Seeking God in science part 10: The Mind-Body Problem

The title of this series, "Seeking God in Science", was meant to be a metaphor.  This should not come as a surprise.  We are (almost certainly) not going to find a literal supernatural being using the scientific method.  But — and I cannot emphasize this enough — this is not because the scientific method is prejudiced against such an outcome.  This is a charge often leveled against science by religious apologists, that science assumes naturalism and so it is structurally incapable of discovering extra-natural truths.  This is absolutely false, and anyone who says this either does not understand the scientific method or they are deliberately misrepresenting the facts (a.k.a. lying).  The scientific method is simply to find the best explanation that accounts for all observations.  If that explanation turns out to be God, the method has no objections.

The fact that God turns out to be unnecessary to explain any observations is far from obvious.  For most of human history the very idea that God was unnecessary would have gotten you laughed out of the room (if not burned at the stake).  Even today the vast majority of people believe that there is some kind of supernatural force of entity at work in the world, either now or in the past, though the sphere of influence of this entity is often much smaller than it was in the past.  Many observations that were once attributed to gods are now widely accepted as natural phenomena: the behavior of stars and planets, weather, earthquakes, and so on.

But there are some observations that remain stubbornly mysterious, the so-called Big Questions.  How did we get here?  Why are we here?  And, possibly the most mysterious of all, what exactly are we?

My original plan for this series was to try to sneak up on these Big Questions step-by-step, starting with intuitively obvious things like the classical mechanics of rigid bodies (with chairs as my canonical example), moving from there to information theory, thermodynamics, and the theory of computation, and then, finally, quantum mechanics for the big finish.  But based on the feedback I've gotten so far I've decided that approach is probably not going to work.  The attention span of a modern audience, especially a young audience, isn't long enough to keep their interest long enough to get through all that.  Rather than throw in the towel I'm going to try something different: a full-frontal assault on one of the Big Questions without having laid all of the necessary groundwork beforehand.  This argument will necessarily have some holes in it, and we won't get all the way to the Answer.  But I hope that it will show that the problem is not beyond the scope of scientific inquiry.

The Biq Question I'm going to tackle here is the problem of consciousness.  What is it?  How does it work?  Can it be explained naturalistically, or is our conscious experience evidence of the supernatural or the divine?  Two spoiler alerts: first, no, consciousness is (almost certainly) not evidence of the supernatural.  And second, if you want to get closer to The Answer than I am going to get here, go read Daniel Dennett's book, "Consciousness Explained."  A word of warning, though: this is an academic work, and not particularly accessible.  One of my goals in this series was to present a more easily accessible account of Dennett's work.

So, to the task.  As always when we apply the scientific method we must start with a Problem, an observation that cannot be accounted for by our current best explanation, with the understanding that all observations are ultimately individual subjective experiences.  The existence of objective reality is not a given.  It is a hypothesis, the Objective Reality Hypothesis, that I have accepted to explain my observations of things like chairs and other humans.  Most other humans that I've met seem to accept the Objective Reality Hypothesis.  At least, they act as if they do.  All of my observations are consistent with the Meta-Objective-Reality-Hypothesis, namely, that most people accept that chairs and other people are real things external to themselves.  I don't need anything extra-physical like "chairness" to explain anything I observe about chairs, nor to explain anything I observe in other people's behavior in their interactions with chairs.  I have never met anyone who denied the objective existence of chairs.  If I were to ever meet such a person I would probably explain their behavior as trolling or mental illness.

Consciousness is different.  Consciousness is something I experience, and the behavior I observe in other people is consistent with the hypothesis that they experience it too.  But consciousness is very different from chairs.  I can see and touch chairs.  So can other people.  So can machines.  So can cats.  Moreover, when I see and touch a chair, someone else can see and touch the same chair.  We can agree on its properties, its location, its size and shape, the material it's made of, what color it is, how much it weighs.

None of these things are true of consciousness.  I can neither see nor touch consciousness, neither mine nor anyone else's.  The only reason I have to believe that consciousness exists at all is my direct subjective experience of it, and (my direct subjective experience of) the behavior I observe in others, and specifically, other people's testimony of experiencing consciousness, which seems to match my own personal subjective experience.

