Why I’m a Dualist

If you disagree, you're a p-zombie

What sorts of things are mental states like belief, love, or pain? The question was first posed to me in a philosophy of mind class I took in sophomore year of college. At the beginning of that semester I was a staunch materialist: I believed that everything that exists is made up of physical stuff. By the end, I was convinced that mental states are immaterial and fundamentally non-physical aspects of reality. You can call this view mind-body dualism. This post aims to explain the arguments that tilted me away from materialism and examine remaining weak points in my views.

Some background ideas

To understand dualism and the arguments for and against it, you need to understand some background ideas. I couldn’t think of a clever way to motivate these ideas, so you’ll just have to take my word that these ideas will be helpful down the line.

1. Mental states

‘Mental states’ refer to a wide range of internal experiences. Things like: cognitions (eg. beliefs, memories, intentions, desires), emotions (eg. anger, happiness), perceptions (sights, sounds, smells, etc), and bodily feelings (eg. pain, hunger).

2. Phenomenal vs functional mental states

We can define particular mental states, or describe what they are, in functionalist terms or phenomenal terms.

A functionalist description of a mental state says that we can define the mental state by pointing to the causal role that the mental state occupies in the mind. Something ‘occupying a causal role,’ just means: it has a certain type of cause and causes a certain type of effect or effects. For example, if I shoot a gun, my finger occupies the causal role of pulling the trigger. If a robot shoots that same gun, its robot finger occupies the same causal role as my flesh-and-blood finger did. As an example of a functionalist account, we could define pain as ‘a process in the brain that occurs when your body is damaged and causes you to say ‘ow!’’ This definition identifies what is supposed to cause pain, and what effects pain is supposed to cause. To the functionalist, it doesn’t actually matter what is being caused and doing the causing: if it has the right causes and causes the right effects, then it is pain.

On the other hand, a phenomenal account of a mental state defines the mental state by pointing to ‘what it is like’ to experience that mental state. For example, a phenomenal account of pain says that pain is just the experience of pain. This sounds circular, but it isn’t. What we’re really saying is that some feeling counts as pain in virtue of the way that it feels. To illustrate this idea, imagine a sensation called shmain: shmain feels like what you would ordinarily call pain, but for some reason, it causes you to jump for joy and yell ‘yippee!’ A functionalist account of shmain would say, ‘Well, since shmain plays a different causal role than pain, it isn’t pain.’ To me, that's crazy! Pain is whatever feels painful. The phenomenal account correctly says that since shmain feels like pain, it is pain.

3. Properties

Properties are the way things are. They are entities that objects can exemplify or instantiate. For example, take a red apple. The apple exemplifies or instantiates the property of redness; red is one of the ways that the apple is.

4. Are mental states properties?

Some philosophers identify mental states with properties of minds. In other words, they claim that mental states are ways that a mind can be, or that mental states are exemplified/instantiated by minds. For instance, when we say that a mind is ‘in pain,’ or that a mind ‘believes,’ we really mean that the mind is instantiating the property of pain/belief. Identifying mental states as properties clarifies what exactly we’re debating when we ask if mental states are physical or non-physical. If mental states are properties, the question becomes: Are these properties physical (like mass or charge), or fundamentally non-physical (irreducible to physics)? This framing helps because it focuses the debate on whether physical explanations completely describe mental states.

My view: Property Dualism

I believe in property dualism: the view that mental properties, as fundamental aspects of reality, are irreducible to, or not made up of, the physical. Since mental states are properties of minds, or mental properties, we can also describe property dualism as the view that mental states, as fundamental aspects of reality, are irreducible to the physical.

Zombies!

This is the fun part: the argument that convinced me to be a property dualist. To get it off the ground, I have to explain what a ‘p-zombie’ is: Imagine a human that is physically identical to you in every way. There is, however, something horribly wrong with your twin: they lack any phenomenal experience. In other words, they don’t feel anything: they are a phenomenal-zombie, or ‘p-zombie’ for short. This isn’t to say that they don’t have functional mental states. When the p-zombie stubs its toe, it yells ‘ow!’ just like you would. Its yell is, moreover, caused by the same causal processes, the same series of neurons firing, that cause you to yell when you stub your toe. The difference between you and the p-zombie is that the p-zombie doesn't experience the accompanying feeling of pain that you experience.

