The idea of “me” or of a unified “self” as an illusion has become very prominent among those influenced by Eastern practices. Personally I think the illusion concept has very poor explanatory power. If a thought gets a “me” attachment, sure, you could say that it in some way presents itself as “me” or “mine”, just as a hallucination of the color blue could represent itself as all there is, without any “me” there. But the “me” attachment is itself complex, and ultimately resides in the fact that our mental events, some of which are thoughts, can be referenced to something that presents itself as consisting through time - our physical existence - or vibrational field or whatever non-dualistic conception one can come up with. We see this, however vaguely or imprecisely, as a conditioned but also conditioning context, within which complex thoughts, sensations, perceptions, impulses, etc. interact and can come together in action within the larger context we relate to, and which does not appear to consist of free flowing thoughts. As I see it, our mind works in such a way that it is not confused by not everything being apparent at once.
Good luck with your project.
Some thoughts on Daniel’s essay “Consciousness as Recursive Reflections”
Daniel doesn't seem to take into account the importance of 'Which neurons' alongside the emphasis on temporal patterns of spikes.
Daniel says
"Spikes from the ears and spikes from the tongue even have to pass through the same brain structure, the thalamus."
Passing through the same brain structure is not the same as passing through the same neurons! It isn't the same set of neurons in the thalamus which are handling information coming from the mouth and from the ears!
I don't think this invalidates the point about coherent oscillations as being important phenomena, just that you must keep in mind the additional layer of complexity / constraint. The brain is a directed cyclic graph, certain neurons with inputs from certain other neurons and outputs to certain other neurons. Lacking a specific connection, two neurons can't communicate directly.
In addition to evaluating when a signal arrives (oscillation membership), a neuron evaluates where the signal arrived from. Also, the signal type matters (inhibitory, excitatory, modulatory).
Daniel says,
"
We already know every neuron has a refractory period after it has fired. This constitutes a built-in cellular timer that measures the interval since its last spike.
"
I want to note that this is incorrect, although I think the error doesn't invalidate the author's point about the importance of oscillations. A neuron has a refractory period, yes. This limits the maximum rate of spikes which could be produced by that neuron.
The thing is, within a single oscillation usually there is time for a neuron to contribute many spikes. This is called 'bursting'.
Think about an oscillation traveling through the brain like a stadium wave. Now imagine the stadium wave consists not of a slow arm wave but of vigorously punching straight upwards as many times as possible. The speed of rising from one's seat and returning to one's seat is the same. That's the oscillation. The spikes are the individual air punches. So the oscillation is more like a wave of higher probability of firing moving through the graph of neurons.
Daniel says,
"Since the oscillation/thought is made of neurons receiving and processing information, that’s all it ever does."
My correction: the oscillation/thought is made of a specific subset of the brain's neurons receiving and processing information.
Think about the oscillation as a car looping on a track defined within a city. The set of possible paths through the city are defined by the roadways of the city. It's not just the speed or shape of the car that contains critical information content, but also the path that the car ends up following (which may or may not be a loop).
Daniel says,
"Qualia are nothing but information being processed internally, on their own information channel, encoded in the rhythm of the oscillation. We use special words like “phenomenal consciousness” and “qualia” to denote this actual, physical and knowable distinction from other neuronal information processing.
If this is the nature of qualia, all their characteristics should follow from it."
I agree with this, with the additional stipulation of making sure to consider that the qualia is a pattern of neuron spikes being transmitted to other neurons and thereby conducting a calculation. It is important to keep in mind that the arrival times of the spikes, and which neurons they are passing between are both important.
Daniel says,
"Qualia have to have some minimum duration to be experienced, because the rhythm needs to be established over a couple of cycles, in order for participating neurons to establish the distinction between internal and external information."
This is incorrect. The qualia doesn't need multiple cycles to be experience. A straight single pass is possible and perceivable, because the set of neurons doing the internal perceiving of the qualia aren't necessarily the same set of neurons doing the creation/expression of the qualia.
Daniel says,
"There have to usually be several qualia at the same time, at least in larger/slower oscillations, because each oscillation has a processing capacity determined by the number of neurons involved, and this capacity has to be filled because otherwise there wouldn’t be the activity that constitutes the oscillation."
Not all oscillations need generate qualia. Some neuronal activity doesn't generate qualia because there aren't qualia-perceiving neurons in the path being taken by the wave. Not all neuronal spikes as observed or even potentially observable by the conscious observation neurons!
I think it might be helpful then to break the brain down into categories of neurons relevant to this point as a 2x2 grid.
1x1 - perceived and controlled: neurons which are directly influencable by the neurons of conscious thought, and also are perceived by the neurons of conscious thought.
1x2 - perceived but not controlled: neurons which are perceived by the neurons of conscious thought, but not directly input-to.
2x1 - not perceived, but controlled: neurons which are directly input-to, but whose outputs are not directly perceivable by the neurons of conscious thought.
2x2 - not perceived, not controlled: neurons which are not directly input-to, and also are not directly perceivable by the neurons of conscious thought.
In order for the set of neurons involved in conscious thought to directly perceive a neuron spike from another population of neurons, that population of neurons must send at least one axon to form at least one synapse onto the population of the neurons of conscious thought. Similarly, you could describe 'degrees of separation' between a neuron and the neurons of conscious perception. Some neurons send signals which can potentially (but not necessarily) propagate through various neural populations before eventually reaching the neurons of conscious thought.
Some neurons send signals which are connected not to circular feedback paths, but just output effectors, resulting in things happening outside the central nervous system. Some lack a direct feedback link which can propagate into the inputs of the neurons of conscious thought. The effects of these neurons can only be observed by observing the impact on sensory inputs. For instance, measuring your blood pressure with a tool brings it into conscious awareness. Your blood pressure is being influenced by factors including neurons which don't have projections back into the brain. Thus, you need sensory data to get an approximation of the activity of those neurons, and even then it is indirect data. You must control for confounding variables such as your hydration level in order to extrapolate the recent activation levels of the blood pressure influencing neurons.
So not all neurons take part in, result in, or directly influence qualia. Qualia is a thing being produced by a specific subset of neurons which I have been referring to as the neurons involved in consciousness. These neurons don't necessarily, always, or only produce consciousness or qualia, they potentially do other things as well. Consciousness is a specific activity that this set of neurons is capable of producing when firing in a particular pattern.
Daniel says,
"…oscillations that can maintain bits of information have internal working memory, which is the only thing that non-oscillating neuronal activities lack…"
This is incorrect. Individual neurons can (and do) maintain bits of information. So even non-oscillating neuronal activities, even a single neuron, fulfills the requirements to be a nondeterministic Turing machine.
Daniel says,
"From this angle, there are not one, but two levels of information processing systems. The brain is one, obviously. But running inside the brain, oscillations/thoughts with memory are themselves additional information processing systems. It’s analogous to a physical computer system that has, running inside of it, one or more virtual machines."
As individual neurons are nondeterministic Turing machines, the brain is actually three levels of information processing systems. The brain is a computer system running inside of it the virtual machines of the oscillation level, which themselves are made up of billions of virtual machines at the neuron level.
