Wednesday, January 02, 2019

Electrons don’t think

Brainless particles leaving tracks
in a bubble chamber. [image source]
I recently discovered panpsychism. That’s the idea that all matter – animate or inanimate – is conscious, we just happen to be somewhat more conscious than carrots. Panpsychism is the modern elan vital.

When I say I “discovered” panpsychism, I mean I discovered there’s a bunch of philosophers who produce pamphlets about it. How do these philosophers address the conflict with evidence? Simple: They don’t.

Now, look, I know that physicists have a reputation of being narrow-minded. But the reason we have this reputation is that we tried the crazy shit long ago and just found it doesn’t work. You call it “narrow-minded,” we call it “science.” We have moved on. Can elementary particles be conscious? No, they can’t. It’s in conflict with evidence. Here’s why.

We know 25 elementary particles. These are collected in the standard model of particle physics. The predictions of the standard model agree with experiment to best precision.

The particles in the standard model are classified by their properties, which are collectively called “quantum numbers.” The electron, for example, has an electric charge of -1 and it can have a spin of +1/2 or -1/2. There are a few other quantum numbers with complicated names, such as the weak hypercharge, but really it’s not so important. Point is, there are handful of those quantum numbers and they uniquely identify an elementary particle.

If you calculate how many particles of a certain type are produced in a particle collision, the result depends on how many variants of the produced particle exist. In particular, it depends on the different values the quantum numbers can take. Since the particles have quantum properties, anything that can happen will happen. If a particle exists in many variants, you’ll produce them all – regardless of whether or not you can distinguish them. The result is that you see more of them than the standard model predicts.

Now, if you want a particle to be conscious, your minimum expectation should be that the particle can change. It’s hard to have an inner life with only one thought. But if electrons could have thoughts, we’d long have seen this in particle collisions because it would change the number of particles produced in collisions.

In other words, electrons aren’t conscious, and neither are any other particles. It’s incompatible with data.

As I explain in my book, there are ways to modify the standard model that do not run into conflict with experiment. One of them is to make new particles so massive that so far we have not managed to produce them in particle collisions, but this doesn’t help you here. Another way is to make them interact so weakly that we haven’t been able to detect them. This too doesn’t help here. The third way is to assume that the existing particles are composed of more fundamental constituents, that are, however, so strongly bound together that we have not yet been able to tear them apart.

With the third option it is indeed possible to add internal states to elementary particles. But if your goal is to give consciousness to those particles so that we can inherit it from them, strongly bound composites do not help you. They do not help you exactly because you have hidden this consciousness so that it needs a lot of energy to access. This then means, of course, that you cannot use it at lower energies, like the ones typical for soft and wet thinking apparatuses like human brains.

Summary: If a philosopher starts speaking about elementary particles, run.


  1. Sabine wrote: I was very proud of my great insight, until I noticed that John Horgan said basically the same thing 20 years earlier, in his case referring to philosophers.

    You addressed this comment to Steven; I'm not sure you actually meant it for me, but it doesn't matter.

    Yes, Horgan has been critical of philosophers. The funny thing is, I think Horgan is a philosopher. Nominally he's a science journalist. Philosophers ask questions and try to find answers. They examine scientific research, interpret it, make sense of it, consider possible ramifications, help the public make sense of it. That's what Horgan does.

    Horgan has said that philosophy is obsolete because science is better at finding truth, but that's only half the story. Scientists and philosophers both seek the truth. Scientists seek it directly by doing research and philosophers seek it indirectly by examining research.

    Horgan mischievously asks "What is the point of philosophy?" But that question is belied by the fact that many of Horgan's books, including The End of Science, are chockful of philosophy.

    Copernicus created a revolution in space (Earth is not the center), Nicolas Steno created a revolution in time (Earth is much older than humans and isn't static) and Darwin created a revolution in identity (humans are animals). In their quest for truth, philosophers accepted these scientific truths, put them in perspective, made sense of them and then helped others make sense of them.

    If someone feels threatened or worried about scientific truths, it's usually philosophical perspective that helps them. If science says we're apes on an insignificant speck of dust in the universe, philosophy says we can still find meaning and purpose, even beyond anything religion offered us. And let's give philosophers credit for creating the philosophy of science.

    Sabine wrote: As I mention in my book, in a discipline where people are getting paid for discussion, they have no incentive to actually solve problems. I was referring to physicists discussing black hole information loss.

    Interestingly enough, in Horgan's book you'll find this:

    "According to Darwinian theory, science stems not so much from our desire for truth per se, but from our compulsion to control our environment in order to increase the likelihood that our genes will propagate. When a given field of science begins to yield diminishing practical returns, scientists may have less incentive to pursue their research and society may be less inclined to pay for it."

    If we ever get to that point, it would be ironic if philosophers end up being the ones who push scientists to do research purely for the sake of finding truth. Finding truth and solving problems is sufficient incentive. Why should scientists climb mountains? A philosopher might say "because it's there."

    In Sabine's book, she examines obstacles to scientific progress, including physicists getting paid to discuss useless theories, while Horgan argues that the great success of scientific progress portends the end of the era of scientific discovery. Sabine and Horgan worry how scientists might react if they and/or society lose faith in progress.

  2. Tim wrote: You owe an apology to everyone who has ignored their financial prospects to go into philosophy

    I like philosophy and I'm accustomed to seeing it get kicked around. I'll defend it, but I don't think the financial sacrifice approach is persuasive.

    I think it's fair to ask a professional philosopher how their work is useful. I rather enjoy telling people how my work is useful. Likewise, why shouldn't a scientist enjoy explaining how their work is useful?

    Sometimes it seems like scientists and philosophers are like farmers and ranchers. From Oklahoma:

    The farmer and the cowman should be friends.
    Oh, the farmer and the cowman should be friends.
    One man likes to push a plough, the other likes to chase a cow,
    But that's no reason why they can't be friends.

  3. @Unknown says Is it the same type of energy for each particle? Why should one type lead to matter and the other type lead to insubstantialness?

    There are not really different "types" of energy, most types you hear about describe something about processes. We have "potential" energy (gravitational energy, stored somehow) and "kinetic" energy. or we use adjectives to describe where the energy originated; solar, nuclear, electromagnetic, electrical, chemical, thermal, gravitational or mechanical energy.

    But in the end they are all the same thing, as Einstein's equation E=M(C^2) tells you.

    Why is it manifest in different forms? The unsatisfying answer is that this is how energy behaves. Searching for more satisfying answers to this question is Sabine's field of fundamental physics.

  4. Unknown wrote: notice I do not use the word "superdeterminism!"

    Well, if someone can explain a useful distinction between the two terms, I'll be happy to accept it. I'm accustomed to thinking of "randomness" in terms of determinism. You think that randomness and determinism are contradictory, while I think of it as a matter of perspective.

    I have a similar issue with the terms agnostic and atheist. An agnostic doesn't "know" if gods exist and an atheist doesn't believe in gods. But if you don't know if gods exist, obviously you can't believe in them. And the very reason atheists don't believe in gods is because they don't know if they exist. So really, what's the difference?

    Unknown wrote: the "universe" in its famous first three minutes "decided" that in our own time railroad lines and railroad stations would always be adjacent to each other?

    I'm really trying to understand why this question vexes you. What you're calling "decisions" are nothing more than events unfolding over time. In a deterministic model, railroad lines and stations don't just appear in nature. Countless events have to unfold first, including events leading to the creation of conscious entities that design and build railroads.

    Again, if you consider your own birth and all the micro-events that had to unfold over the past 5,000 years, you can see that your birth wasn't "decided" 5,000 years ago any more than it was decided a few minutes before you were conceived. In a deterministic model, events follow events.

    If you roll dice, the universe doesn't "decide" what the result will be.

    The deterministic model predicts that if you could go back in time 5,000 years to the same initial conditions, all events, including your birth, would unfold in the same way. There would be no spontaneity or randomness in the way you mean it.

    If you ask if I subscribe to a deterministic model, I would say I don't know enough. On the question of free will, I'm happy with or without it. If we don't have free will, the illusion is sufficiently good to fool us. When I wake up after having a pleasant dream, I still enjoyed it even though I know it wasn't real. Heck, I bet plenty of people would be willing to give up real life if they could exist in a fake but pleasant dream. If virtual reality gets good enough, maybe that's where we're headed anyway.

  5. Steve Mason said

    “If someone feels threatened or worried about scientific truths, it's usually philosophical perspective that helps them. If science says we're apes on an insignificant speck of dust in the universe, philosophy says we can still find meaning and purpose, even beyond anything religion offered us. And let's give philosophers credit for creating the philosophy of science. “

    Correct me if I’m wrong - science is based upon observation and the models constructed from those observations to reflect what science believes is the best possible explanation for something we call reality .
    Here is another model based upon observation , it’s called human civilization and all its attendant structures built upon the model of human experience gained from the collective observations from the past - present - and future . This model tells us that man has “ free will “ - we make choices based upon chance and probability . This model is unable to be falsified by science no matter what theory - equation - axiom or whatever “ technique scientists may use to “prove “ their scientific material determinism or whatever phrase maybe used to describe this belief that all that exists is matter . Science cannot falsify the collective human experience from the beginning of time that says every single human believes they are making free will decisions . This is reality from the human experience and it is a model that will stand until the end of humanity . No scientific “ theory “ is going to change this . If you reply to this post to tell me that I am wrong you will do it believing you are making a free will decision to say that there is no free will .You. along with every single human being that- has -is - will live , do so believing that you are making choices even when you say you are not .
    Science can NEVER prove it’s assertion that free will does not exist .
    The human model built upon observation of reality is that free will exists and it is impossible to prove otherwise .
    This simple truth is not suited to academia as it would take no more than half a page of a book to explain it and 15 mins in a classroom .
    Love is its ultimate truth .

  6. Wow, functionalism and behaviorism wrapped in a nice bow (contradictions and unanswered questions carried over with nothing new) and they (panphycist) still think it's a viable option to "explain" conciousness. ONE can quote at lengths all the phililosophers of antiquity (including 20th century) who offered explanations of consciousness, but it doesnt change the fact that they were WRONG.

    From reading these comments one thing is certain: when you're emotionally invested in an idea EXPERIMENTS (physics, biology etc) AND DATA (neurophysiology) BE DAMMED - LET THE NONSENSICAL ARGUMENTS FROM REASON and "LOGIC" RUN WILD!

  7. @Steven Mason; I believe in common usage agnostics are not certain either way. I am an atheist that is certain, due to logical contradictions I don't believe can be resolved; EVER. But I can understand agnostics that are self-aware enough to realize they are not expert enough at interpreting the claims for and against well enough to reduce the margin of error to personal certainty.

    Like my father that stopped school at 15 (he got my mother pregnant and they got married). He was naturally smart enough to question dogma and reject arguments from authority, and also smart enough to know he was capable of making logical mistakes and being tricked by fallacies.

    Agnosticism seems like good science to me; "I don't know" is a fine answer if you don't think there is enough information and you don't trust what is out there.

  8. @Unknown saying I am the villain who wondered how random events could occur in a deterministic universe. Determinism is part of several schools of physics -- that all things can be traced "backward" in time through a series of causal relationships.

    There is randomness in our universe; specifically wavefunction collapse, or radioactive decay. So NOT all events have a cause; some happen uncaused (to our knowledge). Thus all events might not be able to be traced backward via a causal chain. Uncaused radiation can cause a genetic change in an organism that leads to a heritable mutation. Neutrinos have a high probability of passing through matter, but can interact with it once in awhile, and I don't think why or how it interacts is proven to be deterministic and whether it does interact may be due to quantum randomness.

    So the answer to your question is NO, the current universe (to our knowledge) is not deterministically programmed from the moment of the Big Bang (which is probably not how the universe originated anyway); the evolution of the universe has been influenced by randomness that nobody can demonstrate is deterministic.

    This does not help the issue of free will and making decisions; those may be deterministic up to quantum randomness just like the rest of physics is deterministic up to quantum randomness. So making a truly random decision (e.g. basing your decision upon whether an alpha particle is emitted in an odd second or an even second) is no more free will than making a deterministic decision.

