Friday, May 31, 2019

The Dualist's Quadrilemma -- and All of Ours?

Old-school substance dualists hold that people not only have material or physical bodies but also immaterial souls, and that the soul rather than the body is the locus and origin of conscious experience (qualia, "something-it's-like"-ness, phenomenality). Two great advantages of this view are (1.) that it avoids the puzzle of having to explain how dumb, bumping matter can give rise to something as seemingly ontologically radically different as consciousness (at least many people find this puzzling), (2.) that it promises hope of an afterlife, since possibly the soul could continue to exist after the body has died.

However, substance dualism faces a scope of ensoulment problem, or what I'll call the dualist's quadrilemma. The more I think about this quadrilemma, however, the more I think some version of it might trouble virtually all theories of consciousness.

Who has a soul? I see four possible answers, each of which is problematic:

(1.) Only human beings have souls. (At least on Earth. Let's bracket Martians and angels.) At the moment of conception or at the moment of birth, God or nature gives us a soul, but no dog or chimpanzee or raven has a soul. From this it follows, since souls are the locus of conscious experience, that dogs and chimpanzees and ravens have no conscious experiences -- no emotional experiences, no experiences of pain or hunger, no visual or olfactory experiences. There is nothing it's like to be a dog or chimpanzee or raven -- they are, so to speak, entirely experientially blank, as blank as we normally assume a toy robot to be. They emit behavior similar to the behavior we emit when we experience pain or hunger, and they have nervous systems that closely resemble ours, but that is misleading. They lack the soul-stuff that turns on the lights.

This view is difficult to accept, both on commonsensical and on scientific grounds. Ordinarily, we think that dogs, chimpanzees, and ravens do have experiences, even if their experiences are not as cognitively complex as ours. And scientifically, this view seems to overestimate the gulf between us and our nearest biological relatives -- and furthermore seems to require that there was some discrete moment in our evolutionary history when we changed from unensouled to ensouled creatures (Australopithecus anamensis? Homo habilus?), despite, presumably, no radical saltation in our physiology.

(2.) Everything has a soul! Maybe we all are subparts of a single, grand, universe-sized soul; or maybe there are many, many, tiny souls for tiny objects such as electrons.

Although panpsychist views of this sort have received increasing attention in the philosophy and psychology of consciousness recently, most people in our culture appear to find panpsychism too bizarre to accept. My own view is that one of the main pressures in favor of panpsychism is the seeming unpalatability of the other three horns of this quadrilemma.

(3.) There's a line in the sand. Somewhere between electrons and humans, there's a sharp line between the ensouled and the unensouled creatures. Maybe mammals have souls but no other animal does. Or maybe toads have souls but (cognitively simpler) pond frogs don't.

The problem with this view is that physiology and cognitive sophistication comes in degrees, with no sharp dividing line among the species. If having a soul matters, then there ought to be some radical difference between the souled and unensouled creatures -- at least in their cognition, and probably also in their physiology. But the only place it seems at all plausible to draw a sharp line is between human beings and all the rest -- which puts us back on Horn 1.

[illustration of one possible theory of ensoulment; image source]

(4.) Having a soul isn't a yes/no thing but rather a matter of degree. Some creatures are half-ensouled or 6% ensouled.

The problem with this view is that is requires an entirely novel metaphysics that is difficult to envision. What would it be to be kind of ensouled? Some properties lend themselves to in-between, indeterminate cases: a color might be on the vague boundary between blue and not-quite-blue, a person might be in the vague region between being an extravert and being not quite an extravert. We can imagine how such cases go and build a metaphysics of colors and personality traits to accommodate in-between cases and matters of degree. But souls seem like the kinds of things that one either has or doesn't have, with no in-between cases. I am not aware of any philosopher who has attempted to construct a metaphysics of half-souls, and it's hard to see how this would go.

Now you might say so much the worse for substance dualism! But I think non-dualists face the same quadrilemma, even if not quite as vividly.

What kinds of creatures are conscious? If we don't want to say "only humans" and we don't want to say "everything", then we need either a bright line somewhere or we need a concept of in-between consciousness. A bright line seems implausible given the continuity of cognitive capacities and physiology across living species, in the course of fetal development, and in the course of evolution. So are we (even non-dualists) then pushed into conceptualizing consciousness as the kind of thing that one can kind of have, or half have? I, at least, find this difficult to conceive, and I know of no good attempts to make theoretical sense of the idea. On the face of it, a stream of conscious experience appears to be something you either have (however small or snail-like) or fail to have -- like a soul. There's either a center of subjectivity where experiences arise, or there isn't.