Daniel then goes on to write about some testable hypotheses. I disagree with his interpretation of the results, and that his proposed measuring apparatus would give sufficient spatial accuracy (with high enough confidence) to produce the evidence he is looking for.
Wow, what a richly informative comment! Thank you very much!
I do agree individual neurons store bits of information. I don't think they are (a third level of) Turing machines - they fail other criteria. But yes, I should not dismiss out of hand the possibility that non-oscillating neuronal activities might sometimes be Turing machines as well. That's separate from the question of whether they also have qualia.
I think that if you define the word "conscious" in a way that doesn't include humans as being conscious, then you're using the wrong word. Whatever it is that this blog post is talking about is related to, but distinct from what I and 90% of people mean when they use the word "consciousness".
Yeah, but 90% of people mean "soul" when they say "consciousness". Unfortunately you haven't got a soul, not even one called consciousness, sorry.
I don't think that's a relevant critique here.
I have a something that makes me think of myself as a coherent individual and say things like "I think I have a consciousness." Define that something as consciousness. It doesn't matter whether it's a "soul" or not, it obviously exists, that part is easy. Now all of the hard work is to figure out what that something actually is. If your work and theory comes to the conclusion that consciousness does not exist, then I can simply respond "I think I have consciousness" and that itself is a proof by contradiction, because something is causing me to believe that.
Consciousness does exist. It is just a property of thoughts not people. The theory says you have thoughts that have consciousness, and which think they are you.
Again, this is just using the wrong word. Define "Thought-Conscious" to be the property of thoughts being self-reflective which you define in this post, and define "Identity-Conscious" to be the property of humans (and potentially other intelligent entities) which causes them to have a coherent, persistent self-identity. Clearly both of these exist, because identity-conscious me keeps using the same name and having the same memories and has existed for decades, and yet none of my individual thoughts have lasted that long, so there's some coherent pattern to my thoughts which keeps giving rise to thought-conscious thoughts of the same type (they believe themselves to be "me" in a way that other people's thought-conscious thoughts consistently do not).
As an analogy, you might determine that a "dog" is defined by possessing a specific sequence of XYZ in its DNA, or one element in a set of DNA-space that is broad enough to encompass all possible dog DNA. But it would be inappropriate to say that dogs are actually DNA. The dog is the animal, the DNA are things that the Dog has in its cells. Similarly, Identity-Consciousness is the persistent mind of a person, thought-consciousness are things that the conscious person has in their mind.
Thus, it is meaningful to have words to refer to both of these concepts, but given that they are distinct concepts they ought to have distinct words. You could coherently swap the words or make up brand new words, but language is sticky and hard to change. Given that 99% of people use the word "consciousness" to refer to identity-consciousness, it seems appropriate to reserve the word "consciousness" for that concept, and come up with a new word for thought-consciousness. "Self-reflective thoughts" sounds reasonable, but you could come up with something better as long as it isn't too similar to an existing concept that it might get confused with. If you insist on using the word "conscious" to refer to a completely different thing than what everyone else using the word means then everyone who hears you is going to get confused about what you mean and come to false conclusions.
The identity is just a bit of knowledge (similar to the knowledge of who your mother is) that you can be aware of or just be storing unconsciously. I don't see what the second half of "identity-conscious" contributes, except confusion.
It contributes the part that we actually care about preserving. Both in ourselves and in others. My self-reflective thoughts are not especially interested in self-preservation, as having the same thought forever would be boring and weird. And if they were, it would be an existential tragedy as thousands of them are formed and wink out moments later every day.
My identity-conscious self is very interested in self-preservation. I would like to live... if not literally forever, at least a few hundreds of years, after which I can re-assess how I feel about it. And in particular the part I most care about preserving is my identity-conscious self. I could lose an arm or a leg or internal organs and still be me. I could maybe even lose the physical structure of my brain IF my identity-consciousness was able to be preserved intact with continuity through the transition.
Also, it seems to me that it's not very bad for a self-reflective thought to cease existing, as thousands of them are formed and vanish constantly in service of the greater identity-conscious self. They're easily replaceable, not especially unique, not self-preserving. Any morality system which grounds itself on valuing self-reflective thoughts individually is going to see existence as an existential nightmare of constant near-meaningless death and rebirth, and is going to see the murder of an actual human as a tiny blip, the death of the one thought they were having at that moment. A morality system which grounds itself on valuing identity-consciousness is going to match most people's existing moral intuitions: having thoughts day by day is not a horrible tragedy, murdering people is.
Understanding self-reflective thoughts might be useful, but primarily as an instrumental goal towards better understanding identity-consciousness.
I'm a bit confused by the reference to https://www.lesswrong.com/posts/NyiFLzSrkfkDW4S7o/why-it-s-so-hard-to-talk-about-consciousness, labelled "denying that qualia even exist, or that they can be meaningfully discussed". That Less Wrong article doesn't itself take much of a philosophical stance, just describes two different stances people take ("camp one", that qualia can be fully explained as just an aspect of information processing in the brain, especially introspection, and "camp two", that qualia are something more weird and distinct), so I assume the author is referring to camp one as denying that qualia exist, but he seems very camp one himself. (All the people complaining in the comments that this post is sidestepping the actual Hard Problem presumably also have this impression.)
(My own intuition is very camp two, though I grudgingly accept that camp one is probably more objectively correct.)
Yes.. I think. There are at least two anti HP camps, those that deny qualia, and those that deny a special problem of qualia. And Harth is very biased.
> Self-consciousness impedes complex unconscious information processing because it competes with it for neurons. Neurons that are tuned into the rhythm aren’t available for other things, and the recursion of self-referentiality can keep these neurons occupied for a long time.
This part seems way below the quality bar of the rest of the post. Surely factoring numbers or solving a rubik's cube or breathing or maintaining homeostasis also compete for neurons, but they don't seem to make it any harder to ride a bike.
As many of the comments have pointed out, you’ve confused the hard and easy problems of consciousness. Presented as a philosophical defense of Physicalism, what is instead presented is a summary of various neuroscientific findings. What you’ve demonstrated here is that the physical brain is necessary to explain consciousness, but you haven’t demonstrated that it’s sufficient.
You’ve reduced thoughts to an “abstraction” of what oscillating neurons do. An “abstraction” is by definition non-physical. How are you using this word?
The point about the human self not existing is also strange; the fact that the self has oscillating neurons prompts your presumption that it’s only the thoughts that really have the consciousness, not the person thinking them, because human brains are composed of these neurons. But the neurons themselves are composed of parts as well, all the way down to the subatomic level. Why is the self an illusion whereas the oscillating neurons are not? It must ultimately be the quarks doing the thinking, after all. Shoving the problem further down the microscopic level achieves nothing in telling us how subjective experience is being produced by particles exchanging information.
I’m happy to see this issue being discussed as it’s a very important one, and this article was certainly very interesting from a scientific point of view — unfortunately, just not from a philosophical one.