  9. Steven Mason said:

    "An agnostic doesn't "know" if gods exist and an atheist doesn't believe in gods. But if you don't know if gods exist, obviously you can't believe in them. And the very reason atheists don't believe in gods is because they don't know if they exist. So really, what's the difference? "

    Without referring to "official" definitions of the two terms, I'll offer my take on the difference between them:

    Agnostics are open to the possibility that God exists, atheists are not open to the possibility that God exists, thus removing the apparent commonality of a "non-belief".

    Define by what they believe (or do) rather than what they don't believe (or don't do); agnostics believe it is possible that God exists, atheists believe that it is not possible.

    I wouldn't say: "What's the difference between an engineer and a machinist, since neither one cleans homes for a living?"

    OK, off to check the OED now (better late than never : )

  10. To: Mr. Mason

    1. If random and predetermined events are not contradictory then biological evolution goes out the window. Either genetic mutations are random or they are destined to be. In the latter case, we are dealing with a teleological universe.

    2. Both you and our esteemed Dr. H avoid the point. In a deterministic cosmos, the universe was busy as a bee during its first three minutes mandating that railroad stations always be adjacent to railroad tracks, that toothpaste always occur in tubes and that the traffic at my corner always stops at a red light. Is THIS what you believe was on the metaphorical "mind" of the universe before our own galaxy even existed?

    3. If the universe is NOT deterministic then free will is entirely possible.

    4. If free will IS possible then the universe is not mechanistic and individual agency exists.

    5. If individual agency exists then any purely physicalist explanation of existence is incomplete.

    Don't dodge the question. Do you really believe (Dr H as well) that 14 billion years ago the infant universe was preoccupied with making sure that those streets on a planet that circles a not-yet born star in a non-yet created galaxy on which traffic is only allowed in one direction would ALSO be the very streets with ONE-WAY signs? In a deterministic model, that must be so. So you think it is? If so, the universe had entirely too much time on its hands during those First Three Minutes.

    Dr. H and others scorn philosophy. But, without it, science is just data.

  11. Tim wrote: if you think that regular methods of scientific procedure are adequate to address it, then you still don't understand what it is.

    I'm going to ask for your bottom line. You think there is a hard problem of consciousness? You think consciousness (with accompanying experiences) is beyond scientific materialism? And if someone doesn't share your certainty, you conclude, a priori, that they don't understand what consciousness is?

    You told Sabine that she "writes as if there is no problem at all." Clearly that's not the case. She isn't certain, as you seem to be, that consciousness is beyond scientific materialism. You might be able to take some statements out of context to make it look like she's saying there's no problem, for what it's worth.

    You brought up the lobster as if that supports the assertion that "regular science" can't possibly explain consciousness. You could bring in a thousand other animals for all the difference it makes.

    If you feel certain that regular science can't ever possibly explain consciousness, fine. But you have no evidence - let alone proof - that you are right. As for me, I'm happy admitting that I don't know. I'm open to the possibility that the human brain is not capable of explaining consciousness, even if - this is important - consciousness is purely materialistic. It's oddly presumptuous to think that humans are capable of understanding everything about the material aspects of the universe. We're apes. We see the cognitive limitations of our ape cousins but we're not so comfortable with our own.

    I'm trying to recall all the sci-fi stories that involve aliens with cognitive abilities far beyond ours. Does it ever end happily? If there are any hardcore sci-fi fans here, please help me out.

    And finally Tim, I'm mildly curious what you mean by "regular" scientific methods. This suggests the tantalizing possibility of irregular scientific methods. Please clarify.

  12. Unknown,

    I do not "scorn" philosophy, I dislike pseudoscience that masquerades as philosophy.

  13. Steve Mason,

    Yes, I think there is a Hard Problem. If you haven't gleaned that from what I have written then you simply are not not paying any attention. Yes, accounting for consciousness is beyond the scope of "scientific materialism" as that is presently understood, and beyond all variations on it that I have been able to conceive of. The pantheist wants to change the notion of "materialism" as it is presently understood by attributing properties to all matter that are not part of the present picture, but I can't even see how taking that radical step helps.

    No, if someone does not share my views I do not conclude a priori, or even a posteriori, that he or she does not understand what consciousness is. Indeed, I assume that everyone who has ever felt a pain knows perfectly well what consciousness is. It is not that a person who thinks there is no Hard Problem does not understand consciousness, but it has been my uniform experience that such a person does not understand the arguments that there is a Hard Problem, because he or she is unable to adequately answer those arguments. Pious hand-waving about how "the scientific method" will eventually work it out is not an answer to the detailed arguments being given. There may be some lacuna or false premise in the arguments, but Sabine sure has not pointed one out. Until someone does, I will continue to believe there is a problem.

    I brought up the lobster to make a particular point, namely that all of the study of my own brain states in the world (where I know from first-person experience which are accompanied by pain and which are not) will not in any obvious way help answer the question about whether the lobster is in pain. The lobster's nervous system will exhibit some "similar patterns" to mine and some "dissimilar patterns" to mine. So the advice to just "look for the patterns" is in principle worthless without some way to identify what it is about a "pattern" that makes it give rise to (or be) a pain or painful. Yes, I could have used lots of other animals as examples. I also used silicon-based computers as examples. So what? Each example makes the point.

    The idea that our cognitive limitations make it impossible for us to solve the problem has been extensively defended by Colin McGinn, in case you are unaware. The arguments I have been giving imply that the problem is even deeper than that. It is not that we don't yet happen to have a solution, it is that we cannot even imagine what it would be to have a solution and to be able to recognize by any means that it is the solution.

    I said "regular" scientific methods just to acknowledge that I have been considering present scientific methods as they exist. It is not that I can even imagine a change to the methods that would help: that is the task for the person who claims to be able to solve the problem. But according to Sabine the same methods as those we presently use are up to the task. That claim indicates that she has not understood the nature of the problem to be solved.

  14. To Tim Maudlin : I agree that there is a hard problem of consciousness, and I believe that the reason the hard problem is hard is that it has no solution within a materialist framework.
    If materialism is ruled, out the main remaining options seem to be panpsychism and dualism.
    You seem to reject panpsychism (is this what you called "panthesism" in your recent comment dated 1:18 AM, January 13?).
    So would you advocate dualism instead?

  15. Pascal wrote: " I agree that there is a hard problem of consciousness, and I believe that the reason the hard problem is hard is that it has no solution within a materialist framework.
    If materialism is ruled, out the main remaining options seem to be panpsychism and dualism."

    Haven't you forgotten Idealism. This is the idea that ultimately reality is made only of consciousness. This produces a topsy-turvey view of reality, but there is no equivalent of the Hard Problem. There is also no equivalent of the big problem with dualism - that there has to be some form of interaction between the two realms - so they aren't actually separate.

  16. Pascal,

    I am not advocating any position, because every position seems to lead to intractable problems. My main aim is to make the nature of those problems clear.

    There is clearly a very strong correlation or, supervenience relation, between physical brain states and conscious states, which argues against Cartesian substantial dualism. I suppose a "dual aspect" picture seems the most acceptable, but just accepting that does not really get at the heart of the problem. It is, for example, hard to understand how to reconcile any form of epiphenomenalism (such as a dual-aspect view) with the mechanics of evolution, since evolution only depends on external physical behavior of the sort that pure materialism can handle.

  17. “Well, if someone can explain a useful distinction between the two terms …”

    Just to get the elephant out of the room, maybe I may just illustrate what the super in superdeterminism (SD) is and nothing more. I hope Sabine or Tim will correct me, if something is wrong.

    When I first heard SD I also thought: What the heck is this? If it is already deterministic, how can it be more than that?

    Let us start with statistical independence (SI).
    A prime example for SI is: throwing two dice. If the two dice would be correlated, we would be utterly astonished and seek for an explanation like assuming “they are rigged”. But not rigged dice are not correlated – this is SI.

    The standard explanation for SI are non-linearity, chaos, Stosszahlansatz and H-theorem - purely classical terms in a classical deterministic universe.
    Like a dice a pseudo random number generator (pRNG) using e.g. digits of pi is also a deterministic object. Two pRNGs are not correlated, if not seeded with the same number (kind of initial condition). Also, two sequential numbers of one pRNG or two pRNGs are not correlated.

    Bell in his inequality assumes SI for the free variables, e.g. the polarizer orientations in an EPR experiment.
    In a deterministic universe Alice’s “decision” (A) of course depends deterministically on initial conditions, same for Bob (B). We also could substitute A and B by pRNGs seeded differently.

    SD denies SI between the state of the particle pair in EPR and A respectively B, i.e. the state is not independent of the detector settings.
    This would be analog to dice A respectively B being correlated with another dice.

  18. R. Taylor wrote on 12 January 2019:

    " ... WELL, I have often wondered: What if, when we set off fireworks, the little sparks, which are at first are white-hot then rapidly cool into little "globes", go through a rather short-lived "Goldilocks zone" during which civilizations rise and fall on their surfaces before they even hit the ground. Sounds crazy, but..."

    I had a similar fanciful idea but about particles rather than bonfire sparks. My idea requires particles to be like universes. This is clearly a non-Standard Model as it requires a particle to be divisible. The universe maybe starts and finishes at a point (see CCC model) just like a particle. Maybe viewed from outside it would appear to be quantised. And by that I mean not divisible without first bringing it to a point (by wave function collapse or whatever). So in a way quantised may mean only temporarily indivisible. But the universe would be permanently quantised into one quamtum until the end of (our) time. But internally divisible within the quantum. Any maybe more than one quantum is involved if those quanta cannot be separated until being brought to a single point.

    So in a non-Standard Model maybe there is structure enough for a particle to have intelligence within it. But a question is 'does the universe itself think?". And the answer must again be 'no' ... that is if the universe has too few quanta in its structure, when viewed from outside [and if it were a particle it should be measurable from its exterior, but only at the end of the universe]. And that is despite the many quanta internal to the universe. And that is also despite the universe having sentience within it.

    As for civilisations rising and falling (within particles) in the twinkling of an eye. Well, yes a twinkle in our external eye but I assume the internal civilisations think that a long time has passed for them. There is maybe also a case for removing time from its specially privileged unitary position. String theory models AFAIK normally only have one time dimension. But should that assumption always be the case? Now I really don't mean the following too seriously, but is there a case that CERN is interfering with the due process of civilisations within particles? My answer is again 'no' because of the quantised nature of particles. Nothing we can do affects what goes on within them. And they are oblivious of our external framework.

    One problem is that if two particles are oblivious (to their internal occupants) of each other, then how do they interact? They only do so when they are points and therefore no longer have internal structures.

    But what about off-shell masses of particles? Presumably there are a small number of quanta tightly bound in a normal particle which are temporarily less tightly bound in an off-shell particle. So again the internal occupants will not have structures at this stage. And an off-shell particle should not be able to think and probably has no sentience within it, either.

    BTW I am not anti the Standard Model. It is fine and possibly complete or nearly so. When the table of chemical elements was completed, the answer was not to find ever more elements (though that went on too) but to look for structures within. Now that the Standard Model is nearly complete, the answer is to ... [well when very high energies are obtainable, if ever, ...].

  19. Reimond,

    Your explanation is close to the mark, but understates the problem.

    Rigged (i.e. weighted or shaved) dice are not fair but they are also uncorrelated. Over the long term, seeing how one die comes out give no predictive advantage for the other die. That is, over the long term the frequency with which the dice show a certain outcome for both (e.g. die A '4' and die B '3') is just the product of the frequency with which die A shows '4' and die B shows '3'. To break statistical independence the dice would not merely have to be unfair (easy) but would have to interact or communicate during the throw (hard). That is why postulating a breakdown of statistical independence *for all possible ways of determining the setting of the apparatus* is, to use a technical term, insane.

  20. Sabine, Tim,

    Unknown asks some good questions about plain vanilla determinism.
    I also cannot imagine how “… railroad lines and railroad stations would always be adjacent to each other …” in an ‘exclusively deterministic universe, where all is set by the initial condition’ (EduwaisbtiC)

    Where in the initial condition would this information about where to place the railroad tracks be engraved? And CMB looks a bit too random.
    And it must be a lot of information also including this tiny bit of me just writing this sentence right now.