So one possibility that needs to be considered is that consciousness is an illusion.  There are subjective experiences that really don't correspond to anything objectively real.  I have to stress the distinction between an illusion and a delusion.  A delusion is a subjective experience that is not shared with others.  It may or may not correspond to something objectively real.  The voices that a schizophrenic hears might be coming from a supernatural realm, or they might just be the neurons in their brain misfiring and injecting signals directly into their auditory system, but either way they are literally hearing voices.  It's just that those voices aren't produced by vibrating air molecules.

Illusions are different.  They are common across most humans.  When most humans look at the rotating-snakes illusion, they (we) literally see motion even though the image is actually static.  The neurons in our brains that normally detect motion are being stimulated, just not by things actually moving in objective reality.  Or consider watching a movie, or being in a virtual-reality simulation, or even just looking at a photograph.  You are seeing things that aren't really there.

Of course, there is still something there — the screen.  It is just that the nature of the thing you perceive is different from the nature of the thing that is really there.  You — along with most other humans — perceive motion even though in objective reality there is none.

So the idea that consciousness is an illusion is not falsified by either the fact that you perceive it, nor the fact that everyone else seems to perceive it.

There is still a Problem, though: in order to have a subjective experience of anything at all there has to be something that is having the experience.  Consciousness is not just any old subjective experience, it is the subjective experience.  The very idea of "subjective experience" is inextricably entangled with consciousness.  It seems fundamentally impossible for a non-conscious entity to have a subjective experience.

And yet a non-conscious entity can act as if it is having a subjective experience.  Nowadays we can build an anthropomorphic robot, show it a chair, ask "What do you see?", and be completely unsurprised when the robot replies "I see a chair."  We can easily program the robot to respond "yes" when asked "Are you conscious?  Are you aware of your own existence?"

Let us stipulate that such a robot is not actually conscious, at least not necessarily.  It is at least possible (in fact, it seems likely, at least to me) that such a robot would not in fact be conscious, not actually have subjective experiences, notwithstanding its replies to inquiries.  It is possible that such a robot would be a philosophical zombie, an entity whose behavior is indistinguishable from a conscious entity but which is not in fact a conscious entity.

Is the reverse possible?  Is it possible to build a robot that is actually conscious, or is there something about being a biological entity that is required for consciousness?  Of course we don't yet know the answer to that, but we can observe that we can replace a lot of our biological parts with non-biological alternatives without threatening our claim to being conscious.  We can have artificial limbs, artificial hearts, at least partially artificial ears and eyes, and all that just with today's technology.  The only thing that is not clear is whether an artificial brain can be conscious, and, if so, how could we distinguish an artificial brain that is actually conscious from one that just acts as if it is.

We can also ask the converse: is it possible to have a conscious entity that does not act as if it is?  The answer there is almost certainly yes: there is a rare and horrific medical condition called "locked-in syndrome" which renders a person unable to move (except for their eyes) but they remain (as far as we can tell) fully conscious.  It is not hard to imagine a similar situation where even eye movements are affected.

In a situation like that, where a person's behavior was literally indistinguishable from someone under general anesthesia, how could we possibly know if they were conscious?  Well, one way is if they somehow recovered from their condition and were able to report their experience.  But note that for this to be possible they would have to remember the experience.  If you think about it, this is the only possible way we could distinguish between someone being fully immobile but conscious, and being under general anesthetic.  One of the distinguishing features of general anesthetic is that you can't remember what happened while you were under.  How can we be sure that a person under general anesthetic isn't actually conscious, but just loses the capacity to form memories?