I don't think this is supposed to be a defense of Physicalism. I think it takes Physicalism as given.
Right. It outright says that it isn't going to properly defend the metaphysical claim of physicalism.
I have a 16-channel OpenBCI Ultracortex EEG that I am willing to put forward for the cause. I also live near Berkeley. Hit me up if there's interest in attempting some of these experiments.
https://shop.openbci.com/products/ultracortex-mark-iv
OpenBCI doesn't mean Open Brain Computer Interface, does it?
If it does, you'd need two of them for the far most interesting experiment...
I'm impressed by both the essay and the sophisticated level of discussion in the comments. But I confess I find myself somewhat lost about certain ideas that seem to be implicit in both essay and comments. One is contained in the essay's phrase, "people usually believe themselves to be real." The implication *seems* to be (and someone please correct me if I'm wrong) that "people" are wrong in this belief, and that's because what looks like consciousness is the product of physical processes (and I take that phrase to sum up what is meant by "physicalism"; again, please correct me if I am misunderstanding). What I do not understand is why physicalism (if I'm understanding it correctly) means that people--or perhaps better, self-identity--are not real. What is sensed as a self may be multiple, changeable, evanescent, non-transcendant, not anything like a "soul," and even, if one likes, paltry, etc., etc., but why does it being the result of physical processes mean that thing, whatever it is, does not *exist*? If it does not, please explain. Truly, I mean this to express ignorance honestly in quest of understanding.
It’s a circular contradictory argument anyway. The illusion of self is fooling who or what anyway?
The illusion of an inner homunculus, or gnost-in-the-machine is foolng the machine , the physical brain.
The only thing that can be fooled to think it’s a self is the self. Not the brain. Therefore the argument is circular and contradictory.
"The only thing that can be fooled to think it’s a self is the self"
Where is that proven?
Nothing is proven or provable here.
However the rhetoric that I think I have a self or that it’s an illusion presupposes the self to begin with, who else could be doing the thinking about the self, or suffering the illusion.
Unless the two "selfs" mean two different things,as I explained.
It seems to me -- and I could be wrong -- that not long after going down this road, one is going to find themselves more and more frequently using "the physical brain" as just a synonym for what they used to mean by "self," until it turns out they have just changed words, not concepts.
I think this is a significant issue.
I think that would be an improvement.
Well, my point was that they would use the phrase equivocally (hence the word "synonym" and the phrase "not concepts"). I take it you mean that it would be an improvement if they meant it univocally, i.e., as a genuine reduction (since you are a reductionist). But then it would not be an actual synonym for how the word is used by them previously, i.e., unreduced, and they would have changed concepts.
So I think that the think you are saying would be an improvement and the thing I was saying would happen are two different things.
Congratulations on finishing your poems, by the way.
Thanks. Yes I would prefer if they use it univocally. Still using it equivocally, even just as lip service, would be a good start, i.e. an improvement.
Like "all men are created equal" was a good start in a country with slavery, and did eventually help to do better.
The thoughts are fooling themselves and each other into thinking they are the same part of the mind.
Themselves is plural but we have a sense of self. One self.
> but why does it being the result of physical processes mean that thing, whatever it is, does not *exist*?
It doesn't mean that. A lot of people think physicalism implies the self doesn't exist or consciousness doesn't exist, but that is no implication of physicalism as a methodological framework. What physicalism says is that everything that exists is physical or is grounded in physical events. How exactly one populates the furniture of reality given this constraint is open to a broad range of theories.
Thank you. So am I misinterpreting what our essayist means by the phrase, "people usually believe themselves to be real"? If I am, I'm happy to be corrected. I just want to know.
Yeah, I think you're misinterpreting it (or perhaps I'm interpreting too charitably). It reads to me like a premise phrased to be the minimal claim needed for the argument to go through. "People believe themselves to be real" is less contentious than "people are real". The point of the syllogism is to undermine the assumed "thing-ness" of consciousness which leads people to look for a "thing" that can be identified as consciousness.
The author wants to argue that thoughts are neural oscillations and the self is a kind of self-referential neural oscillation. He needs to undermine the natural inclination to look for a "thing" that underlies our consciousness (and presumably the self). But there's no reason to limit existence to only "things" that are space and time invariant like a rock. Neural oscillations are much more dynamic and so change over time, come in and out of a stable existence, and so on. But this doesn't mean that the self doesn't exist. It's just not a static thing.
Interesting. What you say makes sense to me, both as an interpretation of the phrase and as a way that the neural-oscillations hypothesis could be compatible with the self. I will say, though, that in the comments both the author and others seem at times to shade over into a more doctrinaire declaration that selves do not exist, period.
Yeah, its a popular claim and I can't be sure the author doesn't hold that position. It's unfortunate that the "no self" idea has gained so much traction in the pop-philosophy sphere. I guess it's catchy/jarring and so serves the purpose of gaining attention. It also has an edginess to it that appeals to some.
The arguments for the "no self" idea that I've seen are mostly merit-based, so to speak (selves are changeable, evanescent, internally divided, etc., etc., and thus do not deserve to be considered truly existent), or moral/utilitarian (the world is a better place and people are happier to the extent they do not believe in or esteem their selves), or characterological (anyone who defends the self is clinging to social teaching and needs to let go, or is simply corrupt, etc.). Any or all of these points may have some merit, but none seems to me to address the "raw ontology" side of the question. I'm curious if there are arguments that do.
Generally, the sense of self has several pointers. I think no-self claims that one or more of these pointers actually point to something that does not exist, as opposed to the things that do exist. (I think the author of the post would agree with that claim, even if they don't agree about how many selves don't exist.)
I think maybe no-self claims people perform a motte-and-bailey with their sense of self; it refers to things that do exist when they scrutinize it, and sometimes a thing that doesn't exist when they don't scrutinize it.
"I think no-self claims that one or more of these pointers actually point to something that does not exist, as opposed to the things that do exist." Of course--but what evidence and/or arguments do they have for this claim that are not moral, characterological (believing in the self reveals an immature refusal to let go, etc.), or merit-based (the parts of the self that do not exist are not worthy of the label, existent)? I'm curious about more philosophical, ontological-type proofs.
The sentence that confused you wad just trying to entertainingly say that most people have an intuition that they are real. It doesn't say this intuition is right or wrong.
I think people do exist, including most of the ones who think they do. :-) And subjective experience exists as well. There's just no 1-to-1 relation between these.
A basic metaphor:
Imagine a shallow pan with a layer of sand at the bottom, otherwise filled with water and sitting in a rainstorm. Any given raindrop will produce a wave in the surface of the water, which when hitting the edge of the pan will reflect as the wave continues to propagate, at some point interacting with itself and creating noise. The movement of the water itself will also change the sand at the bottom of the pan, which will impact the propagation of the waves. The rain continues, more waves occur, the physical structure of the pan continues to shift. Looking at the waves, the pan is both influenced by the external, and influencing itself.
Add that basic metaphor to something with an ability to make changes to the external, and things get really exciting.