    And would Bohm really assume to put the tracks already in the initial pilot wave and particle positions?

    Sure, gravity is anti-entropic until a black hole forms and it finally is the driving force as Sabine said here .
    But the question is how all the complexity is generated.

    It would need much less information engraved in the initial condition, if someone could decide to shout “Hey, Joe, put the tracks over here close to the station.” But in an EduwaisbtiC there simply is no decision to make, which is of course absolutely consistent with no “free will”.

    I disagree with Unknown on this point “… If individual agency exists then any purely physicalist explanation of existence is incomplete …”
    With a tiny bit of randomness one can achieve the feast of having individual agency based on a purely physicalist explanation.
    This probably connects to this mind-body problem: how can the mind influence the body?
    With a bit of randomness complex systems are not exclusively causally/deterministically coupled and I guess this also would make “free will” possible in the sense of Sabines requirement here:
    “1’) An agent in possession of free will is able to perform an action that does not inevitably follow from all in principle available information at any time preceding the action.”
    “2) The actions of an agent in possession of free will cannot be consequences of fundamentally random processes.”

    The second point is just there to exclude, that already “… radioactive material has a free will”.

    Of course, in an exclusively deterministic universe “free will” is not possible.
    Here might the philosophical viewpoint be different. I am just referring to the physical one, but with a bit of indeterminism included.

  21. David Bailey there is an "interaction problem" either way. Idealism is nothing more than a subjective panpsychism, but apart from that how does physics become subjective?

    See my

  22. Correction: anti-entropic is of course wrong, I should have said tends to clump just as Penrose describes here

  23. Wow.

    What a journey this post by Dr. Bee has precipitated for myself. I keep landing on the musings of others which have a profound effect on my own thoughts, whether they enhance, destroy, or redirect. For example, a few months ago, I came up with this little illustration of how some scientific methods might appear to the uninitiated:

    Scientist #1: The numbers are just not working out for this here "Big Bang Theory".

    Scientist #2: Well, fudge the numbers. Change a constant. Something.

    Scientist #1: And how will I explain THAT?

    Scientist #2: Well, fudge is dark... I know! Call it "Dark Matter"!

    Today, I happened upon the video linked below. The link will bring you into the timeline near the sequence that relates to the "two scientists" exchange, but note that the nearly the entire video is relatable to the discussions surrounding Electrons can't think, so I would recommend viewing it in its entirety.

    Based on synopsis alone, Lost In Math has gone to the top of my reading list due to its apparent treatment of issues that I have wondered about myself... but now it appears I must read The Science Delusion by Sheldrake immediately following, for the same reason. One analogy I've come up with in this area:

    Suppose you discovered an automobile on the Earth. You don't know everything about its capabilities, but you realize that it has a mileage counter (sorry, no km for me! : ) which registers 240,000 miles. Calculating backwards, you conclude that it must have circled the Earth 10 times. Next, you perform an experiment that "confirms" your conclusion (fuel up the car and let her rip!).

    But you would be incorrect if the car had racked up those miles without having ever left the county. You just weren't aware that it was steerable. You would then be totally in the dark as to what actually transpired in the past of that automobile.

    Thus, formula + theory + experimental confirmation = ???

    Maybe I'll attack the concept of constants in a later post; after all, the entirety of Human experience amounts to perhaps a 1/1,000,000 shutter-speed snapshot of the cosmological timeline...

  24. Part One:

    Tim wrote: I think there is a Hard Problem. If you haven't gleaned that from what I have written then you simply are not not paying any attention.

    When people argue for weeks, writing thousands of words and digressing all over the place, I find it useful to validate the core issues in plain language and see where people actually stand. As you know, sometimes people argue a side just to demonstrate there are other sides and pros and cons on multiple sides.

    My two cent opinion is that you can argue for years that there is a hard problem, but you can't claim that your position is scientifically or logically authenticated and settled.

    Tim wrote: I said "regular" scientific methods just to acknowledge that I have been considering present scientific methods as they exist.

    That's what I would have guessed, but I wanted you to confirm it. But as you know, it's not just "methods." Sure, we might discover new methods, but there's also new discoveries, insights, knowledge, theories, tools, etc. If by "methods" you mean all of these things, great, we're on the same page. But if you're hinting at alternatives to scientific materialism, you'll need to clarify.

    My position on the hard problem is "We don't know." I've been paying attention to all sides of this interesting question for years. You haven't said anything I haven't seen before. What catches my eye is that you seem to be so certain that your position is correct, and anyone who disagrees with you "doesn't understand," a priori.

    I think it's great that people have robust discussions about unsettled questions. But I've noticed that sometimes people get attached to certain positions to the point where they can't admit they don't really know if they are right.

    Tim wrote: if someone does not share my views I do not conclude a priori, or even a posteriori, that he or she does not understand what consciousness is.

    There have been multiple instances where you claim that someone who disagrees with you doesn't understand consciousness. This is a common debate strategy that sometimes hints at ad hominem. It's a minor point, more observation than complaint on my part.

  25. Part Two:

    Tim wrote: all of the study of my own brain states will not in any obvious way help answer the question about whether the lobster is in pain

    If it's not "obvious" to anyone right now, so what? How is that relevant? How can you even say that the lack of "obviousness" is evidence for a hard problem? I acknowledge the possibility that there might be a hard problem, but you seem to want a hard problem.

    I've met plenty of people who want a hard problem because they think it's a way for humans to be special in one way or another. Some people say the hard problem is evidence for God or panpsychism. So naturally, when people insist that there is a hard problem, I wonder why. I'm not suggesting that you have ulterior motives, but I could ask if it would bother you in any way if consciousness could be explained with scientific materialism.

    Tim wrote: The idea that our cognitive limitations make it impossible for us to solve the problem has been extensively defended by Colin McGinn, in case you are unaware.

    McGinn's not the only one. I wish I could claim that my views on human limitations were original; then I could be special. :-) I assume you know that McGinn admits his proposition isn't certain. In any case, the position "we might never figure it out" is not evidence for "we can't possibly ever figure it out," which is the position you seem to be taking.

    I often quip that if humans were smart enough to understand the universe, it couldn't exist. That should tell you something.

    Tim wrote: It is not that we don't yet happen to have a solution, it is that we cannot even imagine what it would be to have a solution and to be able to recognize by any means that it is the solution.

    Yes, I get it. When push comes to shove no one really knows if there is a hard problem.

    Sabine wrote: according to Sabine the same methods as those we presently use are up to the task.

    I'm not sure about that. Have you asked her that specific question? Maybe I missed her answer. And of course a lot depends on what you mean by methods and what Sabine means by methods.

    In 1900, scientists were using the same scientific "methods" that we're using today, in the most general sense of the word. Scientists in 1900 were not up to the task of doing many things scientists do today, even though they used the same methods. So we need to be clear about what we mean.

  26. Just to clarify what I mean by methods. I mean that we do experiments, record data, and develop models, until we have found a useful way to predict experience from measurement.

    I do not think that the current technologies or the current models are remotely sufficient to understand consciousness. Indeed I think we are missing a big part of the necessary math (and some days I am far from optimistic we will ever figure out how it works). Also of the technology, of course, may be hard to come by, but as a theorist I know little about that.

    I have no reason to think there is anything about consciousness that defies these methods. And the only reason that other people have given me for why it might be otherwise is the insistence that certainly there must be something mysterious about it. I do not think this is a convincing argument.

    I also frankly think it's irrelevant that some philosophers prefer mysteries, because science will proceed anyway. I'm here to simply tell you that the idea that consciousness is fundamental will not survive much longer. (And also, electrons don't think.)

  27. @Sabine @Tim

    A wonderfully succinct statement from Sabine. Tim's and Sabine's positions are getting closer. Tim said that present scientific methods are not up to the task and Sabine agrees.

    If Tim acknowledges that future scientific methods might be up to the task, I'm inclined to say that Tim and Sabine agree on the most substantive part of this discussion - assuming they agree on what is meant by methods.

    Sabine wrote: I'm here to simply tell you that the idea that consciousness is fundamental will not survive much longer.

    Tim and Sabine both reject panpsychism - agreement.

    Sabine wrote: I also frankly think it's irrelevant that some philosophers prefer mysteries, because science will proceed anyway.

    I'd bet that Tim and Sabine would agree that science will proceed anyway. As for philosophers preferring mysteries to scientific explanations, I'll say again that I'm used to philosophers getting kicked around. Sabine told us that she's grumpy and anti-social, so to maintain her credibility she's got to act like it once in a while. :-)

    I'm going to drag out my good friend Michael Behe to demonstrate that scientists, not just philosophers, can prefer mysteries. Behe creates the pseudoscientific mystery of irreducible complexity in order to fabricate evidence for intelligent design.

  28. Steven,

    I did not say it requires a new scientific method or something like that. To make a long story short, I said we need better math.

  29. Reimond wrote: Just to get the elephant out of the room, maybe I may just illustrate what the super in superdeterminism

    It's my fault for not being clear on what I mean by a useful distinction. I'm casually acquainted with the concept of superdeterminism, but I don't see how it's useful (or relevant). In the context of scientific observations and predictions, how is it useful?

    I'll put it another way. Let's say hypothetically that superdeterminism is true but we incorrectly assume that determinism is true. Would it make any difference? Would our deterministic models not fit superdeterministic observations?

    My knowledge of superdeterminism is not comprehensive. I know that relatively few physicists subscribe to it and some physicists disagree on interpretations of it. If some physicist can demonstrate that superdeterminism is true, I'll stop asking people to explain how it's useful (maybe; it depends).

    Reimond wrote: SD denies SI

    I'm not sure that's necessarily true. In the back of my mind I have a vague recollection of an argument against that interpretation. Since I'm going to bed I'm not going to try and dig it up, and besides, my vague recollection might be bogus.

  30. Sabine wrote: I did not say it requires a new scientific method or something like that.

    "New" was a bad word for "better" or "more advanced." Good catch.

  31. Dr Bee,

    Some time ago you made it clear to me that atoms don't think alhough they behave in an individual way as is demonstrated by the fact that it is impossible to predict when an excited atom will spontaneously emit a photon. It acts as if it is aware of its abnormal high energy and decides to get rid of it at its own pace but that is only our narrative, just as it is our own narrative that other people are conscious and thinking beings. In the end it is all physics, panphysics.

  32. Steven,

    I appreciate your patience, but I do not mean a better or more advanced scientific method either. The current scientific method is just fine. What we are missing are sufficiently powerful mathematical tools. We simply do not currently understand how to quantitatively treat chaos and complexity, the latter isn't even well defined (or the well-defined notions aren't all that useful, whichever way you look at it). Maybe you could say that it will require a new scientific discipline, yes. But it won't require to change scientific methodology.

  33. @Steven Mason: I don't know what people mean by "superdeterminism", but I can see a useful place for it alongside "determinism".

    For "determinism" I believe we live in a deterministic universe that contains some uncaused truly random events; like radioactive decay and the eigenstate chosen by wavefunction collapse. That means we cannot trace from the present, either back in time or forward in time, indefinitely, because some causal chains will dead-end on an event that was not caused by anything else. For examples; say an alpha particle emitted by random radioactive decay altered a gene in a cell and produced a cancer. Or an alpha particle was energetic enough to cause a neuron to spontaneously fire. (I'm not claiming randomness is a route to free will or consciousness, I don't believe that.)

    So "superdeterminism" could be used to mean these uncaused events are not truly random, the randomness of QM is just a stand-in for the operation of currently unknown mechanisms or relationships, and there are no uncaused random events.

    Then they would be distinct words classifying two distinct belief systems; I am deterministic but I am not superdeterministic.

  34. @Reimond says Where in the initial condition would this information about where to place the railroad tracks be engraved?

    First, truly random quantum events happen (every wavefunction collapse must choose a single state from a probability distribution) so not all information must be encoded from the start.

    But even in an exclusively deterministic state, the information for the future is encoded in the near-infinite possible combinatorial combinations of particle relationships.

    I am a CS guy, I've written dozens of "genetic algorithms" and helped other scientists working in bioinformatics to understand them and use them. These things work. They are entirely deterministic, using only pseudo-random numbers, but they can find relationships between variables and figure things out, and after that discovery these found relationships can be scientifically verified.

