Monday, November 22, 2021

Carroll on Consciousness

Sean M. Carroll is very good at explaining textbook physics, but when he discusses his own beliefs, he has some wacky ideas. He believes in determinism, and many-worlds theory.

(Yes, many-worlds is not deterministic, but that is not my point here.)

Here he debates panpsychism:

Theoretical physicist Sean Carroll joins us to discuss whether it make sense to think of consciousness as an emergent phenomenon, and whether contemporary physics points in this direction.
I see 3 possibilities.

1. There is no such thing as consciousness. Yes, we perceive all sorts of things, and act on those perceptions, but that's all.

2. Consciousness is real, and has a physical basis that may eventually be understood in terms of fundamental physics, chemistry, and biology.

3. Consciousness is in the mind or soul, and not the body, and it best understood in spiritual terms.

As a scientific reductionist, I lean towards (2), but the others are possible, especially since we don't even have a good definition of consciousness.

If (2) and scientific reductionism are true, and humans are composed of 1030 or so quarks and electrons, then it seems plausible that each quark and electron has a little bit of consciousness.

Carroll's answer to this is that the behavior of electrons is completely determined by physical law, and so very strange changes to those laws would be needed to explain partially conscious electrons.

But our best laws of physics are not deterministic. Not in this universe, anyway. Those elections could be partially conscious without any change to known laws.

Carroll has elaborated on his own blog:

The idea was not to explain how consciousness actually works — I don’t really have any good ideas about that. It was to emphasize a dilemma that faces anyone who is not a physicalist, someone who doesn’t accept the view of consciousness as a weakly-emergent way of talking about higher-level phenomena.

The dilemma flows from the following fact: the laws of physics underlying everyday life are completely known. They even have a name, the “Core Theory.” We don’t have a theory of everything, but what we do have is a theory that works really well in a certain restricted domain, and that domain is large enough to include everything that happens in our everyday lives, including inside ourselves. ...

That’s not to say we are certain the Core Theory is correct, even in its supposed domain of applicability.

He then launches into a discussion of zombies who appear to be just like conscious humans, but are not.

I don't see how this proves anything. He cannot define consciousness, and when he takes it away, peeople behave just the same. No. If conscious means anything, it means that people would behave differently if they didn't have it.

Carroll calls his viewpoint physicalism, but it is really the opposite, as he refuses to accept a physical basis for consciousness.

I get why non-physicalists about consciousness are reluctant to propose explicit ways in which the dynamics of the Core Theory might be violated. Physics is really strong, very well-understood, and backed by enormous piles of experimental data. It’s hard to mess with that.
He is assuming that a theory of consciousness would violate the Core Theory, but I doubt it.

Dr. Bee explains why particles decay, and adds a consciousness argument:

the tau can decay in many different ways. Instead of decaying into an electron, a tau-neutrino and an electron anti-neutrino, it could for example decay into a muon, a tau-neutrino and a muon anti-neutrino. Or it could decay into a tau-neutrino and a pion. The pion is made up of two quarks. Or it could decay into a tau-neutrino and a rho. The rho is also made up of two quarks, but different ones than the pion. And there are many other possible decay channels for the tau. ...

The taus are exactly identical. We know this because if they weren’t, they’d themselves be produced in larger numbers in particle collisions than we observe. The idea that there are different versions of taus is therefore just incompatible with observation.

This, by the way, is also why elementary particles can’t be conscious. It’s because we know they do not have internal states. Elementary particles are called elementary because they are simple. The only way you can assign any additional property to them, call that property “consciousness” or whatever you like, is to make that property entirely featureless and unobservable. This is why panpsychism which assigns consciousness to everything, including elementary particles, is either bluntly wrong – that’s if the consciousness of elementary particles is actually observable, because, well, we don’t observe it – or entirely useless – because if that thing you call consciousness isn’t observable it doesn’t explain anything.

So you could have two identical tau particles, and one decays into an electron and 2 neutrinos, and the other decays into a muon and 2 neutrinos.

And we have a Core Theory that explains the dynamics of everything that happens!

No, this is untenable. I see a couple of possibilities.

1. Those taus are not really identical. They have internal states that determine how they will decay.

2. The taus are identical, but they have some sort of conscious free will that allow them to choose how and when they decay.

Some physicists would say that the taus are identical and intrinsically random. But saying that is just a way of saying that we don't know whether it is possibility (1) or (2).

Dr. Bee gives an argument for the taus being identical. But we can never be sure that they are truly identical. Maybe they just appear identical in a particular quantum field theory, but that ignores a deeper reality.

I know it seems crazy to say that a tau particle has a little bit of conscious free will. But the alternatives are stranger.

Now where does human consciousness come from? Carroll says that it cannot come from anything in the Core Theory, because it is dynamically complete and there is no room for any panpsychism.