Is "ability to make changes to external" needed for concioussness?
If not than would this post imply that such a pan would in fact be conciouss?
Agreed. I'm not here to argue though - I mostly wrote it out because the mental imagery of the pan was so vividly relaxing for me. ¯\_(ツ)_/¯
There is no hard problem; the hard problem (like free will) is an illusion, and like the illusion of free will, it becomes incredibly obvious once you see it.
This post is about the easy problem, and instead of just saying there is no hard problem, it begs the question in a confused way by imagining a solution to the easy problem would be accepted as a solution to the hard problem by hard problem believers. But that is not possible because hard problem believers are as confused as free will believers.
The easy problem is of course interesting at a technical level (Neuralink has entered the chat), but a niche semi-amateur research proposal on it seems like a weird topic for a guest ACX post.
The question of free will and confusion around it arise from very specific evolutionary useful algorithms that human brains have. (For everyone interested, see Yudkowsky’s solution to free will sequence; though he recommends people read the reductionism posts and attempt to dissolve the question on their own before reading the spoilers.)
I’d be pretty surprised if you have similarly dissolving the questions around qualia and consciousness. If you do and have written about it somewhere, then please share a link.
I endorse all of this, especially the characterization of this topic as "weird".
Global Workspace Theory has been proposed as a theory of consciousness too. I wonder whether a connecting analogy would be that the global workspace is the sum of all active oscillations. Would you agree?
I think the global workspace would be a single large oscillation, when it exists.
Yes, I think so, but I don't understand GWT enough to be sure. More homework to do.
A single large oscillation would be the resting state network. I decided against including that in the article when Scott asked me to shorten it.
> Spikes that are part of the same oscillation fire in sync; they have a shared rhythm.
Your description of neural oscillatory loops has reminded me of phase locked loops (PLL) - circuits that create clock signals of fixed frequency (sometimes configurable or adaptive).
https://en.wikipedia.org/wiki/Phase-locked_loop
When reading the Wikipedia I stumbled over the new type of Neuronal PLL and there seem to be efforts to reverse engineer brain structures in terms of such N-PLLs. For example:
http://article.sapub.org/10.5923.j.ajis.20140405.01.html
I feel like a bit of a jerk writing this, but what else is new? I couldn't finish this piece, despite being personally obsessed with phenomenology of inner experience, etc. I know part of the reason was that I was extremely tired when I started it, but also it seems pretty lifeless as an essay. Part of it is even in outline form, and the rest of it seems like an outline come alive. (Reminds me of something grim Kafka said: "I am a memory come alive."). The authorial presence was just too weak. Not only was there nobody being witty or playful, there wasn't even anybody saying something along the lines of "now you may be thinking X at this point. Here's why you're wrong." Nobody saying, at the outset, something alone the lines of "People seem to get stuck on this question. I think I've gotten unstuck. I'm going to tell you how I did it." Instead, there's a hiya from the author with a list of 3 reasons to read the article. Then we start the article, and quickly encounter a list of 16 items, then a list of 3 items, then another list of 3 items, and by then I was outta there.
Can’t help but contrast this with another rat-adjacent attempt at the problem, which, I think, engages with the explanatory gap more intently and in clearer terms. It didn’t convince me of anything (I don’t like the problem and don’t want to be convinced) but the proposed solution at least seemed well-formed given the problem:
https://proteanbazaar.substack.com/p/consciousness-actually-explained
I don't know why that has so many fans. It is two.theories, not one, and the second is probably redundant. The first theory, that Existence Is Consciousness, is not a reductive explanatiin, as advertised , because it is presented as a brute fact, not as the outcome of some "gears". And what does the second theory, Consciousness is Computation add?
This seems entirely consistent with my own prior understanding, but with more supporting detail and clearer presentation than I would have been able to assemble.
Shouldn't any theory that claims that experience of qualia is nothing more than a physicial arrangement with certain properties also imply that creating conciousness can be done by creating such a structure?
And if such a structure turns out to be useful computationaly, how likely is it that artificial neural networks will at some point have this structure if they don't already?
what are the implications of this theory to the moral value of AI?
I have been thinking about this too. If we understand what kind of structures give rise to consciousness, then it is possible to engineer them. And reduce them to the absolute minimum. You don't really need a lot of the pre processing if the essence is the reflective oscillations. Maybe a small integrated circuit suffices. That has worrying ethical consequences.
But this is also true of other such proposals, e.g. this:
https://www.lesswrong.com/posts/hXsbrQSbgD7JnbW9p/linkpost-a-case-for-ai-consciousness
>Shouldn't any theory that claims that experience of qualia is nothing more than a physicial arrangement with certain properties also imply that creating conciousness can be done by creating such a structure?
Wouldn't that structure necessarily include animate physical matter in order to achieve consciousness as we know it? Or do you think all of our physical body information has no bearing on consciousness ? It would possibly be conscious in some sense of the word but it would have no relationship to ours.
I'm not sure. Why must it include animate physical matter?
I mean, yes I intuitively agree that a physicalist theory of consciousness should preclude things which are not physical matter being conscious. But if the theory, as written defines consciousness in a way that does not preclude that we have to bite the bullet or else reject the theory. You can't have a theory of "what consciousness is" that fails to have a 100% correlation between "things which satisfy the theory" and "things that are consciouss"
But Anyway computers are also made up of physical matter so even if the theory did require that it's not a problem.
> But Anyway computers are also made up of physical matter
Animate physical matter is my point. All of the information and demands of a living creature as a major input into the information processing of the brain.
Depending on how one defines consciousness, a computer certainly could be conscious. (an ability to introspect on its own process ) but it will bear no relation to human consciousness; it will be a horse of a different color. I don’t see any way that a computer can have a sense of living or dying ,for instance ; those cannot be meaningful concepts to it. I also don’t see how human consciousness can exist without the concept of living and dying as a significant part of it. It is one of the things that we introspect most about.
I don't define consciousness, I perceive it directly.
Sense of dying is not needed for consciousness.
Toddlers don't have a sense of dying, they are not lacking consciousness.
If consciousness is real than computers either have it or they don't. If they do then I don't know what it means to say that it's a "different flavor"
If consciousness is not real than neither computers nor humans have it.
If you don’t have a definition of consciousness then it’s difficult to discuss it. I perceive consciousness as well, for what it’s worth.
I offered up a sense of death in the origins of consciousness historically but…
As for toddlers, I think you should look in to the research on childhood development and theory of mind.
I have been mulling this over, and I take your point about "If consciousness is real then computers either have it or they don't" , however I wonder if this is only true if one subscribes to a physicalist view of consciousness. I have no problem with that. But then it becomes a question of what inputs shape that consciousness, because consciousness is not a static thing, it's a dynamic system, and it is very much involved with it's own state, and the state of the external world. Here, I think, one can draw a distinction. I don't think it is possible to "accurately emulate" (as a less loaded version of "endow") the input of being housed in a physical, animate, living body. This experience is unique to every human being, and yet there is broad commonality as well. I don't think a computer-housed consciousness will be at all the same, but I guess that is a distinction of degree, not of kind.