    But because it is exclusively deterministic, where exactly was the solution encoded? It is partly in the program that chooses whether one "gene" is better than another, and partly in the relationships inherent in the data samples chosen to "train" it.

    But there is no practical way to "jump ahead" and just solve the genetic algorithm, that would require too much knowledge of the data and too much computing power.

    But the result is encoded in the rules and relationships that exist when I press Enter; if I run it again with the same random number seed, then 24 hours later I get the exact same answers.

  35. "It acts as if it is aware of its abnormal high energy and decides to get rid of it at its own pace but that is only our narrative"

    In such cases, as in all forms of radioactive decay, the probability for a transition is constant in time. In other words, there is a certain probability that the transition will happen in a certain time, which is independent of how long the object has been in its current state. This leads (in the large-number limit) to the well known exponential law of radioactive decay.

  36. Sabine,

    “What we are missing are sufficiently powerful mathematical tools. We simply do not currently understand how to quantitatively treat chaos …”
    Is it what you are missing that we cannot predict beyond the Lyapunov exponent - this lack of predictive power?

  37. Reimond,

    No, is not what I mean. What I mean is that you need a way to figure out when a system is, vaguely speaking, interesting in a stable way, or at least parts of it are. I know that this is not a good definition, which is exactly the problem. And if you have a definition you need to be able to actually compute something with it, preferably so that you get results faster than just simply observing the system you try to model.

  38. Castaldo,

    Yes, I absolutely agree with you. The initial condition is of minor or no importance, important are the rules. And the philosophical important point is that the algorithm, the process does not “know” it is using only pseudo random numbers. Just as you said, if it is run the second time with the same seed the result is predetermined.
    From one perspective it looks random, from the other it is deterministic.
    From one perspective it is the magnetic, from the other it is the electric field.
    From one perspective it is before, from the other it is after.
    It is often only the perspective that makes the same thing look different.

  39. Dr Bee,

    Is using theory to compute the time required that half of an unstable system has become stable not enough? Like I said, such a system seems to be interested in itself to become stable. What makes this behavior interesting for us is when such a system is complex enough.

  40. Sabine wrote:

    > And if you have a definition you need to be able
    > to actually compute
    > something with it, preferably so that you get results
    > faster than just simply observing the system you try to model.

    In theoretical computer science there is the notion
    of a "P-complete problem" which captures in a precise way the intuition that you need to simulate (or observe) a system step by step in order to predict its behavior.
    See for instance

  41. Sabine,

    Ok, then I guess it is about limit cycles. But all interesting complex, dynamical, pattern creating systems I know on this planet like water eddies or weather when they hang around in a limit cycle lose their fascination.

    “… you need to be able to actually compute something with it, preferably so that you get results faster than just simply observing the system you try to model.”
    Maybe this is not possible. The behavior is engraved in the rules of the system and maybe you just have to let it run.

    Here the simplest of all non-linear systems with an already fascination rich behavior.

  42. Tim,

    I only have first-person experiences, and for all I know that's the only thing that exists. Yet physics works just fine for me. I fail to see what's supposedly different about consciousness. We'll look for regularities, develop models, and if those models are reliably predictive we'll call them useful descriptions of the world. It's how science works.

    I am not ignoring your arguments. I just fail to see what you think the problem is.

  43. Pascal,

    Yes, but of what use is this if you do not know the math to describe the system at hand? (Or if the math you know is definitely of the uncalculable type, at least for the time being.)

  44. Reimond,

    Yes, it may be that it's not possible but I consider this unlikely. The reason is that natural selection rarely gives you the best possible state, it gives you whatever is good enough to come out ahead. Even if, you do not need to actually predict the system (faster than the system itself) to be able to understand its properties. It's just that if you could that would be preferable. (Though, come to think of it, some people may think otherwise.)

  45. How can you talk about consciousness without talking about the content of consciousness?

    For a particle, the content of consciousness is necessarily what we human beings would symbolically (mathematically) represent as law of nature relationships, and as numbers representing the variables in these relationships. For a bird or a human being, we would symbolically represent the more complex content of their conscious experience with words; though you could also presumably attempt to represent some of the content of their conscious experience with variables and numbers, and as relationships between the variables.

    In other words, the content of consciousness is information: lower-level information for particles; and higher-level information for birds and human beings, that has been wholly derived from analysis and synthesis of huge amounts of the lower-level information. When talking or writing about information, we symbolically represent information via equations, numbers and words. (Obviously, a computer/ robot only processes our symbolic representations of information; a computer/ robot does not process information per se.)

    Any free-will actions that a particle or a person might make is based on the available information but free-will/ action/ creativity/ “randomness” is a separate aspect of reality to the knowledge/ experience/ information aspect of reality.

  46. Sabine wrote: What we are missing are sufficiently powerful mathematical tools.

    I'm pretty sure we agree, even if some of our words don't match exactly. Here's what I originally wrote about "methods":

    "It's not just methods. Sure, we might discover new methods, but there's also new discoveries, insights, knowledge, theories, tools, etc. If by 'methods' you mean all of these things, great, we're on the same page."

    Here's what you wrote:

    "I mean that we do experiments, record data, and develop models, until we have found a useful way to predict experience from measurement . . I think we are missing a big part of the necessary math."

    In my list, when I wrote "tools," I include mathematical tools, not just hammers and screwdrivers. :-)

  47. Amazing that this discussion is still going on, so here is my handful of two cent pieces.

    I haven't read all 600-odd comments, so someone may already have mentioned that Sabine's argument against panpsychism is quite similar to the one Sean Carroll gives in his book The Big Picture. Panpsychism is the reductionist equivalent to dark energy, it's a patch to make the equation balance when you can't find anywhere else to place the phenomenon.

    People have to reduce consciousness all the way to fundamental physics because there's no proper anti-reductionist metatheory that explains how emergence works. Richard Feynman gave a bunch of wonderful examples of emergence reasoning in QED, but he wasn't trying to be rigorous.

    Once your emergence theory can explain the entropy reduction in the origin of life and the information processing in bacterial chemotaxis, you might be ready to attack the recursive processes of homeostasis and locomotive path planning. You might find consciousness naturally evolving from those phenomena. But that's an awfully long chain of reasoning and there are multitudes of ways to go astray.

    As Sabine wriote, "I did not say it requires a new scientific method or something like that. To make a long story short, I said we need better math." I suspect that that better math might look something like homotopy type theory, but I'm not a mathematician. I have no idea how fundamental physics could be recast into HOTT terms. I'm not even very optimistic, given the apparent failure of the axiomatic QFT program.

    It may well be that understanding consciousness, either in general or in any particular instance, including your own personal consciousess, is beyond the capability of the human brain. Colin McGinn's arguments against comprehensibility drift between incoherence and nonsense, in any case. A proper argument against the understandability of consciousness would look more like a proof of the undecidability of the halting problem or an incompleteness theorem.

  48. Bee, you've got a lot of conversations going on here but if you're sincerely interested in understanding panpsychism, just read Griffin's book Unsnarling the Worldknot, and in particular Chapter 7. It's extremely well done and is basically his exposition of Whitehead's panpsychist philosophy for the 21st Century, contextualized within other thinkers' views on the nature of mind and matter.

    I've offered you a lot of feedback on your OP and your comments and you don't seem to have understood or internalized much of what I've offered, including my query about why you ignored and continue to ignore the serious oversight in your OP in assuming that any psyche in an electron would have to lead to different kinds of particles in scattering experiments, rather than simply a choice about position and manifestation in each moment ("choice, not chance").

    In your last response to me you expressed confusion about why I posted my abstract from our latest paper. This was because you stated: "I still don't know how you think that mind "complexifies" if all you have done is attaching a new word to elementary particles without changing anything about the physical implications. In your theory, for all I can tell, the brain works exactly like physics tells us it works." My posting my abstract and the explanatory notes around it was to answer your questions about how mind complexifies as matter compllexifies and why purely physicalist accounts can't provide a place for consciousness.

    I've been reading in the philosophy of mind for 30 years now and began writing seriously about it a decade ago. These are complex issues. There are no right answers. But there are more plausible answers and more explanatorily powerful answers. The notion that today's physics, which categorically excludes mind from the fundamental aspects of nature, can at this point be discarded as implausible and unhelpful.

  49. Tim, would love to see your response to the 1990 paper by Bohm that I posted here where he makes his panpsychist views explicit. I also highly recommend Griffin's book Unsnarling the Worldknot if you haven't already read it. It's probably the single best rigorous case for panpsychism out there.

  50. Steven,

    Ok, then. Doesn't really matter to me whether you want to call it a tool or a method as long as we understand each other. Thanks for your patience.


  51. G3 McK,

    It's been a while that I read Sean Carroll's book, so I can't recall exactly what he wrote there, but I noticed on several occasions that Carroll's and my views on emergence are very similar, so it's quite possibly the case. Same thing with the above-mentioned quote from Brian Kox. Actually I think anyone who has a background in particle physics would tell you the same thing. Reductionism isn't a philosophy, it's an experimentally well-confirmed fact that one can't just throw out if one doesn't like the consequences.

  52. Tam,

    What you call "feedback" I would call ignoring my questions and repeating the same things I have debunked over and over again.

    You are welcome to call a particle's position a "choice" but I don't know what that's supposed to be other than verbal gymnastics. I told you several times, Tam, write it down. Write down your theory. Define "particle", tell me what mathematics it is described by (or if not, how you want to recover the standard model), define "choice", tell me what mathematics it is described by. (And, on that occasion, please also explain why you think it is justified to call it "choice".)

    You cannot talk about particles without explaining how you want to get back all the correct predictions of particle physics. This would be a step back by several hundred years.

    I also already told you above what's the problem with your resonance idea. You did not even bother to reply.

  53. Castaldo wrote: So "superdeterminism" could be used to mean these uncaused events are not truly random, the randomness of QM is just a stand-in for the operation of currently unknown mechanisms or relationships

    I originally objected to the term superdeterminism in the context of "truly random" dice. I didn't think superdeterminism was useful in that context; I thought determinism was adequate.

    With deterministic models there can be unknown variables, in contrast to the question of hidden variables in superdeterminism. I made the point that if deterministic models of dice behavior were complete enough, we could take the randomness out of randomness. I figured that deterministic models would be adequate for most of our scientific observations and predictions, hence my questions about its usefulness.

    I don't know enough about QM to have a deep discussion of superdeterminism as a theory. If that's where this discussion is going, I'll try to follow it.

  54. I’ve enjoyed following this discussion.

    This year’s Science of Consciousness meeting is being held in Interlaken Switzerland. The deadline for early registration and abstract submission is today, the 15th. It looks like there will be some focus on dual aspect monism. I’m pretty enthused about that!

  55. Sabine wrote: We'll look for regularities, develop models, and if those models are reliably predictive we'll call them useful descriptions of the world.

    Nice to see a theoretical physicist singing "don't worry, be happy."

  56. Sabine,

    “The reason is that natural selection rarely gives you the best possible state …”
    Yes, absolutely, biological natural selection (BNS) only provides a good solution in relation, relative to the environment.
    Imagine BNS evolved a genius who is capable to fabricate the absolute best screwdriver in every of the possible worlds. Made of chrome-vanadium with self-sharpening edges, a soft rubber handle with a touch of cooling gel, … simply a dream …. But BNS did not yet evolve a genius who invented a screw - the perfect screwdriver would be absolutely worthless, totally screwed up ;-).
    This is precisely what the video/talk mentioned here is about.

  57. @Reimond: You are ignoring cause and effect. The screwdriver gets invented because the screw is invented first; the screw is invented because somebody's brain realizes an inclined plane can wrap around a cylinder.

    Determinism would not mean everything is invented at once, nor does it do away with causality. A computer program's internal state of memory is modified every cycle an instruction is completed. The code doesn't have to be changing (just like the rules of QM and GR don't have to change) but the effects of repeatedly apply the code changes the state.

    If the code eventually succeeds it can produce an answer that was not known when the code began; and did not exist anywhere in the code or in the data when the program began. The answer is a result of the code exposing a particular relationship that existed in the original data. But with enough data, there is a combinatorial explosion of possible relationships, and the code exposes one of them.