So maybe I need to reconcile myself to one of the other three horns? But they're all so unattractive!


Marcus Arvan said...

Why isn't higher-order cognition a plausible dividing-line for (3)? Only some creatures (humans, some primates, some birds, etc.) have higher-order capacities necessary for representing *themselves* representing things. Given that conscious experience intuitively seems to be just that (experiencing that one is experiencing things), this seems to me a plausible way to defend 3 (though I'm inclined to panpsychism myself).

Anonymous said...

Every aspect of the dualist's qaudrilemma is problematic. Whenever entertaining metaphysical speculations, I've established four criteria which have to be met in order to consider any postulate viable.
1. Logical consistency: It has to make sense, which means that any theory has to be able to stand up under the scrutiny of analysis.
2. Ease of explanation: The postulate has to be simple, simple enough that a six year old could apprehend it.
3. Inclusiveness: The hypothesis has to include every phenomenon though out all of time without any exceptions.
4. Anti-Anthropocentrism: The postulate does not revolve around homo sapiens, in other words, it is not about us.

Eric Schwitzgebel said...

Thanks for the comments, folks!

Marcus: I higher-order theories have some plausibility as a basis for sorting the creatures with consciousness from those without consciousness. However, that doesn't make the bright-line problem go away. In our messy, gradual world, it's unlikely that there's a sharp saltation from creatures with to those without higher-order cognition (though of course to defend this in detail would require addressing the details of particular HO theories). I'm inclined to think that the problem is aggravated by thinking about babies, who very plausibly are conscious but who also have higher-order cognition, if at all, in the implicit and probably graded manner that non-human animals do.

Lee: 1 seems plausible. But why hold that the world is as simple as required by 2? My Stanford-school philosophy of science rebels against 3, in this complex world that defies simple modeling. 4 seems true of some phenomena but not of others.

howard b said...

Many religious folks accept the reality of eternal mysteries. Why can't we accept the inconvenient existence of unsolvable contradictions at the heart of our world?

howard b said...

Your pan psychism sounds compatible with Lucretius in a strange way How is pan psychism different than animism, presuming the world to be alive?

William said...

An alternative to the "bright line" and the "in-between" of consciousness is simply to admit different kinds of consciousness. Perhaps a horse or a bat is conscious, but in a qualitatively different way (no language of thought perhaps) than a human is. This becomes even more likely with lifeforms that have very different brains from ours, such as squid.

This might fit Aristotle's idea that nonhuman animals and humans have different kinds of souls.

SelfAwarePatterns said...

Interesting post Eric, as always.

I think many of us, when discussing consciousness, whether intentionally or not, are talking about a soul. Even among professed non-dualists, remnant Cartesian dualism heavily influences the discussion. And most people who've given up dualism haven't come to terms with all the implications.

One of those implications is the 4th horn you describe. It's actually not that hard to accept if you read about neurological case studies of brain injured patients.

Phineas Gage lost a lot of his self control due to massive frontal lobe loss. Henry Molaison lost the ability to form new memories, at least without repetition. People can lose the ability to perceive colors, the left side of the world, to perceive motion, faces, or even pain. They can lose the ability to understand language or a sense of self. No aspect of the mind seems to be immune.

Often when I lay this out, someone says I'm not talking about *experience*. My question for anyone who says that is, without using synonyms of the word "experience" such as "quale", what precisely are you talking about that can't be lost due to a brain lesion in the wrong spot? (Aside from a lesion in the midbrain or thalamus region knocking the whole thing out.)

I think the idea of an irreducible psyche is a holdover from classic dualism. If you accept that the mind is what the brain does, and that there are many states of partial functionality the brain can be in, then the idea of a partial consciousness shouldn't be a large leap.

Anonymous said...

To avoid misunderstanding, I should have qualified that my four criteria are required for a universal theory, a theory of everything. The "hard problem" of both causation and consciousness are intrinsically linked and are at the center of the mystery. It appears, that mankind has no choice but to accept the inconvenience of the unsolvable contradictions and the heart of our world. It would appear that way, and skeptics by into that paradigm without question as do most people. But I was no skeptic; so I soldiered on with my research for over forty years until my work suddenly coalesced into an unexpected conclusion.

Just to be clear, according to my model, a model which satisfies all four of the criteria previously outlined, the hard problem of both causation and consciousness is resolved. At its core, the mystery shrouding causation and consciousness is fundamentally simple, so simple, that it will literally blow one's mind. The entire equation can be written on the front of a t-shirt. There is only one latent paradox built into my model: Homo sapiens will not be happy with the theory because the theory does impart to us any more control.

dave said...