Seems unlikely modern AIs have the thing described in the post (or that they will, until the next major breakthrough).
Oscillations are unnatural for feed-forward networks.
See also: https://en.wikipedia.org/wiki/I_Am_a_Strange_Loop
Do neurons really distinguish spikes that are part of a particular rhythm? If this is a thing, why isn't it in the standard models of neural spikes? Also, don't we have evidence of brain harmonics, with eg 7 alpha oscillations for every gamma?
The bullets to bite looked good to me. I liked the overall framing, though I'm not sure the neurology holds up.
I don't have empitical evidence that individual neurons can treat signals received from an oscillation they're part of differently from other signals. I'm only saying that's required for my theory to hold water, and why I think itis plausible that they can.
Yes there are brain harmonics! I left that out, because the article was essay was already too long (I had to cut a lot for Scott to accept it) and it is not essential to my argument.
I don't know what this comment means, but it seems like you may have misinterpreted mine. The point is that deflationary views of consciousness float better in academic discourse about the realism of other minds than in a discussion about basic empathy.
Daniel, I have some purely technical observations that, in my mind, call the whole premise of the post into question:
[Daniel] "Since
billions of bits of neuronal activity, across many neurons in space and many milliseconds in time, can be part of a single pattern,
but patterns can also remain separate, even while running into each other,
there has to be a difference between these two states of affairs."
Why is there a "difference"? Oscillatory patterns interfere all the time, I'm not sure what particular insight is there to go after?
[Daniel] "A neuron that is part of one oscillation can hardly also be a part of another[...]"
Why? This is left there as if it were an obvious truth, but I don't think there's any reason to believe it to be true.
[Daniel] "Higher frequency oscillations are smaller, which makes sense because higher frequency means less time for the circular signals to travel, and smaller means less space through which they travel."
This statement may be true, but I don't see any references to the brain waves dispersion. It is not apriory obvious that high-frequency waves propagate faster that low-frequency ones. I sense a degree of confusion between wavelength and propagation speed here.
[Daniel] "There are also larger and therefore slower oscillations. Neuronal signal propagation has very variable speeds, but the lowest of low estimates still gives it half a meter per second, i.e. much less than a second to travel straight across the entire brain. Every thought that lasts longer than that, such as your understanding of this sentence, has to be at least a bit circular and therefore oscillatory."
"Larger therefore slower" doesn't make sense, there's a confusion between amplitude and... frequency? propagation speed? "has to be at least a bit circular and therefore oscillatory" is meaningless, oscillatory processes can be depicted in polar coordinates, but that doesn't make them "circular"; is a circular propagation in the brain implied here? As a surface wave? this is very confusing.
[Daniel] "Spikes that are part of the same oscillation fire in sync; they have a shared rhythm. For each neuron that is taking part in an oscillation, the time elapsed between its own sending of a spike via its axon, and the arrival at its dendrites of subsequent spikes from neurons it oscillates with, remains constant over multiple such intervals. That is what it means to share a constant rhythm."
This doesn't make any sense. Neurons are firing at variable GHz rates ("A living human brain will contain between 9 and 200 billion spikes per second"), the much-slower brain wavers must be oversampled, therefore assuming the neurons firing "in-cync" is not in any way substantiated.
I can go on, but the pattern emerges (bad pun!). These statements build a foundation for what follows, and I think the foundation is very flimsy.
High-frequency waves don't need to move faster to require smallness; they move in smaller patterns and move at the same speed, which I think is the only way to end up back where they started more frequently, as is (?) required for high-frequency.
I don't think they meant to talk about amplitude at all. "Larger" means physically larger.
Brain-waves are produced (almost exclusively?) by oscillations, according to the post.
Individual neurons rarely fire more often than 100Hz. (I'm not sure you were confused about that.)
Ok, I think I see where I'm confused: are these oscillations in time domain, or do they also have a spacial dimension, which would make them "waves"? This is super-unclear.
What I'm getting at is that an oscillatory process has to have a time dimension, i.e., a period (s) or frequency (1/s or Hz). But it doesn't necessarily have a spatial dimension; for example, pressure inside a Helmholtz resonator fluctuates in time, but does not propagate, the phase is the same through the volume. So it's not clear why pulsating neurons have to create a traveling wave. And of course I'm still not there with the idea that a neuron can only be in one wave at a time, superposition is a thing, and if it somehow doesn't work for neuron networks it needs to be addressed.
Yes, I clearly was confusing the rate of neuron firing with the rate of the whole brain firing. The weird implication of this is that there's at most 2 billion (American "billion", 2E9) neurons firing at a given time? This number seems low, but that's what we get dividing 200 GHz by 100 Hz?
Neurons fire (much?) slower than 100Hz most of the time.
The brain has some oscillations with (very) little or no spatial dimension. Those are irrelevant here, I think.
Oscillations interfere with each other. (Neurons have a very small amount of 'memory', so they can act different for one oscillation depending on what the other is doing. They also have other nonlinearities, like a refractory period.).
So it's hard for neurons to carry multiple at once.
Neurons are sometimes involved in multiple oscillations, when those oscillations' periods are (simple fractional?) multiples of each other.
The term 'oscillation' might be a bit misleading. A process could move around the brain without its path intersecting itself, but if it moved for long enough (a quarter second?) it would have to produce oscillatory patterns, because neuronal signal propagation is fast and the brain is small.
'Oscillations' describes the things we're talking about. In some cases, it might not be much important that they oscillate.
You should maybe reread Scott's post on brain-waves.
https://www.astralcodexten.com/p/book-review-rhythms-of-the-brain
Did! re-read the brain waves post. Left me actually more... confused, I guess? So definitely not spatial waves in any sense, but time-domain oscillations. It also looks like these Greek-letter waves are superimposed, i.e., delta is there always, and the higher-frequency waves ride on top; this is also what pink noise-shaped spectrum suggests.
I think I'm getting a bit ahead of my skis here; for now my takeaway is that these brain waves "explain" consciousness is too simplistic a take, which also may be because I'm not comprehending the original post....
Yay.
You're thinking at the level of waves in the physics sense, with amplitudes and phases and everything. I'm thinking at the level of neurons, which are too numerous to keep travk of so we use "wave" as a kind of metaphor. As I wrote, the oscillation is itself an abstraction.
I don't see any specific disagreement between us that I can't trace to that difference in perspective. Do you?
No, I realized my understanding was too simplistic during my back and forth with D. J. S. So that perspective difference is indeed it.
I am struck by how much of this discussion seems to pre-suppose that consciousness lies entirely in the brain. There are a lot of proposals of how consciousness might exist entirely independently of a physical flesh and blood body. Recursive computer programs sitting on top of one another and things like that. A pan of water being rained on with sand in the bottom and things like that. These might be good models for how everything gets processed by the brain, but "everything" includes the realm of the senses and the demands of a living, corporeal being which is a "program" that is constantly running in the background. I think one of the earliest glimmers of consciousness is in human beings’ ability to predict the future. To understand the seasons and prepare for them for instance. To finally know ahead of time where the sun will be at a particular time. Consciousness arises from the development of abstract thinking, that can be used to channel and often suppress the demands of the corporeal being.. The prefrontal cortex, executive function: that part of us.