The first part of the ‘p-zombie argument’ in favor of property dualism is meant to establish that such a p-zombie is metaphysically possible. Something is metaphysically possible when the laws of metaphysics don’t rule it out. Take Superman for example. Superman is clearly physically impossible: the laws of physics rule out a person flying unassisted or shooting laser beams from their eyes. But Superman is perfectly metaphysically possible; the existence of Superman doesn’t, for instance, involve any logical contradiction or contradiction in concepts¹.

Part one of the p-zombie argument goes like this:

1. P-zombies are conceivable.

2. Anything that is conceivable is possible.

3. Therefore p-zombies are possible.

Let’s break down each premise. The first premise asserts that you can conceive of a p-zombie existing. Although some philosophers challenge this claim, I suspect my written description of your p-zombie twin helped you conceive of such a being. I certainly have no trouble conceiving of a p-zombie. The second premise is quite controversial. However, the source of the controversy lies outside the specific issue, mind-body dualism, that this post is about. So rather than get into the nitty gritty of debates about the epistemology of modality (don’t worry if you don’t know what that means), I’ll just provide two examples that illustrate the intuition behind premise 2, and then explain why a weaker version of premise 2 still allows the argument to go through.

Example one: Consider a square circle. Certainly such an object is impossible. There could not be a circle that is square. Now try to imagine a square circle. It’s tough, right? This example, and many others, illustrate that things that we know to be impossible often correlate with things that we cannot conceive of. This seems to support the idea of a link between possibility and conceivability.

Example two: Consider a blue giraffe. You can conceive of a blue giraffe. Presumably, blue giraffes are possible. This and many other examples establish that, at least often, possible things correlate with conceivable things.

Suppose, however, that conceivability and possibility can come apart. Some philosophers have argued that it is nonetheless the case that conceivability involves the “appearance of possibility.”² If we endorse the principle that something appearing to us in a particular way provides defeasible evidence of it being that way³, then we can get to the conclusion that, at least, we have defeasible evidence that p-zombies are possible. The burden then rests on the opponent of p-zombies to affirmatively argue against the possibility of p-zombies. I am personally agnostic between the strong (anything conceivable is possible) and weak (if something is conceivable we have defeasible evidence that it is possible) versions of premise 2, but I’ve never been convinced by an argument that p-zombies are impossible, so I’m pretty comfortable accepting part one of the argument.

The second part of the p-zombie argument is where the rubber hits the road. It goes like this:

1. P-zombies are possible.

2. If p-zombies are possible, then property dualism is true.

3. Therefore property dualism is true.

Premise 1 is just the conclusion of part one, which I already defended, so I don’t have to defend it here. The tricky part is premise 2. To understand it, it's useful to first switch out ‘property dualism is true,’ with ‘mental properties are physically irreducible.’ This is just a restatement of the thesis of property dualism. The key insight is that if X is reducible to Y, then the Y-facts entail the X-facts. For example, hurricanes are reducible to particle physics: if you knew the position, velocity, and physical state of all the particles that make up a hurricane, you could, in principle, deduce any fact about the hurricane. Finally, consider the following: if p-zombies are possible, then it is possible for the same physical facts about a person to coexist with two different sets of mental facts: they may either be a p-zombie or a normal person. Therefore, the physical facts about a person do not entail the mental facts about them, and thus, the mental properties of that person are not reducible to the physical. Voila! I’ve just proven property dualism.

Revenge of the physical

Or have I? There are, of course, strong objections to property dualism. One of the strongest, in my view, is the objection from the ‘causal closure of the physical.’ The ‘causal closure of the physical’ is just a fancy way of saying that all physical effects have physical causes. Seems intuitive enough. This principle is, after all, the foundation of modern physics. If it weren’t true, we’d be in trouble. The dangerous part is this: mental states are, or at least seem to be, causally efficacious. In other words, mental states cause physical effects. For example, being in pain as a result of touching a hot stove seems to cause you to pull your hand away from the hot stove. But if physical effects, like pulling your hand away from a hot stove, can only have physical causes, then it must be that mental states are physical. This, of course, contradicts property dualism.