The bright dividing line between species does seem problematic, given that physical/neural complexity is continuous all the way down the phylogenetic tree. Any reason we give for drawing it between one species and another would be subject to fuzzy edge case issues where the capacities of both blur.

But maybe thinking in terms of the complexity of species is the wrong way to go. Humans are highly complex, but when we're in a dreamless sleep, or knocked unconscious, our highly complex brains are not instantiating any phenomenal properties.

This makes a bright dividing line feel a bit less problematic to me. When I consider my own experience of falling asleep, I can sort of understand myself as crossing this line every night. One moment I'm conscious, the next moment I'm not. In the morning something "kicks back on" in my brain, instantiating phenomenal consciousness and bringing me back across the line.

Wherever the physical state of affairs that cause my brain to "kick back on" are present elsewhere in the world is perhaps where we should draw the line. This makes thinking about species a little easier. Perhaps there are species who are perpetually straddling the line, moving back and forth from one side to the other. Maybe there is a frog or a dragonfly somewhere whose complexity is such that at one moment there is something it is like to be them, but the very next moment not. This feels extremely counterintuitive, but should it be any less intuitive than falling into a dreamless sleep every night and then waking up?

James of Seattle said...

If consciousness is about certain kinds of processes, then each of these categories can be explained. If you can define the requirements for a conscious-type process, then you have created the line in the sand. A system is conscious if it can perform a conscious-type process, and only such systems are conscious. If you define the process narrowly such that only humans have that capability, then you have made humans uniquely special. If you define the process broadly (say, ability to interact with the environment), then you get panpsychism.

But it is unlikely that you will define the process such that different systems can perform that process to varying degrees. Instead, some systems will be capable of more variations of that process than others. Some systems may have a repertoire of a single such process. It wouldn’t make sense to say such systems have some fraction of consciousness. It would be better just to say they have a simpler consciousness.


Anonymous said...


I am quickly reaching the end of my short career as a commentator on internet forums, so I am going to close that career out with a final comment. As an older gentleman, I've lived long enough to offer constructive criticism. Skepticism is a badge of honor worn by its recipients with pride and honor is a gift that a man gives to himself. It takes courage to follow a bolder course and keep one's heart open to possibility. You already know how a feel about rationality and its limitations as a discrete binary system. It is because of the limitations of rationality that you have chosen to be a skeptic.

Whenever one is convinced by a rational argument, one does not know more but one knows less, because a conviction closes the door to any other possibility. Intellectual honesty and spiritualism make the perfect union. Make no mistake, spiritualism is not religion, spiritualism is courage. Intellectual honesty keeps the door open to possibility and spiritualism gives one the courage to walk through that door.

Good luck Eric, you've been an honorable host.
Lee Roetcisoender

Callan said...

I guess a question is does consciousness always seem complete to itself?

Would a half consciousness actually use itself as the ruler for measuring what is a whole consciousness and find itself a complete one?

So is it hard to imagine a half consciousness (or a one and a half consciousness) because when one reflects on oneself, one uses oneself as both the subject and the ruler to measure the subject by?

'Does my consciousness look big in this?'

Philosopher Eric said...

Cheers Lee. The scourge of addiction is one of the things that you commonly discuss, and I do realize that I’m addicted. I’ll stop blogging when they pry the keyboard from my cold dead fingers! In order to not relapse though, you’ll need a hobby which provides similar intellectual stimulation. And if you end up finding that nothing sufficiently fill the void, well even some alcoholics are considered “functional”. The highs and lows may be unsettling, though you might nevertheless learn to get by. At least that’s my plan.

Philosopher Eric said...

Agreed professor, it’s not just the dualist that has a problem here, but the naturalist as well. I presume that the vast majority of the people who read your writings are naturalists, and yet who among our tribe is comfortable enough to claim, “Unlike all else in the universe, here is the causally produced stuff by which conscious mind exists….”?

(….Crickets…. )

Well I do propose something in that regard. Evolution obviously created central organism processors (or “brains”). It may be that they caused the Cambrian explosion of life 541 million years ago. Regardless, at least initially these mechanisms should have been similar to the computers by which our robots are instructed, or non phenomenal machines. And why did such life need more? I suspect because agency was required in order to effectively deal with more “open” environments. Regardless, apparently these non-conscious brains began to produce a punishment/ reward dynamic from which to drive a conscious form of function, or sentience. Just as electricity drives the function of our computers, and neurons drive the function of our brains, sentience is the stuff that I believe drives consciousness. I consider this stuff purpose itself — it’s inherently valuable for anything to feel as good as it possibly can. Thus instead of evolution figuring out exactly how to program us under our open environments (as we must for our computers, and so program me to write the words that you see here) it forced us to figure things out for ourselves. Evolution somehow causally created sentience, and thus agents. Here we’re driven to function by means of the punishments to rewards that we feel.