I submit that this rhetorical question is actually a key to understanding consciousness.  I can't think of any way to make this argument rigorous because consciousness is such a slippery eel, but I submit that being able to form memories is essential to consciousness, a necessary (though obviously not sufficient) condition.  I'll make this argument in anther way: when I speak of my subjective experiences, what is the referent of the word "my"?  What is the thing having the experience?  Well, it's me, and a big part of what makes me "me" is my history.  The person I am is in some sense defined by my knowledge and experiences, which is to say, by things I remember.  Even if I suffer from amnesia (or simply old age), if I can still function as a human being, if I can talk or even just brush my teeth, I have to be remembering something in order to perform those tasks.  I can't conceive of any way that it could possibly make sense for a thing to be conscious without remembering anything.

As I mentioned before, memory might be necessary, but it alone is clearly not sufficient for consciousness.  Many non-conscious things have memory, with computers being the obvious example.  But human memory has an interesting feature which can be demonstrated by psychophysics experiments: it is malleable.  People can have the subjective experience of remembering things that demonstrably did not happen, that could not have happened.  This plays out on time scales both large and small.  On large time scales it is possible to implant false memories through suggestion.  On small time scales it is possible to show that your conscious awareness of your surroundings lags reality by a good fraction of a second, which is the time scale your neurons operate on, and that what you consciously perceive does not always correspond to objective reality.

So maybe your subjective experience of consciousness itself does not correspond to objective reality.  Maybe consciousness itself is an illusion.  That would explain why we can't measure it objectively, but on the other hand it raises another Problem.  When I say, "I have a subjective experience of seeing a chair" I mean more than the mere fact that I can see the chair, and that my behavior reflects the fact that I can see the chair.  I can build a robot that can see chairs and behave accordingly, but that robot won't (necessarily) have a subjective experience of seeing the chair.  The subjective experience is more than the mere seeing of the chair, it is being aware of seeing the chair.  But it is even more than that.  It is being aware of being aware of seeing the chair, and being aware of being aware of being aware of seeing the chair, and so on in a seemingly infinite loop.  A robot can be aware of seeing a chair (a better example would be that a self-driving car can be aware of objects around it like pedestrians and other cars).  But a robot (or a self-driving car) is not (necessarily) conscious.  The difference (or at least one of the differences) between awareness and consciousness is this recursive self-application of awareness to itself.

This observation would seem to eliminate the possibility that consciousness is an illusion.  If the subjective experience of consciousness is real (and it is, at least for me) then there has to be something real backing up that experience, some "substrate" for the emergent experience to emerge from, some base case for the recursion.  An old friend of mine in my JPL days, Joe Provenzano, put it like this: if consciousness is an illusion, who is being illused?  An illusion is a disconnect from objective reality, but there still has to be something at the other end of that disconnect.

That "something" is usually called a "mind", and the question of how the mind relates to the brain is called the mind-body problem.  There are two possibilities: the first is that minds can be explained entirely in terms of the material properties of people's brains in a manner analogous to how chairs can be explained entirely in terms of the material properties of the atoms that comprise them.  The weird and seemingly magical properties of minds are a consequence merely of the vastly greater complexity of brains compared to chairs.

The other possibility is that minds require something immaterial, something outside of the physical properties of the brain, to explain them.  I'm going to follow convention and call that a "soul", though I want to be clear that I don't intend for that word to carry any of the usual theological baggage associated with it.  A soul in this case may or may not be supernatural, may or may not have anything to do with any deity, may or may not be a natural phenomenon.  The only stipulation is that they exist outside of people's brains, including the possibility that they could exist outside of space and time itself.

I am, of course, ultimately going to argue in favor of material minds and against souls, though, again of course, I am only going to be able to make the tiniest dent in that argument here.  I'm going to just point out a few things that I hope will sow some seeds of doubt in you if you are currently an advocate of the soul hypothesis.

First, there are clear correlations between brain states and mental states, including consciousness.  The most obvious is that your conscious experience is strongly bound to the physical location of your body and hence to the physical location of your brain.  If your soul exists, its conscious perspective on the world is almost always the perspective of your physical body.  What is "here" and "now" to your consciousness is also "here" and "now" to your body and brain.  There are anecdotal reports of "out of body experiences" but there is no evidence that these are actually real in the sense that the experience corresponds to objective reality.  (It would be pretty easy to do an experiment to test this.  Most out-of-body experiences seem to happen in hospitals.  So take a piece of cardboard, write a random number on it, and mount it on the ceiling so that it can be seen from above but not from below.  If anyone claiming to have an out-of-body experience can tell you what that number is, that would be strong evidence that the experience was actually real.)