    Cause and effect are not eliminated. The can-opener, historically, was invented years after canning became popular for preservation; people used to open the cans with knives. The can-opener was invented because canning was invented.

    Just because the results of cause and effect in a deterministic system can be computed, doesn't mean they can proceed in any order. The "effect" of one cause becomes the "cause" of another effect; and so far we can "compute ahead" only very short times; or get a look at some of the rules governing cause and effect; which is QM and GR (and apparently some we haven't seen yet that would explain the anomalies we detect).

  58. Castaldo,

    I didn´t mean this example to be real. Of course, you just described correctly cause and effect in an indeterministic system - do not forget that the pseudo randomness is real randomness for the algorithm.
    Just watch the video in here to see the problem. Many including Sean Carrol believe the Brexit vote can in principle be predicted, if they only would have known the initial condition, let´s say 10 years ago. They believe in a deterministic, unitary evolution. Your algorithms only pick one possibility with the help of randomness or as you say “there is a combinatorial explosion of possible relationships, and the code exposes one of them”.

  59. Dr. Castaldo said:

    "Determinism would not mean everything is invented at once, nor does it do away with causality."

    How about a view wherein determinism initially applies to the inanimate, but the emergence of life is a wild-card?

    But for the appearance of life-forms, it is easy to imagine everything unfolding in a specific sequence; not planned but inevitable.

    But, life happened, and causality still applies in both domains, animate and inanimate, albeit with the possibility that causality from the animate side can impinge on the determinism of the inanimate side, e.g.:

    Dr. Bee finally realizes her dream of flipping the switch on a space-based, globe-encircling linear accelerator in her quest to solve, once and for all, whether "Electrons can't think".

    We all know what happens next. But for the emergence of life, the Earth would have continued on its determined/causal path, but...

  60. Tim Maudlin, Steve Mason, and others,

    Re “the hard problem of consciousness”:

    Is consciousness an item of information? I.e. is consciousness like temperature: something that might be said to emerge from the local complexity of the universe; yet another entry in the long list of items of information that physicists can represent as a variable/variables and measure, where the measurement results can be represented as numbers?

    Or is consciousness OF information? I.e. is consciousness a thing that potentially applies to ALL items of information, including the variables and numbers that represent the items of information?

  61. Castaldo wrote (to Reimond): You are ignoring cause and effect. Determinism would not mean everything is invented at once, nor does it do away with causality.

    It would be helpful if you quoted the part of Reimond's comment that you're responding to. I'm trying to follow the discussion, but I don't recall where Reimond said anything that would warrant your response.

    Unknown talked about things getting "invented" at strange times and places, but I don't recall Reimond doing that.

  62. @Castaldo

    If you're responding to Reimond's humorous story about a "dream," I'm pretty certain it was just a natural selection joke. Nerdy jokes sometimes don't come across as jokes. :-)

  63. Speaking of nerdy jokes, should understanding jokes and laughing at them be a part of a robust Turing test? I've known plenty of people who aren't very good at recognizing jokes. If AI gets better than humans at recognizing humor, does that mean they surpass us in some meaningful way? If AI can distinguish between good jokes and lame jokes, and if they get bored listening to a hundred knock-knock jokes in a row, so much the better, eh?

    A while back Sabine told us that she doesn't appreciate certain kinds of popular humor. If AI laughs at all kinds of humor, does that make them smarter or dumber than humans? Does that make them no better than laugh tracks on TV sitcoms?

    Random thoughts inspired by Reimond's joke. Of course, if Reimond wasn't joking, it's because I'm a dumb human. :-)

    (Yes, I know Reimond was making a point and it wasn't just a joke. I'm not that dumb.)

  64. Jeff wrote: This year’s Science of Consciousness meeting is being held in Interlaken Switzerland.

    Getting to that conference would be a hard problem for me. :-)

  65. @ Lorraine Ford, who said:

    "Or is consciousness OF information?"

    The link below will bring you in at the most pertinent sequence of the presentation, but I recommend watching it in its entirety, even for those who may view it as stand-up comedy:

  66. Lorraine Ford,

    I can't see that the concept of information (Shannon or otherwise) has any conceptual connection to the Hard Problem at all. The point is that pain *hurts*, that *there is something it is like to be in pain*, but (according the Sabine and me) there is nothing it is like to be an electron, and (according to me) even a pair of entangled electrons. The pain state may or may not contain Shannon information about anything. That is irrelevant to its nature as a first-person conscious experience.

  67. Tim Maudlin,

    In order to survive, human beings and other creatures need to acquire information [1] about their local surrounding environment. “There is something it is like” to experience that more subtle information, just like “there is something it is like” to experience glaring pain. (The pain caused by a wasp bite is presumably survival-related information.)

    But re the connection between information and the Hard Problem: In order to survive, a creature will likely need to be able to handle a large amount of simultaneously occurring information relating to its surrounding environment. But the creature doesn’t draw up a diagram or a table of variables and numbers on a piece of paper to keep track of all the information: instead the creature is able to easily handle multiple pieces of information only because all information categories feel different.

    I’m contending that the main content of consciousness is information: “higher-level” information has a logical relationship to, and has been wholly derived from, huge amounts of “lower-level” information coming from interactions with particles, atoms and molecules from the surrounding environment. This lower-level information is the sort of stuff that physicists represent with equations, variable and numbers.

    You say “there is nothing it is like to be an electron”, but I would say that “there is nothing it is like to be a universe”: the universe knows nothing; in no sense, abstractly or otherwise, does the universe know any “law of nature” relationships, or any information whatsoever. It is more reasonable to think that the carriers of information (that we would represent as variables and numbers) and “law of nature” relationships are things like particles.

    1. Information: “Facts provided or learned about something or someone”, “What is conveyed or represented by a particular arrangement or sequence of things”, .

  68. R. Taylor,

    I think Rupert Sheldrake might have some pertinent things to say, including criticisms of science, but I don't agree with his views about consciousness.

  69. Lorraine wrote: is consciousness like temperature: something that might be said to emerge from the local complexity of the universe

    I don't consider temperature an emergent phenomenon because we understand the mechanisms that cause temperature. We normally refer to emergence when the mechanisms aren't apparent.

    Imagine that you're surrounded by a circle of people and all of them are throwing tennis balls at you. Imagine that you give a name to being hit by all those tennis balls, and that name is "temperature." If more balls are thrown, or if they are thrown harder, the temperature goes up. There's an obvious link between the tennis balls and the temperature. Actually, this simplistic analogy isn't a bad way to explain temperature in the context of weather.

    It wouldn't be logically or grammatically incorrect to say that temperature emerges from the movement of air molecules, but it's not the usual way scientists use the term. If we could somehow emulate consciousness by throwing billions of tennis balls around in a chamber, we could rightly say that consciousness emerges from the interaction of tennis balls. That's because the link between tennis balls and consciousness isn't apparent.

    I'll say again that this is a simplistic analogy. If someone wants to devise a better analogy that explains why temperature isn't considered to be an emergent phenomenon but consciousness is, go for it.

    And please, let's not rekindle the tiresome debate about emergence. Anyone who is nostalgic for that can re-read the previous debate.

  70. Taylor wrote: I recommend watching it in its entirety, even for those who may view it as stand-up comedy

    Why do you recommend the video? What point are you trying to make? It took two minutes to see that it's rubbish. If I watch the entire presentation, would you expect me to change my mind?

  71. Sabine,

    The last four days I had a cold, this might explain the nerdy, natural selection joke. Sorry, I only was able to grab your first part. At that time, I did not get your important second part.
    “Even if, you do not need to actually predict the system (faster than the system itself) to be able to understand its properties. It's just that if you could that would be preferable.”
    With that I absolutely agree. So, let us expose some properties.

    “… complexity … isn't even well defined …” Yes, right, but when we see it, we recognize it.
    Some examples:
    - Mandelbrot set: iterating non-linear z²+c; threshold - the number of iterations; no randomness.
    - snowflakes: non-linear dynamics; threshold – the breaking of symmetry of water vapor at surface spots; depending on the initial condition at the moment of crystallization; heat dissipation at crystallization;
    - turbulence: non-linear dynamics, dissipation across scales (defined by thresholds down to the molecular level);
    This here reminds me of renormalization group flow, Beta-function (0).

    What are the complexity generating properties (1) in these examples?
    A non-linear process; one or many thresholds; heat dissipation.
    Mandelbrot set has no time dependence, therefore no dissipation, also no randomness (2).

    “entropy locally can decrease while globally it can increase, it's a process mostly driven by gravity.” Yes, complexity and life is finally driven by gravity and the standard model. For every visible photon from the sun we get rid of about 5 infrared photons. Getting rid of entropy, i.e. dissipation is all important. (3)

    (0) Until recently I did not know that Gerard ´t Hooft was among the first to discover asymptotic freedom in QCD as he says in here “When I heard about the publications by H.D. Politzer, D. Gross and F. Wilczek, I was surprised, not about the result, which I had known all along …” – could have been his second Nobel price.
    (1) Kolmogorov (-Chaitin) complexity is a measure for compression. For the Mandelbrot set this is just a handful of bytes: the threshold and z²+c. In physics all we do is to find the laws, i.e. the correct compression.
    (2) Of course, to generate this (Shannon) information, i.e. the correlated pixels, CPU time is used and heat dissipated, but that´s another layer. A clear separation of concerns (SoC) always helps as mentioned here.
    (3) only since we are a driven by the flow of photons (low entropy in, high entropy out), i.e. (1 dof in, 5 dofs out) planet earth is a spot of low entropy. This enables the origin of more or less stable structures that are complex and contain high amount of (Shannon) information i.e. the correlation of their constituents, like proteins, brains. It is all about entropy and information. No net energy intake (well, at least if we believe climate change deniers ;-).


  72. Thermodynamics (TD):
    The heat eq. connected to dissipation is a deterministic linear partial differential equation. It is apparently paradoxical that it is intimately related to a random, a Markov, a Wiener process, like Brownian motion. It is not retrodictive, but perfectly predictive. More or less independent of the initial condition all becomes squishy as shown here (4).

    From the heat eq. by Wick rotating t -> it we get the Schrödinger eq., which describes a unitary evolution. All is perfectly retrodictable and predictable. We also need a factor ℏ/m, which might remind us that we are talking about light stuff. The imaginary time, that transformed a diffusion (5) into a wave eq., also is screaming now “square it to get probabilities”. Somewhere the measurement problem is lurking.

    Formally e^(iHt) and e^(-Hß) with ß=1/kT are just analytically continued. But why is Boltzmann and Schrödinger connected in equilibrium, such that temperature T becomes cyclic imaginary time t or equivalent why Euclidean quantum field theory in spacetime, with cyclic boundary condition becomes quantum statistical mechanics in space?
    A. Zee says Some physicists, myself included, feel that there may be something profound here that we have not quite understood.

    You said here that a measurement must be a non-linear intervention that cuts through a linear, unitary evolution. A measurement in e.g. the double slit experiment distributes energy and momentum from the source to the sink. E.g. in the photon case it does so since in a laser higher energy levels (static Schrödinger) are populated (Boltzmann). QM distributes it further from the source to the sink without any loss, since the unitary evolution in QM guarantees, that probabilities add up to 1. One possibility is realized with fundamental QM randomness. In the measurement energy/momentum vanishes from the source and shows up at a spot somewhere on the screen.

    Distributing energy/momentum is a property of TD.
    Might it be that TD is just an emergent property of myriads of tiny observer independent triggered measurements/reductions (R) that redistribute energy/momentum everywhere and all the time? With unitary evolutions (U) in between to calculate probabilities.
    Might this be the reason why temperature becomes cyclic imaginary “time” in equilibrium?
    We are not living in a world of classical matter, molecules are no billiard balls. If we would set ℏ=0 this world would not exist. QM matter only looks classical, because Schrödinger cats do not get too heavy. This world is a subtle interplay between QM (QFT+SM) and GR.

    We have to add to the list here yet another dichotomy: irreversible/reversible.
    This one gives us the flow of time.