This argument seems solid to me. I confess that I don’t have a strong justification of the causal closure of the physical. It just seems true to me. To make up for my lack of argumentative rigor, I promise to circle back to this issue in a later post. Of course, I can’t accept this argument and be a property dualist. Since I’m not ready to abandon the causal closure of the physical, I am left with only one option: deny the causal efficaciousness of mental states.

Is Consciousness Mere Seafoam?

Epiphenomenalism is the view that mental states have physical causes but cannot have physical effects. Mental states are, in this view, like so much seafoam, churned up by waves but unable to affect the waves in return. I am (tentatively) an epiphenomenalist. I am not, however, a happy epiphenomenalist. I think the position has serious issues. I adopt it only because my commitment to the causal closure of the physical forces me to. I’ll explore two major challenges to epiphenomenalism in this section.

The first challenge comes in the form of a question: How do you know you’re not a p-zombie? Well, it seems obvious: you can feel things. You’re feeling things right now. So of course you’re not a zombie! Let’s think about this more closely though: epiphenomenalism claims that mental states have physical causes. Belief is a mental state. So your belief that you are feeling things must have a physical cause. Moreover, epiphenomenalism claims that phenomenal mental states are physically causally inert, so whatever feelings you believe yourself to be feeling could not be the causes of your belief that you are feeling things. Most accounts of knowledge stipulate that justification is required for knowledge. That is to say, if you know some fact P, you must be justified in believing that P. The challenge then is this: how can you be justified in believing that you have feelings if those feelings are not causally connected to your belief in those feelings? It seems like your belief isn’t justified. This yields the insane conclusion that, actually, you don’t know whether you’re a p-zombie or not. This is problematic for epiphenomenalism to say the least. I don’t know how to resolve this issue⁴. If you’ve got ideas, please comment them. I could use some help!

We can also state the second challenge as a question: Why do particular mental states correlate with particular types of physical events if they do not cause those physical events? To get an intuitive grasp of this challenge, recall the hot stove example. When I feel pain, I pull my hand away from the hot stove. There is, in other words, a correlation between the feeling of pain and aversive behavior towards the source of that pain. We can explain the correlation easily if pain is causally efficacious; pain tends to co-occur with aversive behavior because it causes the aversive behavior. Without this causal connection, it becomes much harder to explain the correlation.

I’ll take a stab at it anyway. My rough solution to this challenge involves what you can call ‘psycho-physical laws.’ Think of these like the laws of physics, but for mental states. These psycho-physical laws fix which physical states give rise to which sorts of mental states. In other words, these laws explain why certain physical states reliably correlate with certain mental states. Let’s go back to the hot stove case. Suppose that according to the psycho-physical laws, physical state X is the sufficient cause of your pain. Now suppose that physical state X is also the sufficient cause for you to pull your hand away from the hot stove. Now we have an explanation for why pain and pulling your hand away reliably co-occur: they share a physical cause.

What remains

There remain, of course, more unanswered questions. I’ll list a few topics here which I neglected to discuss in this post, but may be the subject of future posts:

1. Kripkean cases of a posteriori necessity as a challenge to the conceivability-possibility link.

2. The question of substance dualism: are mental states properties of a physical substance, like your body, or of a distinctly mental substance?

3. An explanation of psycho-physical laws: What they may be, and how we may come to know them.

4. An examination of the best arguments for and against the causal closure of the physical.

5. Discussion of the interaction problem: how can physical events cause non-physical events (ie mental states)?

Regardless, I think I’ve established that if you’re willing to bite the bullet on epiphenomenalism (which admittedly is a big bullet to bite), then there is a very strong case for being a property dualist like me. If you find yourself an unconvinced materialist by the end of this post, well, you’re probably a p-zombie. Cheers!

1. See: https://plato.stanford.edu/entries/modality-varieties/ for a more complete description of different varieties of possibility.

2. https://plato.stanford.edu/entries/modality-epistemology/#VariModaTargEpisMod

3. For example, if I see something, and it appears blue, my seeing it as blue provides defeasible evidence that it is blue.

4. One approach might involve invoking a ‘reliabilist’ account of justification, where reliability is understood in a frequency sense (pertaining to what occurs in the actual world). See more here: https://plato.stanford.edu/entries/reliabilism/#BasiCompReliMentProcDeteTrueVsFalsBeli