There is room. Quantum mechanics is not deterministic. It predicts probabilities because there are mysterious causal factors that it cannot account for.

Jerry Coyne says that Carroll decisively refutes panpsychism. I disagree. I say Carroll has the worse argument.

After writing this, I am surprised to see Lubos Motl take the side of panpsychism.

But even before QM, it was rather clear that panpsychism was needed in any scientific world view simply because there can't be any "metaphysically sharp" boundary between objects like humans that we consider conscious; and other objects. So some amount of the "consciousness substance" must be assigned to any object in Nature, otherwise we end up with a clearly scientifically ludicrous anthropocentric or anthropomorphic theory. ...

OK, Carroll hasn't noticed that the current "Core Theory" is actually quantum mechanical and therefore needs conscious observers to be applied. Much of his article is a circular reasoning ...

For the 9,877th time, he can only be a "physicalist" because he doesn't do science. If he were doing science, he would be abandoning theories that conflict with the observations. And because all classical i.e. P-world theories conflict with the observations, they are dead....

Carroll behaves exactly like you expect from a zombie: Sean Carroll is a simulation of a generic zombie

This isn't fair because Carroll's Core Theory is not classical mechanics. Carroll would say that he is very much a believer in quantum mechanics.

But Carroll doesn't really believe in textbook quantum mechanics. He believes in many-worlds theory, where there is no wave function collapse, no probabilities, no predicted events, no free will, and no correspondence with any scientific experiments.

Yes, I do think that many-worlds theory is fundamentally incompatible with science. It is owrse than believing in astrology or witchcraft.

Nautilus has more on the pros and cons of panpsychism.


  1. GIGO from so many overpaid fat heads with nothing intelligent to say. Embarrassing really, having to choose between inanimate meat puppets on one side and secret magical soul sauce ingredients on the other. Pray tell, what the heck does anyone even mean by 'consciousness'?

    Would be kind of nice if anyone decided to specify exactly what they were talking about... which might also reveal how little they know about what they are talking about.

    Consciousness as an awareness of the immediate self and surroundings like many animals, or more abstract levels of awareness involving half-assed pretzel-knot reasoning that really don't do anything useful ...except get silly papers published in academic journals?

    As with many things, there is more than on or off, 0 or 1. There is a matter of degree in a spectrum of consciousness that might be considered almost mechanical on one end (such as with insects or most woke college activists), and more involved, devious and unpredictable on the other end (such as the average little old lady trying to cheat at bingo).

  2. Dear Roger,

    The primary evidence of consciousness is via introspection, by directly observing (i.e., by directly perceiving) that one is conscious --- that one is aware of reality including of one's own mental states; and then reaching the idea that one has a faculty of being conscious, of perceiving reality (i.e. of the fact that one indeed has a self, with a free-will).

    The term "consciousness" refers to both a state of being conscious (being aware of reality including of one's own mental states), as also a faculty of being conscious.

    Observations of one's own consciousness (in both senses of the term) is the primary evidence for consciousness.

    Each conscious being is *directly* aware of his own consciousness. Leaving aside claims of telepathy and all that, we normally do not have a direct experience of someone else' consciousness. But the point is, a direct experience like that is not at all necessary. We can also infer, from "external" observations, that other people are conscious, that they are aware, that they have the faculty of consciousness.

    Further, we can also infer, from observations, that people [and even animals or living beings] are conscious only when they are living. Dead bodies feel no joy or pain. [That's why (at least in India) we even cremate them.]

    So, the very blatantly obvious conclusion is this: *Life* is necessary for consciousness.

    Now, consider, e.g., your statement:

    >> "If (2) and scientific reductionism are true, and humans are composed of 1030 or so quarks and electrons, then it seems plausible that each quark and electron has a little bit of consciousness. "

    If you replace "consciousness" with "life", would it continue to look plausible? After all, if an electron has to be conscious, it has to be living, no? ... So, does it look plausible to say that we always live in a universe that is *entirely* made of "animals" (in the sense: animate beings)? That not just people and animals and insects and trees and fungi and amobae and viri (when in living action), but also tables and chairs and buildings and roads and rocks are, in a way, "animals"? At least in some elementary or rudimentary sense of the term "life"?

    And if panvitalism (just to coin a term) doesn't seem plausible, then why does panpsychism?


    PS: If things continue in the same direction, it won't be long before they introduce topics on Quantum Computing in "Psychology 101". ... Also, PhD topics: "Survey of psychological difficulties faced by ions in the trapped-ion quantum computer." Another PhD topic: "Psychological difficulties of trapped ion and optical qubits --- a comparative study." Etc. [Also biological difficulties, really speaking... Also medical remedies...] QC enthusiasts would love it too. "Potential applications of the quantum computer include biological, psychological, and medical fields". You get the idea...