    (4) The heat or diffusion or flow eq. tends to smooth out sharp edges, which we could call “curvatures”. Real spacetime curvatures are smoothed out via Ricci flow.
    (5) The imaginary part of a solid-state Greens function is the dissipative part of the system, that comes from a bit of randomness in the lattice. With this imaginary part conductivity is calculated.


  73. The factors of the actions of GR and QM are about 77 orders apart. If the fundamental QM randomness is real, then also GR and thus the block universe must inherit some tiny, tiny bit of it. Where to sneak in localized QM matter? Well, in EFEs the time component G°’= T°’ contains no second time derivative, it is just a constraint on the initial values analog to Gauss law as mentioned here in footnote (3). This would break the symmetry between the future and the past. This would be a realization of George Ellis’s evolving/crystalizing block universe.
    This would neither change GR nor QM - they just would work together via measurements.
    “There is a crack in everything, that’s how the light gets in” (L. Cohen - just an argument by authority ;-)

    QM and GR work like a charm, we just need to integrate QM and GR into a process, instead of exclusively either putting an action into a path integral as Feynman taught us or just integrating differential equations as Newton taught us to do.

    “I think we are missing a big part of the necessary math." (You referred this to consciousness and complexity)
    “… if those models are reliably predictive we'll call them useful descriptions of the world.”
    Well, I think we already have almost all of the right math and all of particles (6), we just need to change the interpretation and lower our expectations. If we would be able to predict all we would be god. But I do not believe in god, but I believe that reductionism works.

    (6) dark matter could now be explained without new particles, once QM and GR are incorporated in a non-local process, since now de Sitter entropy can be part of the dynamics, once the acceleration falls below a certain threshold. Just following Verlinde´s proposal (he already breaks with ‘spacetime entanglement entropy’).
    For life on earth the tiny “TD” entropy plays the dominant role.

  74. Reimond wrote: the nerdy, natural selection joke

    I called it, so that's one brownie point for me. By the way, it was a funny joke.

    Reimond wrote: QM matter only looks classical

    Yes, but it's a great look. Dare I say it's a beautiful look?

    Reimond wrote: I think we already have almost all of the right math and all of particles

    Maybe, maybe not. In my opinion it's way too soon to have a (meaningful) opinion. Maybe I'm too eager to take the copout position of "I don't know."

    In Oscar Wilde's play The Importance of Being Earnest, Lady Bracknell asks Earnest if he knows everything or nothing. He admits that he knows nothing, to which Lady Bracknell replies, "I am pleased to hear it. I do not approve of anything that tampers with natural ignorance."

    Reimond wrote: we just need to integrate QM and GR into a process

    While you're at it, you can throw in determinism and superdeterminism. :-)

  75. Steven Mason,

    I agree: temperature is a category of information that is explainable in terms of other, known, categories of information like kinetic energy. Temperature, and seemingly metal conductivity and fluid viscosity are not genuine emergent phenomena: they are merely a logical consequence of the physics. A truly emergent phenomenon would not be a logical consequence of the physics.

    Existing physics is representable in terms of “law of nature” relationships, variables/categories of information and numbers. But seemingly, to represents consciousness, you also need algorithms. Where are the algorithms coming from?

  76. @Lorraine Ford: Where are the algorithms coming from? From natural selection; they are evolved capabilities that "emerge" when we start combining neurons (pattern recognizers) that process nerve signals to create new nerve signals indicating recognition of a generalized sensory pattern.

    A single neuron is a sophisticated classifier of a collection of signals. Arranged in a network they can become, more than linearly, better modelers of what is being sensed in the outside world. If they become recurrent (because nature does not have any reason to prevent that) then they become, like a simple IIR (Infinite Impulse Response) filter, constantly self-activating.

    There does not have to be a mystery here; the algorithms are in the hardware, just like on this computer I am using. The "software" is encoded as hardware signals and the rest of the hardware is designed to be infinitely recurrent (via interpreting certain signals as "branches") until the power is shut off. (Which it can do itself, but when it does, it continues to loop until it dies.)

  77. “I don't consider temperature an emergent phenomenon”
    Well, yes, ok, it is simple, but I would call temperature an emergent phenomenon (weakly emergent as always), since you need a bunch of microscopic classical particles rattling around.
    Heat is explained nicely here in the turbulence video. And the change in heat dE and the change in entropy dS is given by: TdS=dE. Where entropy can be defined as S=ln Ω and Ω is the number of microstates.

    Where does TdS=dE come from?
    Consider two systems, e.g. two cups of tea with Ω’ and Ω’’. We bring them in contact then Ω=Ω’Ω’’ and S=ln(Ω’Ω’’)=lnΩ’+lnΩ’’=S’+S’’. After a while heat was exchanged and the combined system is in equilibrium, i.e. entropy S is at maximum. Further exchanging a tiny bit of heat dE’=-dE’’ means 0=dS/dE’=dS’/dE’+dS’’/dE’=dS’/dE’-dS’’/dE’’ ⇔ dS’/dE’=dS’’/dE’’. This property dS’/dE’ which is the same in both systems is now defined as temperature via dS/dE=1/T (⇔TdS=dE).

    Thus, temperature has also a relational aspect. It will be the property when two systems stay in contact. But for this it needs a process.
    And getting rid of entropy (dS=dE/T) is all-important for complex systems that are generating information (information is the correlation in their constituents like protein and brains, very special points in phase space, much more than limit cycles, but also stabilized via feedback, non-linearity)

  78. Dr. A.M. Castaldo,

    There is an assumption that (what we would represent with) algorithms [1] derives from a universe representable as equations, variables and numbers (and no algorithms).

    But an approach which uses algorithms to “prove” that algorithms “evolve” is not good enough; and an approach which uses already highly-evolved and ordered neuron-cells to “prove” that something-or-other evolves is not good enough: you need to show from the ground up that simple algorithm-rules are a logical consequence of a situation representable by nothing but equations, variables and numbers, a situation where nothing new has been added (the new addition will represent a new rule having been added to the situation).

    And of course, you can’t prove any such thing, even in very simple cases. There is a hierarchy here: algorithms are higher in the hierarchy than equations; as in computer coding, the algorithms control the equations; the algorithms do not derive from the equations, variables and numbers, no matter what the situation.

    By the time cells appeared on the planet, the something that is representable by algorithms was already there.

    1. Algorithm: “A process or set of rules to be followed in calculations or other problem-solving operations, especially by a computer”,

  79. I should have emphasized above that we best start with a hot and a cold cup of tea.
    We live because the sun is hot, T≈6000K and outer space is cold, T≈4K.
    Life feeds on bits of low entropy energy. This cascades through the organisms and in every level (photosynthesis, sugars, proteins, …) entropy adds until finally for each visible 5 infrared photons are radiated off.
    This cascade of energy in life is analog to the cascade here in the turbulence video.

    “While you're at it, you can throw in determinism … :-)”
    That´s easy. Let us assume that in each QM measurement the “decision”, the dice is realized by a pseudo random number generator (pRNG). Further let us assume that our universe is so poor that it has only a single pRNG, a single QM dice where all of fundamental QM randomness is taken from. Then if someone, let´s call her/him god restarts the system with the same seed all would be predetermined. But of course, this holds only from her/his perspective.

  80. Btw since Castaldo just mentioned “IIR (Infinite Impulse Response) filter” I would like to bridge over to responds theory, which basically can be comprised as: If you have a system the best way to analyze it is to take a hammer and bang on it and listen to the response. The hammer is a Dirac delta function δ, the system L is e.g. defined by a partial differential eq. and the response is the Green´s function, the kernel G which is: LG=δ.
    E.g. the heat eq. has a Gaussian as kernel orhere the famous 1/r for the Laplacian.
    In QFT the Green´s function becomes the propagator for electrons, photons.

  81. Tim,

    Here in the exchange between Sabine and Wojciech about naturalness the term “radical ontic structural realism” popped up with reference to this post “electrons don’t think”.
    I never heard this term before, so I checked it here and obviously it connects quite well to this discussion.
    I suspect that it is intimately related with Bell’s local beables, but maybe you could comment on this.
    “Tim Maudlin argues against Lewis's Humean Supervenience on the basis of quantum entanglement and argues that this means the end of ontological reductionism, and abandoning the combinatorial conception of reality…”
    Especially since the terms supervenience and combinatorial appear.

    Btw when I talked above about “localized QM matter” I meant this to be analog to Bell’s local beables. Since in a reduction an entangled QM state and its mass/energy is mapped onto spacetime at least in a GRW-like theory.

  82. @Lorraine Ford:

    I disagree. First, the biological literature already contains detailed steps using extant organisms for a plausible route of neuron evolution in increasing complexity, from simple sensory cells to nerves to neurons. We don't need something to start.

    Second, algorithms are not "primary" in any sense, any algorithm, mathematical, statistical, neural, electronic, is just a model of a system. It is we humans that develop rules, and we follow them in order to understand how the universe works. The universe doesn't "follow" rules, it has no choice to follow or not follow; it doesn't do any calculus or averaging. Matter interacts and transforms itself from one package of energy to others without computing anything.

    Thus "algorithms" are always secondary to nature; our neurons evolved to model the world around us and make predictions of probable futures from those models, which caused our ancestors to make choices that selectively increased our survival and reproduction. It just so happens that when the models become good enough, we can model many other things that may have nothing to do with survival; like multiple dimensions, black holes, the physics of particles, etc. Perhaps someday they might matter, in solving energy problems, health problems, letting us expand to other worlds, or who knows what.

    The universe does not follow, obey or do anything with algorithms. It does no math. It can make no mistakes. Humans devise algorithms, and compute with them to model whatever it is that the fields of the universe are doing.

  83. Reimond wrote: but I would call temperature an emergent phenomenon (weak emergence)

    I have no objection. My general point was that when scientists refer to "emergence," especially in the context of things like consciousness and abiogenesis, they are usually referring to so-called strong emergence, and there are significant differences between the two.

    Reimond wrote: Then if someone, let´s call her/him god restarts the system with the same seed all would be predetermined.

    I was making a nerdy joke about integrating GR and QM, which could in turn settle questions about determinism/superdeterminism. But I appreciate your response.

  84. Lorraine wrote: But seemingly, to represents consciousness, you also need algorithms. Where are the algorithms coming from?

    If I understand your question correctly, the default answer for scientists is that the "algorithms" would be natural mechanisms. Life and consciousness would be "coming from" the same sources as all the other phenomena in the universe.

    I'm not sure, but if I were to generalize your question, are you asking how entropy can decrease? Do you think that the decreased entropy of life and consciousness are special?

  85. Dr. A.M. Castaldo,

    Biology is not physics, even though it is physics underlying the biology. This because the complexity of biology means that you can’t necessarily have genuine cause and effect like you have with physics (Even physics is not 100% cause and effect: e.g. the (assumed to be) random outcomes of quantum events.) So “the biological literature” can only approximate cause and effect, when it comes to the behaviour and evolution of cells. You do need to look deeper than biology if you want to find genuine cause and effect.

    We represent reality via written and spoken words, and mathematical symbols [1]. Models are not so special: they are basically just another form of representation. Physics uses “law of nature” equations, variables and numbers to model/represent aspects of the universe-system; biology might also use mathematical models, variables and numbers to model/represent aspects of the universe-system; but because of the complexity of what is being represented, the biological models can’t necessarily get at real causes. On the other hand, it is justifiably assumed that physics’ “law of nature” equations ideally represent genuine cause-and-effect rules that apply in the universe.

    And what physicists represent with the delta symbol in their equations is an algorithmic step. So (what we would represent with) algorithms are a primary aspect of what the universe is doing.

    1. We use written and spoken words, and mathematical symbols, to represent the content of our consciousness. We represent reality via the content of our consciousness.

  86. Steven Mason,

    I wasn’t referring to entropy, an indirectly measurable quantity, or any of the variables of physics. In general, physics uses categories of information that can be represented by a variable, and whose quantity is represented by a number, and whose relationship to other such variables is represented by an equation. The variable (on one side of the equation) is a relationship; the variable doesn’t have behaviour; the variable’s fixed “behaviour” is representable by the equation.