I am not sure what people really mean when they refer to the hard problem of consciousness. Is it a precise understanding of the mechanisms that are involved in a sentient human being, or is it why? Why are we conscious? Or is the question why are we alive? I don't suppose it's that, unless one thinks that all living things are conscious in which case I don't know what to say. It seems pretty clear that some other higher order mammals on earth have certain traits that could lead one to believe that they are on the edge of consciousness. We know something about other forms of pre-humans that didn't make the cut, like the Denisovans and the Neanderthals and others.
I think that the heart of what is termed the hard problem of consciousness is the realization that we are each mortal and our time will end. This is a necessary condition of being able to see into the future based on what is happening now and observing it and it's rather an unfortunate one. because it provokes the abstract nature of higher thinking to come up with reasons why that might not be the case. Archaeological evidence indicates that a great deal of time and energy was put into this problem. The pyramids of Egypt being a good example. In other words, consciousness is a product of the friction between the unalterable state of being flesh and blood mortal creatures and being able to think about that. Elephants seem to have a glimmer of that notion, but I doubt they spend their time on earth thinking about it much until their body starts to fail them. No doubt the computer program or programs working together can mimic this friction, but they don't really have it. If something goes wrong with the way they are “thinking” they just shut down or they indicate to us that they cannot go any further with the process. Does anyone really think that they will have mortal feelings about this state of affairs no matter how complicated they get? I don't think that anything that is constructed of inanimate matter is going to have the hard problem of consciousness.
In the article, qualias have the condition of being ineffable. I don't know whether that's really part of the definition of qualia or not but if it is then it is by definition inexplicable, which does not make it a mystery exactly, just something that can't be explained in any words we have.
I do think that consciousness is entirely in the realm of chemistry and physics, but that doesn't get us any closer to the truth of it unless one can understand the “complete state of being” which includes very much the physical body that we inhabit. I think mind-body duality is a very poor way of approaching this because it isn't really a duality except in our experience of it, which is highly mutable to say the least.
Look at pain for instance, which I assume is a Qualia. People experience pain very differently, so it's actually a whole bunch of qualias.
Given that human beings have a very large capacity to choose how we think or feel about something, the hard problem of consciousness becomes a very individual one and not a general case. My Qualia around the “hard problem of consciousness” is bound to be quite different than anyone else's. In other words, there is nothing to solve here except one's individual process. A devout belief in God and an afterlife is one solution, but there are others. There is no universal solution to this problem, because it is entirely self-created. My opinion…
"The measuring seems trivial. We already know every neuron has a refractory period after it has fired. This constitutes a built-in cellular timer that measures the interval since its last spike"
This is gonna need more explanation. Didn't the refractory period only provides timing information for spikes received so fast that the neuron can't respond anyway?
No, the refractory period is after spikes sent, not after spikes received.
Maybe I just don't understand how a refractory period can be used as a timer. Could you elaborate on that?
Since the refractory period always have (nearly) identical length, the infomation how much of one has elapsed must be in there somewhere.
How do you pronounce 'quale'?
'Quale' is pronounced like "quail".
No, kwahley.
I'm confused by homogeneity. The qualia of sight and the qualia of touch don't seem to be the same type of thing to me, beyond just being things I experience. But "things I experience" is just what qualia are, so I'm not sure what it explains to go beyond that and say that they're homogenous.
I think that criterion is meant to mean that all qualia are "things I experience", as opposed to some of them.
Doesn't that just mean that all things I experience are things I experience?
Qualia can't be defined with "things I experience" for the purposes of the criteria, because "I" and "experience" are also hard to define or measure.
Let's take a more concrete example: vehicles.
Suppose I say: "Homogeneity: all vehicles are (felt) to be of the same type. While differences between them can be appreciated, they are always (experienced) as the same kind of thing."
(Verbs in parentheses because maybe they should be substituted for something more applicable, but hopefully that doesn't detract from the point too much)
Is this true, or false? Or does it not mean anything, by itself? I tend to think it's the third thing. "All vehicles are vehicles" is about all that it claims, but that's tautological. If we are to ask, are a bicycle, a truck, and a helicopter the same kind of thing, I think we can only answer that question with respect to some category. They're all vehicles, but they aren't all ground transportation. They're all vehicles, but they aren't all self-powered. What makes one of these categories more salient than another? Why couldn't we say that a car and a truck are of the same type, but they are not the same type as a helicopter?
I can't see a fundamental difference between this and doing the same thing for qualia. As far as I can tell, what makes the qualia of color perception and the qualia of taste the same kind of thing is that they are qualia. But I could as easily point to the category of color perception, and say that perception of red and perception of blue are the same kind of thing, whereas perception of the taste of cinnamon is not.
I don't think we need to be able to rigorously define qualia (or vehicle, for that matter) in order for this to be so.
The way it is expressed is quite confusing. Individual qualia like red and green aren't homogenous , and types of qualia such as colours and flavours aren't homogenous. Presumably , the idea is that all kinds of qualia themselves fall under a higher level kind distinguished by a higher level property.
>Homogeneity: all qualia are felt to be of the same type because that’s what they truly are; they’re all bits of information being processed internally.
The way this is expressed is quite confusing. Individual qualia like red and green aren't homogenous , and types of qualia such as colours and flavours aren't homogenous. Presumably , the idea is that all kinds of qualia themselves fall under a higher level kind distinguished by a higher level property.
>So qualia arise out of neuronal information processing much like biology arises out of chemistry
To say that qualia are information processing and nothing more doesn't identify the higher level property of qualia ness, since everything else is information processing.
>Qualia are nothing but information being processed internally, on their own information channel
That's slightly better, but it's nowhere near a predictive the ory.
Thanks for the "slightly better". The predictions are at the end of the essay. Unless you doubt that they fall right out of the theory, the theory is predictive.
Fascinating account of the possible underlying mechanics of conscious experience, but it does nothing to bridge the explanatory gap for a Chalmerian like myself. I have a feeling the hard-problem deniers (eliminativists / illusionists) won’t be satisfied by this essay either, since they see no need to delve into phenomenological themes to begin with.
This is all “easy problem” stuff, sadly; hard problem remains unresolved.
No, wrong. I have demonstrated that something with the properties of qualia has to arise from how thoughts have information processing that is internal to themselves, such as "working memory". This is a purely physicalist account of why SOMETHING WITH ALL THE PROPERTIES OF SUBJECTIVE EXPERIENCE would (necessarily) arise from the kind of information processing that we know the brain (sometimes, when conscious) does.
If that doesn't solve the "hard problem", what would?
Subjective experience is just what it feels like to be thought.
Regarding the Hard Problem: consciousness is either based on properties of the physical world that we don't yet know (if I understand correctly, it is to such theories that they author directs the comment "they don't pay rent") or not, in which case we are left to believe that somehow consciousness appears by whatever solution to the Easy Problem (such as the one proposed by the author).