Second, using modern technology we can see, at least in broad brushstrokes, what kinds of brain activity correspond with being conscious as opposed to being asleep or under general anesthesia or in a coma.  Third, we can alter our subjective experience of consciousness with various chemicals (with general anesthesia being an extreme example).

So if souls are real, they have to interact with the brain somehow.  If nothing else, there has to be some mechanism by which the material things that influence consciousness exert that influence on the soul.  And that has to run both ways.  The soul has to also be able to influence the brain somehow.  Somewhere between the nerve impulses going in to the brain and the nerve impulses coming back out, there has to be something that happens differently than it would if souls did not exist.  If everything the brain does is exactly the same with and without souls, then, ipso facto, souls are not necessary to explain any observations.  That is the scientific definition of a thing that doesn't exist.  There might be technological limitations to our ever observing those differences, but there has to be something in the brain that could potentially be observed in principle that required souls to explain.  Otherwise, material minds would suffice.

There are two other potential explanations for consciousness besides souls and minds which I am going to list here just for completeness.  The first possibility is that I am the only truly conscious entity in the universe.  Despite the behavior of other humans being qualitatively indistinguishable from mine, they nonetheless do not actually have the same subjective experience as I do.  In other words, my consciousness is real, but yours isn't.  Your outward signs of consciousness are fake, a simulation of consciousness, not the real thing.  You are a philosophical zombie (PZ), a robot, an entity that behaves as if it is conscious but isn't really.  A closely related explanation is solipsism, which says that I am the only thing that actually exists and everything else that I perceive is just a figment of my imagination.

The problem with both of these is pretty obvious: they only work for me.  Of course, you can adopt variations on these hypotheses which say that your consciousness is the only legitimately real one, or that everyone (including me) is a figment of your imagination.  But you cannot consciously adopt the hypothesis that my consciousness is the only real one.  The problem with these hypotheses is the same as the problem with last-thursdayism: these are not individual hypotheses, they are a vast family of hypotheses (over 8 billion of them), all of which are consistent with the objective data, but only one of which can possibly be correct.  Solipsism and PZs cannot be falsified by data, but they can be eliminated as good explanations because every individual instance of these hypotheses requires special pleading in order to lay claim to being the best explanation.

I'm going to throw out one last Problem for the soul hypothesis: when do souls come into being, and how do they ultimately get bound to their corresponding brains and bodies?  A common claim among Christians is that souls are created at conception, but there are two problems with this.  First, at conception, the fertilized egg doesn't yet have a brain, or even anything that can be identified as the precursor of a brain.  And second, there is the phenomenon of identical twins, which are produced when a fertilized egg splits physically after its first cell division and produces two individuals with identical genes.  Do identical twins share a single soul?  If so, can one of a pair of identical twins be saved and the other not?  If not, where does the second soul come from?

My point in this article was not to solve the mind-body problem.  Mankind has been arguing over that for millennia and there is obviously no way I am going to cut that Gordian knot in a few thousand words.  What I wanted to do here is simply to demonstrate how even such a deep philosophical question can be subjected to scientific inquiry, and how the structure of that inquiry shares some similarity with the scientific inquiry into the nature of chairs and other seemingly mundane phenomena.

BTW, Grant Sanderson, the brilliant mind behind 3Blue1Brown, just released the second chapter in an excellent series on information theory and large language models.  I encourage you to watch it if you want to get a deeper understanding of information theory.  As some commenters have pointed out, the definition of information I have been using here is not the only one there is.  There are lots of different definitions, and a deep connection between information and entropy, which I have been trying very hard to avoid getting into.  Watching Grant's explanations will make it clearer why I have been trying so hard to avoid these details.  This series is about philosophy, and it's hard enough to write about without getting tangled in the technical weeds.  But there are plenty of resources out there if you want to get into the details.