    On the other hand, life and consciousness is not representable with equations (representing fixed behaviours), but with algorithms representing IF….THEN decisions, made on the basis of numbers (depending on the numeric value for particular variables, the decision tree will branch to an outcome or another decision). With life and consciousness, these decisions are sometimes one-off, and other times the decision pathways have seemingly been built up over time. And whereas an equation is a relationship that says nothing about the numeric values of a variable, an algorithm is very concerned with particular numeric values or a range of numeric values for the variable.

    An algorithm represents 1) a higher, more detailed, level of knowledge than an equation (i.e. knowledge of particular numbers applying to a variable); and 2) decision making and outcomes based on that knowledge.

    So, can an algorithm emerge/evolve out of equations? Despite all the assumptions that something representable by an algorithm can emerge out of something representable by equations, there has never been a proof of concept; and intuitively, such an emergence/evolution is impossible. So, I would say that (what is representable as algorithms) did not emerge out of (what is representable as) equations!!

  87. Reimond,

    Radical Structural Realism (A.K.A. Ontic Structural Realism) is a term of art in philosophy that, as far as I can tell, designates a completely incoherent view. The tag line for that view is "Relations Without Relata". But as is kind of obvious, it is impossible to have relations without relata, i.e. for there to be relations but fail to be anything related by them. It is somewhat akin. to nonsense like "the world is made of information". There is a strain of gnostic pronouncements about quantum physics that starts with Bohr and runs through Wheeler and is basically (like patriotism) the last refuge of a scoundrel.

    There is a reasonable position called "Epistemic Structural Realism", which holds that we can have better and more justified epistemic access to facts about certain relations than we can have about their relata, That's perfectly fine, and was one line of thought used to save Scientific Realism from the Pessimistic Meta-Induction (more terms of art in philosophy of science, which I could explain but don't really see the point of doing here). But Radical (or Ontic) Structural Realism has pretty much always been "full of sound and fury, signifying nothing" (I omit the start of the quote out of politeness).

  88. If only you were only making this point, that all philosophers are happy to accept, instead of trying to do philosophy... That would be a relief.

  89. @Lorraine Ford; I sincerely doubt consciousness has anything to do with IF/THEN decision making; that is just a convenient model we use to describe what is happening. Our best model for neural nets is not IF/THEN, but a simple reactive model; a neuron takes inputs (from sensors or other neurons) and this may or may not produce a reaction; but there is no "decision" made, just like a lever is not "deciding" to lift a rock when pressure is applied to the handle.

    To say "life and consciousness is not representable with equations" is a false assumption. If one makes the most obvious assumption, that life and consciousness require no special "spirit" or new physics, and are examples of biochemistry in a self-stimulating feedback loop, then they can be represented by equations.

    I don't see how this viewpoint provides any pragmatic advantage in solving any problems. As far as I can tell you are arguing for some new religious elements. This is not science or logic.

    So --- have fun, I hope it makes you feel good.

  90. Tim,

    Thanks for the explanation. After rereading “Epistemic Structural Realism” in here I also would subscribe to epistemic, since “We cannot know the individuals that instantiate the structure of the world but we can know their properties and relations.”. And yes, "Relations Without Relata" is cot convincing.
    When it comes to the QM state then I subscribe to ontic, i.e. ψ-ontic, since the world is not about us getting updated. Anthropocentricity never served science well.

    Btw I never understood Carlo Rovelli´s relational QM, but on the other hand I like his relational view very much, e.g. in “Why Gauge?”.

  91. Reimond,

    Hrvoje Nikolić has recently said that "Relationalism, as an interpretation of QM, is nothing but euphemism for solipsism". Having talked to Carlo about his view, I am inclined to think this is accurate. And as Bell said, it is impossible to refute solipsism, but if you take it seriously then it is hard to take anything else seriously (e.g. why have life insurance?).

  92. Lorraine wrote: can an algorithm emerge/evolve out of equations?

    In the context of consciousness, if we ever get to the point where we can model it, would you consider the model to be an algorithm of sorts? We refer to weather as a chaotic system not because it's random, but because it's too complex to make long-term predictions. There are "algorithms," but they only get us so far. The answer to your question might have to wait a thousand years or so. :-)

  93. Tim wrote: if you take [solipsism] seriously then it is hard to take anything else seriously (e.g. why have life insurance?).

    Even people who take solipsism seriously have to play the game or suffer. Forget life insurance. Why should a solipsist eat?

    You know, I'm an atheist and I've heard the same argument from theists, that it's hard to take anything seriously if one is an atheist. The fact is, no matter what one believes, we're all playing the same game: Pursue pleasure and avoid pain. Do you know of anyone who doesn't take that game seriously?

  94. Reimond wrote: After rereading “Epistemic Structural Realism” in here I also would subscribe to . . .

    As a teen, when I became an atheist I also gave up the expectation that humans could perceive or understand true reality. I held on to other "special" qualities, such as our ability to create meaning and purpose and our ability to make useful predictions. I'm happier with my uncertainties about reality than I ever was with my certainties about God.

  95. Dr. A.M. Castaldo,

    Yes, algorithms are a “convenient model”. I would argue that this is because algorithms mirror something of the actual structure of conscious decision-making, in the case that there is more than one possible numeric outcome for a variable representing a physical outcome. I agree that when (what we would represent as) equations decide all the numeric outcomes for the variables, consciousness has nothing to do, and so no algorithms are required to represent the action of consciousness.

    I’m not sure that there is any “pragmatic advantage” in representing conscious decision making as algorithms. But feedback loops are just another type of information “input” to a situation: they have nothing to do with whether or not the situation requires a decision to be made on multiple possible outcomes, so they don’t solve the decision-making problem.

    You are right that decision making does require a new element that is not part of physics. Decision making with multiple possible outcomes can’t be modelled by physics equations, variables and numbers

  96. Tim Maudlin,
    Seemingly you don’t like the idea that the content of consciousness can be thought of as first-person information (not Shannon information). So how would you describe the content of consciousness? Surely consciousness is not so mysterious that no one can attempt to analyse it?

  97. Steven Mason,
    Why wait 1000 years? :-)
    I think that there seems to be two main aspects of consciousness: it’s decision-making action (which could be modelled by algorithms), and it’s knowledge/information content. Unlike weather, consciousness knows something from a point of view; and unlike weather, consciousness seems to make decisions in the case that more than one physical outcome is possible.

  98. Lorraine Ford,

    The nature of conscious experience is not in the least mysterious: everyone who has felt a pain knows what that is like.

    The question is how the existence of such a thing can arise from physical conditions.

    "It hurts" is a pretty good description of the content of one sort of conscious experience. Nothing mysterious about that at all.

  99. Lorraine wrote: Why wait 1000 years?

    Isn't that infinitely better than jumping to unsupportable conclusions, which is what some people prefer?

    Lorraine wrote: Unlike weather . . .

    By definition, analogies can never be completely "like" the things they're being compared to.

    So, how is weather "like" consciousness? First, they are both complex systems. Second, even if we manage to understand consciousness to the degree that we understand weather, the predictive power of our models can only go so far. Third, consciousness, like weather, might not be a hard problem. Fourth, consciousness, like weather, might be "a logical consequence of the physics." Fifth, consciousness, like weather, might not require any "new elements of physics."

    I could offer more, but that's sufficient to redeem the analogy.

  100. Steven Mason,

    It is always necessary to jump to some sort of conclusion, some sort of model. Only time, discussion and experiment will decide whether a conclusion/model is actually “unsupportable”: you can’t know beforehand, especially with something like consciousness where there isn’t even any agreement about the connection between the physics and consciousness. Timid, safe and scientifically orthodox ideas have so far failed to shed much light on consciousness. And it is only very recently that scientists gave themselves permission to even talk about consciousness. Consciousness might be a hard problem, and consciousness might require "new elements of physics", but there is no need to quit in defeat because of a few “mights”.

    Is consciousness nothing but a “complex system”? Is individual consciousness controllable by the individual; or is individual consciousness as uncontrollable as the weather, where the weather is the logical outcome of an interconnected complex system which extends outward to the entire universe? The weather doesn’t control itself, but do individuals control themselves? (I’m talking about genuine control, not about faux control where there might be a surface appearance of control.). Algorithms are used to represent control.

  101. Tim Maudlin,

    I think your seeming insistence that the content of conscious experience is nothing but feelings and sensations, with no information content, is a stubborn refusal to face facts.

    People who have nerve damage and can’t feel parts of their body, and people who have a genetic condition whereby they can’t feel pain, are subject to increased risk of injury and death [1] because they don’t have access to the important information that the conscious experience of pain provides.

    If you take the above disorders to an extreme, you would have a zombie, philosophy’s hypothetical being that does not experience anything, but which philosophy claims could respond exactly like a normal human being. But the above disorders show that a zombie could not respond like a normal human being, because experience itself (e.g. of pain) is the exact thing that makes it possible to respond like a normal human being. The pain experience has an information content which in effect warns a person to avoid things that could harm their body.


  102. Lorraine Ford,

    I nowhere asserted that *all* conscious experience is nothing but feelings and sensations. It is sufficient to establish the character of the Hard Problem that *some* is. Pain is one such example.

    Humans who cannot feel pain will not get certain sorts of (Shannon) information about their bodies that typical people do. And people who feel phantom limb pain will be subject to certain false beliefs about their bodies that typical people are not. Neither of these observations about the informational or misinformational content of pain sensations sheds any light at all on the intrinsic character of these experiences, i.e. their painfulness. It is that intrinsic first-person character that has to be accounted for somehow, and that physics is of zero help with.

    The genetic disorder of not feeling pain, taken to an extreme, yields nothing like a philosophical zombie. Even the less extreme examples do not: people who do not feel pain most definitely do not act like typical people. Indeed, due to that they are prone to dying early. The question whether philosophical zombies are metaphysically possible is hotly contended. I do not believe they are. But the metaphysical possibility of an entity that exhibits all normal pain behavior, including the avoidance of harmful circumstances, yet feels no pain is, I think, obvious. That can be achieved by a lookup table that has none of the features required to produce conscious experience.

  103. Lorraine wrote: It is always necessary to jump to some sort of conclusion, some sort of model.

    I make distinctions between a hypothesis and jumping to a conclusion. There are people who speak with certainty about consciousness, as if the evidence we have now is sufficient to have a conclusion.

    Lorraine wrote: Is consciousness nothing but a “complex system”?

    Well, that would be the materialist model, right? Would you agree that consciousness might have a natural explanation? Is there sufficient evidence to rule it out?

    Lorraine wrote: Algorithms are used to represent control.

    If you're saying something about free will and determinism in the context of "algorithms," I don't understand.

  104. @Tim Maudlin says That can be achieved by a lookup table that has none of the features required to produce conscious experience.

    First, I don't think we can be certain of that claim; and Second, how do you know that the process and biochemical activity and consequences of consulting this "lookup table" does not itself produce the conscious experience?

    On the first point, the nerve circuitry to exhibit normal pain behavior is indispensable; if the nerves are prevented from reacting (topical anesthetic) the pain behavior is not there; if the body is born without the proper circuitry (the CIPA you discuss) the behavior is not there.

    I argue the rest of the brain circuitry is equally indispensable; if suppressed by anesthetic the pain behavior disappears.

    Which brings us to the second point; the way our brain works and neurons react can be the "conscious experience", and there is no reason to believe any other "lookup" algorithm can duplicate the "normal pain behavior" you are talking about. That behavior can include fear and urgency and determination to seek help, for example, which all require predictive behavior of what the pain means for the immediate future. It includes an escape response, which requires identifying the source of the pain, itself a neural modeling operation to map a neuron firing to a physical external location in the body and fire neurons to control muscles to pull away, or somehow remove whatever object is causing the pain, which itself requires a neural model of things that can cause pain.

    And because the brain is not organized like a computer, all these brain signals fly around willy nilly, activating thousands or millions of neural models that may also fire and activate thousands or millions more neural models.

    It may be that the only way to fully duplicate normal pain behavior is to replicate the brain circuitry that produces that behavior, and that cascade of neural models responding and interacting with each other may very be the "conscious experience".

    Proposing that something incredibly complex can be replaced with something else extremely simple is not a valid premise for argument, without any evidence the incredibly simple replacement is even plausible.