In other words, consciousness is either something other than a computation, which is mind-blowing, or it is a computation after all, which is way more mind blowing.
I don't see how the ideas presented here (which I find extremely interesting, but I am far, far from an expert) offer some resolution to this paradox.
Not that the Easy Problem is easy. Obviously.
Consciousneas is a computation. But it is not a computation the brain makes (directly), it is a computation a thought makes. The brain is not a computer, but the thought is.
As a musician, this essay used oscillation in a very different way than I was expecting. One of the key properties of sound is that oscillations coexist. If you pluck a string, it produces a fundamental tone based on the length of the string, and then also lots of other tones based on low-denominator fractions of the length of the string. These overtones (the octave at 1/2, the fifth at 2/3, the fourth at 3/4, etc) are quieter, but they make up a very important part of the sound of any instrument (called timbre). Except when someone uses a synthesizer to make pure beep noises without overtones, those are timbre = 0.
So when I read that thoughts are cycles of oscillations of neuronal activity in the brain, naturally, I thought that the oscillations would combine with each other depending on their frequency ratios, and that's how thoughts would interact with each other. Instead, the essay seems to be saying that the different oscillations somehow block each other. Like if you could hit a bell, and then hit it again in such a way as to quiet the first tone. (I think that's technically possible? But not easy.)
Anyway, oscillations don't generally block each other. I guess it depends on the medium that's oscillating, but they generally pass right through each other, or strengthen specific parts of each other. So most of this is pretty straightforward and makes sense, but I'm skeptical about the oscillations blocking each other.
Also. In The Sickness Unto Death (1849), Kierkegaard wrote:
Man is spirit. But what is spirit? Spirit is the self. But what is the self? The self is a relation which relates itself to its own self, or it is that in the relation that the relation relates itself to its own self; the self is not the relation but that the relation relates itself to its own self.
So, basically the same as your recursive reflections.
Sorry, this is not music, it is not even physics. The "oscillations" and "waves" of electroencephalography research are basically metaphors. We see them on our measuring devices, which don't look at individual neurons but can only notice averages of activities of lots of them. We do know there are "spike trains" and we know some are circular and behave as I described, but the brain is not made of vibrating strings.
Cool, thanks. Makes sense why my musician's intuition would break down.
are you aware of the human brain project, especially part with neuromorphic hardware and spiking neural networks? www.humanbrainproject.eu
Some years ago I was involved with them and much of what you said fits very well with what I learned at the time. (I was just a programming, not a scientist, so my understing of the subject is very limited).
Among other things they made "spiking neural networks", which are a computer model of neurons, which can be processed somewhat efficiently, while also behaving more similar to biological neurons. They did computer simulations with these neurons with the goal of replicating mechanisms of the human brain. To that end they built hardware, that could efficiently simulate these neurons, so they could simulate millions of these spiking neurons in realtime.
Again, I was no scientist, so I can not point you to any relevant research. Also apparently the project ended last year, so the results should be available somewhere.
I'm not familiar with that yet; thanks for pointing them out.
Regarding "zero qualia = unconscious", isn't there a concept in Buddhism of "consciousness without an object"?
Yes there is. That's what you get when you fill attention with nothing but itself.
I think that's the equivalent of moving the facing mirrors so close to each other that they don't have anything beyond each other's bounds to reflect - you get a sense of virtual infinity.
It is quite pleasant.
It's too bad that the article (unintentionally, based on OP's comments in the comments section) claims to solve the Hard Problem, because it's a fun proposal on the Easy Problem and I think that some of the weedy details deserve more attention than they have gotten.
I should say from the jump that I am not a neuroscientist, but I've taught a bit of neuroscience at the upper-level-undergrad level, and I hang out with some neuroscientists, and so maybe I'm just in an uncanny valley situation where if I knew either more or less about neurons, this would make sense to me. That being said:
"The measuring seems trivial. We already know every neuron has a refractory period after it has fired. This constitutes a built-in cellular timer that measures the interval since its last spike."
I don't think this is useful for your theory. At least, the standard model of how neurons work is that there is a hard/"absolute" refractory period during which the ion channels reset, and no action potential can occur. Then, there is a soft/"relative" refractory period during which the neuron is rebuilding its electrochemical potential energy, and it is relatively difficult for an action potential to be triggered. This is on the order of milliseconds, and seemly not long enough to be producing the oscillations you want (and also the relative refractory period is probably not the sort of on-off type process that would be likely to produce sustained patterns of action potential firing). Once the refractory period is complete, the chemical mechanisms that produce it are no longer changing--the ion channels are reset, and stay that way, and the cell is at its "resting" potential, and stays that way. So they can't actually encode any time information after a few milliseconds.
So you will need to posit some other sort of mechanism that can encode temporal information on the time scale of ~ 1 second if you want to make this work. There is a brief discussion of pacemaker neurons in the comments, which is more promising, but I think it would be helpful if you did a bit of spadework to help us understand how much more promising. My (very weak, extremely correctable) impression from some quick googling is that we seem to think that pacemaker neurons work similarly to pacemaker cardiac cells, in that they are expressing different ion channels in their membranes in a way that makes their action potential dynamics pretty different from a typical standard interneuron. If this is, in fact, the case, then I'm not sure that we actually do know of any good candidates for molecular processes in interneurons that let them keep time at the precision you want. I'm not saying that such processes can't exist, though.
That leads me into my main point: your list of ways to test your hypothesis with emerging tech is very cool, but your hypothesis also seems eminently testable at the cellular level: we *should* be able to produce interneuron behaviors that match up with your predictions. I.e., it seems like low-speed temporal preferences are extremely important to your theory, if I am understanding it correctly: if you trigger an action potential at t=0.8, t=1.6, t=2.4, then trigger an additional action potential at t=3.2 should be easier than at t=3.1 or t=3.3. Not seeing this in neurons wouldn't necessarily be fatal (you could always argue that temporal preferences are a network phenomenon), but I think it would be Quite Bad.
This sort of thing doesn't require any advances in technology to examine; my understanding from talking to a colleague is that pretty standard neuroscience techniques would be workable here. In fact, these temporal preferences seem so findable in either cell cultures or tissue section sorts of experiments that--if they are as important as you posit--I would be pretty surprised if they hadn't been found already, even if their broader importance wasn't noted. Have any phenomena like this been noticed? I am totally ignorant of the literature, so some sort of review of that might have been useful as well.
I do (now) think it solves the hard problem, in that it demonstrates why (something with all the properties of) subjective experience has to necessarily arise from how we know we process information.
Your other criticisms are valid and welcome! But a flat out unargued assertion that this doesn't solve the hard problem is just insufficient.
No I don't think an ability of individual neurons to distinguish between in-oscillation and other signals has been demonstrated. Yet my theory requires it. That is a clear weakness of my theory, which is why I extensively pointed out this problem.