Neil Sinhababu said...

Hi Eric! I'm for (3).

A general way I think we can discover the line in the sand is by testing whether humans who have had parts of their brains damaged can have all the experiences they did before. If damage to a brain part prevents humans from having experience of some kind, it's evidence that creatures without that part won't have experiences of the appropriate kind. So creatures without the parts that surgeons damage in successful cingulotomies for chronic pain sufferers are less likely to be capable of feeling pain. But creatures with these parts are more likely to be capable of feeling pain.

Obviously there are a number of problems that need to be ironed out for this approach, such as individuating brain parts properly. And if there's a strong sort of multiple realizability, the approach simply won't work. But it's the best approach I've heard of on this issue.

Howie said...

Only humans have souls- just as there are neuronal correlates of consciousness- perhaps there are neuronal correlates of souls? Just as some substances are radioactive perhaps there is a gland in the brain, perhaps the pineal gland, maybe there is some property of the human brain, maybe humans are the only creatures to understand math or theory of mind and the soul comes in the same package

Arnold said...

Couldn't consciousness just be a reason for a faith in oneself...

Keith Dee said...

Prof-- Are you not a fan of panpsychism? I would've thought you'd be a bit sympathetic to it, given your claims in "If Materialism Is True, the United States Is Probably Conscious"

Eric Schwitzgebel said...

Hi all -- sorry for the slow response on these comments. Chaotic week!

Some scattered replies:

William: On different kinds: What I'm interested in is "phenomenal consciousness" as it is usually called, which I take to be a single, natural kind. I could be wrong about that, though, which would be one way to escape the dilemma.

SelfAware: If I had to bet, I'd go for (4) also -- though I do think it is quite strange and almost incomprehensible given my (our?) general background conception of consciousness. (And maybe that's the point.)

dave: Thanks for that suggestion -- that's an interesting angle into the issues. Of course there is a major behavioral and cognitive shift between sleeping and waking that goes along with the bright line -- though not maybe between dreaming and dreamless sleep.

Lee: Thanks so much for the kind comment. I'll miss having you around here!

Philosopher Eric: That seems not unreasonable -- though there will still be the vagueness vs bright line choice, right?

Neil: Yes, like dave's suggestion, this seems like a possible start on an empirically justified bright-line approach. For us phenomenal realists, one of these horns ought to work out one way or another!

Howie: Right, some sort of generalization of the problem to a materialist view, like a neural correlates view.

Keith: I wouldn't rule out panpsychism, though my credence in it is << 25%. I'm worried about our ability to make much progress on these questions.

Philosopher Eric said...

“Philosopher Eric: That seems not unreasonable -- though there will still be the vagueness vs bright line choice, right?”

Right professor, no “bright line” here epistemologically, though at least a line would finally exist in a conceptual sense. And if related scientists were to find it useful then we’d expect this line to become strong at least conceptually, and even if it’s difficult to understand exactly what is and isn’t sentient. But then if this approach would be useful, why haven’t scientists taken this path already? I suspect because they do not yet have effective enough principles of epistemology from which to work.

My plan is this: First we develop a community of respected professionals with accept principles of epistemology from which to instruct the institution of science (not to mention principles of metaphysics and axiology). Scientists then use these principles to not only continue exploring hard sciences effectively, but now also soft sciences.

Here’s a demonstration of my point (which gets to your assertion to Keith that you have <<25% credence in panpsychism). We need a respected group of professional epistemologists which plainly state that there are no true or false definitions for the “consciousness” term, or any other. Thus this community would agree with panpsychists that all of existence is indeed conscious , that is when the term happens to be defined as a function of causality (for example, and assuming naturalism). But given that we do not yet have a respected community of epistemologists with this common principle at their disposal, today panpsychism and all sorts of funky notions run rampant in our soft sciences.

I don’t entirely blame the “ordinary language” philosophers for this. They wisely saw how divergences in definition screw up academic pursuits in general. But I believe that my solution is far superior to what they’ve proposed, and thus could potentially become adopted to the benefit of scientific endeavors in general.