  105. Dr A. M. Castaldo,

    The point is a conceptual one. Any deterministic system of any kind underwrites an inout-output map. For the human brain, think of it as a map from the nerve stimulus coming into the brain to the motor-control nerve outputs. Let the "input" be the sum total nerve inputs from your birth until now, so for a conversation the "input" is not just the question currently being asked but the entire exchange. Obviously, any system that underwrites the same input-output map will produce the same observable behavior and therefore behavior that is equally fit as far as natural selection cares. And one way to support that map is by a hardwired look-up table. It would be physically impractical, but that does not affect the force of the argument. If we accept that such a look-up table computational architecture would not support conscious states, then all forms of behaviorism, information processing accounts, etc. are dead.

    Dynamical complexity can always be replaced by dynamical simplicity at the cost of a huge number of distinct internal states. That is always a possible way to implement and input-output map. That is indeed a "valid premise".

    Replacing a deterministic input/output map with a stochastic one does not change the situation.

  106. Tim wrote: The point is a conceptual one.

    That's how I understood it. I don't think you're proposing "an incredibly simple replacement" for consciousness. When you refer to "dynamical simplicity," I don't take that to mean simple in any absolute sense. In fact, you stated explicitly that you don't think it's possible to build a p-zombie.

    I think your response to Lorraine is valid.

  107. @ Tim Maudly, Dr. A.M. Castaldo:

    Feeling pain may be a perfectly valid example to explain what "qualia" means, but "strategically" it is probalby not be the best choice. Because pain is usually a complex experience involving cognitive, emotional, and physiological components, most of which may be amenable to reductionist explanation. This gives deniers of a special status of qualia the chance to jump over the real problem. Perhaps it would be better to stick to the other standard example: "how it is to see a color". This is a pure qualia experience that comes, I think, without appreciable cognitive etc. side effects, nor does it provide any advantage in natural selection.

  108. Tim Maudlin,

    Pain and other experiences have nothing to do with coded messages, or calculations made in connection with coded messages: i.e. experience has nothing to do with Shannon information. Messages use strings of symbols, but pain etc. is like a message before it was represented as a string of symbols. Pain and other experiences are not messages, though people might represent their experience with a sentence consisting of a string of symbols.

    Nevertheless, experience seems to have structure, and I would think that the structure consists of categories of information in some sort of logical relationship with other categories of information, forming new categories of information. I would think that the structure of conscious experience is analogous to law of nature relationships, except that law of nature relationships use (what we would represent as) mathematical relationships instead of logical relationships. (Relationships are between aspects of things, not between things themselves.)

    So I would say that consciousness is physics: physics already seems to assume some sort of algorithmic logic in its equations; and physics already assumes some sort of point of view knowledge of aspects of things and their relationships to other aspects of things. This knowledge is awareness/experience because it is not a coded representation. I.e. experience is part of physics.

    A philosophical zombie is “a hypothetical being that responds to stimulus as a person would but that does not experience consciousness” [1]. But a philosophical zombie cannot exist if conscious experience is part of the physics that determines normal human behaviours.


  109. Tim Maudlin,

    Re your reply of 1:39 AM, January 27, 2019:

    You haven’t said what you think carries information in the universe:
    “Law of nature” relationships (which we represent as equations, variables and numbers), might be thought of as primitive information, i.e. what there was to know in the universe before complex molecules and living things came on the scene. So, do individuals (particles and all things made out of particles) carry information, OR is it a global entity (the universe, a God, or a Platonic realm) that carries information? Your contention that “It is that intrinsic first-person character that has to be accounted for somehow, and that physics is of zero help with” only seems to be a problem if you are of the opinion that it is a global entity (the universe, a God, or a Platonic realm) that carries the primitive information, and not things like particles.

    You haven’t said what you think the fundamental nature of knowledge could be:
    Is there something underlying our symbolic representations of knowledge (e.g. words and sentences; “law of nature” equations), and our symbolic representations of symbolic representations (e.g. binary digits that represent words, sentences and equations; Shannon information; “a lookup table that has none of the features required to produce conscious experience”)? Do you think that there is something or some quality, that is not itself a representation, underlying all our representations?

  110. Tim Maudlin,

    Re your 3:16 PM, January 03, 2019 comment:

    With your definition of “free will”, even a billiard ball has “free will” [1]. You are talking about situations that have only one possible outcome, and where that outcome is determined by laws of nature. I.e. you are talking about situations that are constrained by necessity or fate.

    Actual free will [2] can only exist if reality is such that there is genuinely more than one possible outcome for a situation, and if reality is such that a person can make a genuine non-deterministic choice from these possible outcomes.

    1. “I think Locke and Hume nailed free will, and since then there has been no interesting debate about it”, Tim Maudlin, 1 November 2018, . (Hume on Free Will: … “freedom and moral responsibility can be reconciled with (causal) determinism”, . ) (Locke On Freedom: … “Compatibilism is the thesis that free will is compatible with causal determinism…the evidence strongly suggests that Locke would have embraced compatibilism”, .) (“Causal determinism is… the idea that every event is necessitated by antecedent events and conditions together with the laws of nature”, .)

    2. Free will: “The power of acting without the constraint of necessity or fate; the ability to act at one's own discretion”, .

  111. Tim Maudlin,

    Your “free will” is not free will, it is causal determinism. I would like to explain the difference between free will and causal determinism:

    Physical outcomes, for living things or particles, can (ideally) be represented by a set of variables and associated numbers.

    Causal determinism, i.e. where a thing can have no free will over outcomes, is represented by the following two situations: 1) For the next moment in time there is only one possible outcome, and laws of nature determine every numeric value for every variable representing the outcome; and 2) There seems to be more than one possible future outcome, but nevertheless at every point in time, laws of nature determine every numeric value for every variable representing an outcome.

    On the other hand, free will over outcomes is represented by the following situation: For the next moment in time, there is more than one possible numeric value for at least one of the variables representing the outcome; in what might look like a random choice to an observer of the situation, the thing with free will “chooses” one of those possible numeric values; there was no time, or mechanism, for the numeric value to be determined by laws of nature.

    “Laws of nature”, “possible outcomes”, “the next moment in time” and “choose” are just words representing something about the universe: they are not necessarily to be taken too literally. But to have free will over outcomes, it is at least necessary that the universe has quantum possibilities i.e. that for the next moment in time, there is more than one possible numeric value for at least one of the variables representing the outcome.

  112. 'Summary: If a philosopher starts speaking about elementary particles, run.' Even some Mathematicians. When people ask me about the Free Will Theorem I go oh no - not again. I cant fob them off by saying its rot because its a legit theorem. If anyone doesn't know it look it up. See if you can spot its flaw - hint - garbage in garbage out.

  113. 'John Bell's work that seems to finally validate "spooky action at a distance."'. What I am about to say is not true in some interpretations like DBB - but it is true in the majority. Please remember an entangled state of two particles is a single quantum state - you cant say that it can be considered as one particle and another particle. You can only talk of two independent particles once the entanglement is broken - but not before. Now realizing that have a look at the assumptions of both of Bells Theorems. In particular look at the concept of local causality. BTW evidently the two theorems are supposed to be equivalent. I have seen the proof of both but never of their equivalence. Probably missing something basic.

  114. Hydrogen is hydrogen. Not water. Oxygen is oxygen. Not water. We're known that the two will combine to make water.

    "how do elementary particles, which have no degree of freedom for consciousness in your criticism, assemble to form consciousness in human and animal brains?" As of now we do not know, but that is no reason to invoke oogitty boogy.

  115. What we fo know is that it is associated with structures we call neurons. i don't hear any claim that hair follicles or feet age involved

  116. And why pick spin as the analogue of consciousness? Why not charge our mass?

  117. Again. Why single out spin and disregard mass and charge? For that matter why the emphasis on electrons? What have you against protons, neutrons, pions etc?

    Add for quibbling about what spin is versus does, this sound much like Clinton's argument about the meaning of the word "is" itself.

  118. Hi, Sabine,
    I studied Mathematical Physics many years ago and am completely out-of-touch with modern attitudes. Yet your headline attacted me.

    I don't know exactly how humans think. Animals appear to think in a more primitive way. I, personaly don't believe that single-celled animals such as virus' and bacteria think. However, I do wonder how thinking managed to evolve from what are named as "sub-atomic particles" (and which I suspect are probably impossibly small smudges of energy).

    While I agree that sub-atomic 'particles' which we can't see, have no knowledge of how they use space, look, what colour they are, etc are extremely likely not to think.
    But then again, nor can single-celled Life such as virus' and bacteria.
    This Earth we live on has, however, a fantastic evolutionary process. So please tell me why it's impossible that the actual fabric of the universe should have

    1. anngie: A car seat will not take you to the store. Nor will a steering wheel, or an accelerator pedal. But a car in good repair with fuel can take you to the store, and bring you and your groceries back home. The car can do something no individual component of the car can do by itself.

      Your mistake is called the fallacy of decomposition; that what is true for the WHOLE must be true of its PARTS.

      The same thing is true of particles; a brain composed of (who knows how many) atoms can think, but just like the car, this is because the parts are arranged in a specific order that forms structures (neurons, chemicals, support elements) that interact and accomplish something.

      But it is a fallacy to think that just because the whole brain can think, that anything less than a few thousand neurons can think. Or to think that disconnected neurons can think. At some point, you just have inert cells that have no stimulus and aren't signaling each other.

      And certainly, there is no reason to believe atoms think, or electrons think. They can be part of a WHOLE that thinks, like a tire is part of a car, without being able to think themselves.

      The same fallacy applies to consciousness, feelings, qualia, and other mental phenomena.

      (Panpsychism is the notion that consciousness extends down to particles; another example of the fallacy of decomposition.)

      The inverse is the Fallacy of Composition; that what is true of the parts must be true of the whole. One example of that is if any single individual stands up in a theater, they will be able to see better. In fact several people could stand up, and see better. But if everybody stands up, not everybody will see better. What is true for a part is NOT true of the whole.

  119. This argument risks another fallacy for it is not proven anywhere that awareness is like thinking or that both be complex. While thinking is inherently complex since it equals some form/s of logical calculus/i, the same cannot safely be said of certain forms of awareness. Think of Zen as awareness in the absence of thinking. It could be Buddhas all the way down!

  120. "Now, if you want a particle to be conscious, your minimum expectation should be that the particle can change."

    Can you (change) ?
    Can you change into an ant or a Lion or simply somebody else ?
    Can change and stop eating ?

    1. Yes, I can change, witness that now I type an "A" and now I type a "B". That's not the same thing, hence change. An electron cannot do that. Before you submit further comments, please think about what I am saying.

  121. So, does Chalmer's view mean that consciousness is about human feelings? Those unpleasant hurts and joys that are stimulated by chemicals.
    Plants and animals also react to chemicals when suffering damage and also can transmit differing chemicals via fungi threads for different types of damage.

  122. Electrons and other such subatomic particles can't change and their behaviour is highly predictable. Yet, it has no definite position until it's subjected to an interaction (observation) that causes its waveform to collapse. I'd interpret that as a change. First, it's all over the place in the electron cloud and then it suddenly appears to have decided where to be, because of a slight nudge.

    What is an electron anyway?

  123. The electrons consciousness is in the hidden variables silly.
    Joking aside I don't even understand what electron consciousness would mean or why it would be meaningful. I cam here because a philosopher friend of mine was very angry about this post.

  124. Yes, it's clear that the simplistic version of panpsychism described here is a straw man. I admire many things about her thinking, but Hossenfelfer's willingness to attribute nonsensical claims to intelligent people is almost contemptuous. Is she not aware, for example, that John Conway has developed mathematical theorems with implications along these lines? Had she not heard of the free will theorem? That seems unlikely—I have to conclude she chooses to ignore evidence in this case. Evidence, that is, that not everyone who takes panpsychism seriously is a crank. And I don't fully understand the motivation—you don't have to cast aspersions on an idea that you don't find interesting or worth exploring. You can just ignore it.

  125. I am of course familiar with the free will theorem, which you would know if you'd be familiar with my work, which evidently you are not. Neither, for that matter, do you seem to understand the free will theorem, which -- as pretty much everyone who understands it agrees -- has absolutely nothing to do with free will. If you were trying to demonstrate your utter lack of comprehension, congratulations, you succeeded.