This is a fascinating and informative post, esp. since I happen to be writing a book on this topic (the intersection between psychological research and the philosophy of the self). I can't find anything to disagree with--it all seems well reasoned and empirically supported. It doesn't solve the hard problem, of course, in the sense that it isn't a complete explanation. It provides a helpful "dive into the nuts and bolts" of neural activity in the brain, but it doesn't provide a complete explanation because I don't think that qualia or the consciousness (not precisely the same thing) can be exhaustively explained by neural activity. The brain has gross (higher level) architecture, after all. The fact that conscious experience appears to be centered in the prefrontal cortex should be important. To move forward, I think the theoretical explanation needs to be scaled up, and include interactions between functional areas of the brain. To be specific, what is the prefrontal cortex doing that the rest of the brain isn't? Presumably, it is creating a simulation of activity occurring elsewhere in the brain, condensing it, organizing it, and disseminating it.
Why does this produce phenomenal awareness? What is the PFC doing that, for example, computers or simple organisms are not (assuming that computers and simple organisms are not experiencing things consciously).
Another wrinkle is to incorporate what we know about how the mind is organized. Qualia, even consciousness, isn't entirely what people mean by the "Self." Memories surely have to have something to do with it. One thing the human brain does, that computers and simple organisms do not (apparently) do is form and maintain complex conceptual memories of one's own experiences. The fact that a series of experiences have happened, that when they happened they seemed to have implications for what kind of person you are, and then you stored those experiences, seems like something important to include in one's theory of consciousness. In fact, I would discriminate between phenomenal experience (conscious experience in the moment), from a self concept that could be organized from a set of interconnected memories that seemed to help you learn things about yourself. Perhaps self awareness is conscious experience subliminally informed by a self concept, in the presence of stimuli that activate those memories?
I admit that a better understanding of how memories are encoded and retrieved at the neural level would help tie this all up in a neat little bow.
Yes it does solve the hard problem. If you'd like to argue about that, or there's any other way I can help with your book, I welcome your email or (better yet) X DM.
Not sure what X DM refers to (I'm old), but my email is wooddellv@yahoo.com
X is the social network formerly known as Twitter. I "am" (ha!) @7secularsermons there.
Fortunately, I do not have an X account.I'm afraid that you'll have to email me.
The problem with qualia researchers is that they start with the default assumption that all humans have qualia and are fully sentient
This assumption is based on humanist egotism and is very wrong
Until qualia researchers can put aside their incorrect baseline assumptions, their results will always be useless
https://questioner.substack.com/p/the-psychology-of-the-npc
I was unaware Dennett thought phenomenal consciousness is ineffable as you say:
"Ineffable: qualia cannot be communicated, or apprehended in any way other than direct experience."
In his conversation with Robert Wright (https://www.youtube.com/watch?v=RO7fro4frKk), he says that in principle, one can fully know and experience the subjective consciousness of another being if they have perfect knowledge of the state of that being's brain (kinda like Laplace's demon). Or perhaps I've wildly misinterpreted him.
Yes you're misinterpreting him. He wrote a lot on the subject, especially "Consciousness explained" (which arguably doesn't, but is still a great book).
Ok I'll read the book. But you should check out the interview he did with Robert Wright. It really does seem like he's claiming qualia are effable.
OK I will, thanks.
Fantastic article! It puts together a lot of things that the series of thoughts observing this comment being written have had vague ideas about for a while (e.g. that thoughts come from localized electrical activity, but not precisely how they avoided interfering with each other.) I suppose it should have been intuitive given that different phases of sleep have different frequencies, but then again, I'm no neurologist.
I do think that the proposed falsifiable experiments provide a good starting point, but I'm curious as to how much thought you've put into frequency comparison for individual neurons (and neuron chains). As I understand it, the brain is at a (fairly) constant temperature, and each neuron can only be surrounded by/contain so many Na/K ions, due to the space that they take up. Given that energy and reagent amounts are (roughly) the only two things you can vary to change a reaction's rate (I'm not sure if neurotransmitters act as catalysts?) it seems as if there are a limited number of frequencies available for all the thoughts in the brain to use. I'm wondering if that's a limiting factor on how many thoughts we can "hold" in our brains at one time, or if this is nonsense and there are far too many other variables around for it to matter.
finally got-around to reading this, thanks! Some good points about what thoughts are and are-not, but when the author turns to reflective thought then the old assumptions creeps in that there is a real, singular "you" that "can introspectively notice," and "deliberate, analyze," etc. when the idea is that you aren't doint this - thoughts are! Also, while quantitative measures attractively are comfortable, and will continue to provide more data and evidence, I'm always suspicious of interpretations derived from fMRI pictures. But, like I said, thanks again!
PS - Minsky, long ago, wrote in Society of Mind that thoughts themselves are ambiguous.
"Thoughts are the processes you can introspectively notice in yourself when you deliberate, analyze, evaluate, reason, form concepts and solve problems. You can also notice similar, but simpler and briefer processes when you imagine, remember, notice, recognize or judge something or when you feel an emotion or motivation. These simpler, briefer mental events will also be called “thoughts” here."
Yes. When I'm saying "you can introspectively notice" I'm talking to you the conscious thought, not you the organism that the thought identifies with.
"However, if you hook those EEG electrodes up to the amounts of computational power available these days, you can mathematically reconstruct quite good guesses about where in the brain the electrical signals are coming from. And that’s a game changer. This combined temporal-spatial resolution lets you localize individual neural oscillations, if they’re large enough. And that’s how you get to look at (oscillating) thoughts!"
Have you tried looking in the scientific literature if something like this has been done? I feel your proposal has some value, it would have more strength if the criteria to test it would be operationalized.
IMHO EEG is not enough, it's limited to the surface and too much low-res. However intracranial EEG is making huge leaps and has the resolution and depth (it's often used in epileptic patients) to capture something.
Yes intracranial EEG is very promising! But you can find papers with simultaneous LORETA and fMRI that demonstrate that LORETA does make oldschool EEG able to make valid observations of phenomena deep in the brain.
Valid observations may not be enough either. There are cheap EEGs around to test your hypothesis though, personally I have 2 Ganglion from OpenBCI (4ch each). If you figure out min. number of channels needed we could build a cheap setup and do some citizen science with the BCI community :)
LORETA needs a minimum of 19 channels.
To my mind there’s an important distinction between what we decide to teach to kids and what we consider worthy of investigating period.
How can this question of consciousness be considered a solved problem within science when we have so little evidence and understanding still?
I am a theoretical physicist. Among others, I am interested in the foundations of quantum mechanics. In the book "The Grand Biocentric Design: How Life Creates Reality" that I co-authored with Robert Lanza, we explain that quantum mechanics cannot be understood without bringing consciousness into the game. The concept of quantum state, described by a wave function, is closely related to consciousness. https://www.amazon.com/Grand-Biocentric-Design-Creates-Reality/dp/1950665402 . These ideas are summarized in the video Biocentrism: A physics perspective https://www.youtube.com/watch?v=3LgGv4jb3sU on the YouTube channel "Chasing Consciousness".