The kind of dualism we've been working with so far is called Cartesian or substance dualism. There's also another kind of dualism, property dualism.

  • The substance dualist says that there are special non-physical substances, souls. Since these souls are substances, they can in principle exist independently of our bodies. Having a mind requires having a soul.

  • The property dualist denies that there are any souls. The property dualist agrees with the materialist that the only substances there are are physical substances, like rocks and human bodies. But the property dualist says that there are two kinds of properties. There are physical properties and there are mental properties; and the mental properties are independent of, and can't be reduced to, the physical properties. They can vary independently. After God distributed all the physical properties, he had a further choice about how to distribute the mental properties.

    (You may hear some philosophers talking about "emergentism." That's a version of property dualism.)

  • The materialist denies that there are any souls and he also denies that mental properties are independent of physical properties in the way the property dualist says. According to the materialist, what mental properties you have is wholly determined by what physical properties you have. Once God distributed all the physical properties, there was nothing further for him to do. It was already settled who had what mental properties.

    (Some philosophers use the term "physicalism" as a synonymn for materialism. Other philosophers use "physicalism" to refer to a specific version of materialism, called ¥¥"identity theory," that we'll be looking at later. You'll have to try to figure out from the context what each philosopher means by "physicalism." Sorry it's so messy.)

What does it mean to say that your mental properties "are wholly determined by" your physical properties? To understand this better, we need to introduce some new conceptual apparatus.

Necessity and Possibility

To say that something is necessary is to say that it must be the case. When something is not necessarily false, when it could have been the case, then we say that it's possible. If something is neither necessarily true not necessarily false, then we say that it's contingent.

Possible Necessarily true
True, but could have been false Contingent
False, but could have been true
Not Possible Necessarily false

  1. The word "possible" in English has many different meanings. If someone comes up to you on the street and says "It's possible that I was born on the moon," you'll most naturally take him to be making some sort of (crazy) claim about what it's reasonable for him to believe really is the case. You'd be understanding his claim as "For all I know, I was born on the moon." This is a use of "possible" to express claims about evidence and knowledge. It's called an epistemic sense of possibility.

    But there is another thing the guy might have meant. He might have meant, "Look, I know I was born on the Planet Earth. But things could have gone differently. I could have been born on the moon--if for instance mankind had colonized the moon in the 1800s." Here he's using "possible" not in the sense of "for all I know..." but rather in a metaphysical sense. He's not talking about what he knows. He's talking about what could have happened, if the world or history had been different in certain ways. We call claims of this latter sort claims about logical or metaphysical possibility.

    Here's a rule of thumb to help you out. If something of this sort is ever true:

    If the world were to be different in such-and-such ways, then P would be the case
    ...then P is metaphysically possible. If nothing of that sort is true, then P is metaphysically impossible. There's no way the world could be such that P would be the case.

  2. If it's metaphysically necessary that a thing have certain properties, then we say that those properties are essential properties of that thing, and that they are part of the thing's nature or essence. So for instance, oddness is one of the essential properties of the number 3. Some philosophers would argue that being made of wood is one of the essential properties of this table. If they're right--if it is an essential property of the table--then there would be no possible situation in which the table exists, but it's made of plastic. (There may be possible situations in which some other table exists in the same location, and is made of plastic. But if being made of wood is essential to this table, then the plastic table would have to be a different table. It couldn't be one and the same table as this table.)

    Those properties which are not essential, we call accidental properties. For instance, being 3 feet tall is only an accidental property of the table (we could saw part of its legs off to make it shorter).

    Note: For philosophers, talk about your "essential" properties is not the same thing as talk about your most important or most valuable properties. It is talk about properties which it is impossible for you to lose. Your wit and charm may be among your important properties. But it's possible for a person to become un-witty, and un-charming, while still being the same person. So wit and charm are not essential properties.

  3. If something is compatible with our laws of physics--if the laws of physics would permit it to happen--then we say that that thing is both metaphysically possible and physically possible. If something is guaranteed by the laws of physics, then we say that the thing is physically necessary, and that it would be physically impossible for things to be otherwise.

    Many philosophers think things can be possible in a metaphysical sense, even if they're contrary to our actual laws of physics, and so physically impossible. For example, the charge of an electron might have been larger; or the gravitational constants might have been slightly different. We know that, as things stand, the world is not that way. And if the world had been that way, then perhaps life as we know it would not have come about. But it still seems possible, in a broad sense, for the world to have been that way. Physicists sometimes talk about, and advance hypotheses about, what the world would be like if the laws of physics or the fundamental physical constants had been slightly different.

    It's important to be clear about what this bottom category, "metaphysically impossible," means. What it means is that that is not a way the world could have been.

    Let's try to come up with some things that would be examples of this.

    Would it have been possible for the world to contain 3-sided squares? What would such a world be like? That doesn't really seem to be a possible way for the world to be, does it?

    Nor does it seem to be possible for it to be raining and not raining in the same place at the same time. Nor does it seem to be possible for a person to exist and fail to be one and the same person as himself. There are no possible situations in which things of those sort take place. They all seem to be metaphysically impossible.

    Now, so far, the examples we're coming up with of things that are metaphysically impossible all seem to have something to do with logic or definitions or math. That raises an interesting question:

    If something is metaphysically impossible, will that always be because it involves some contradiction or going against some definition?

    This is a hard question, and contemporary philosophers disagree about how to answer it. We'll be returning to this question several times this semester.


Consider the relation between (i) throwing a ball against a window, and (ii) the window's breaking. Or (iii) striking a match, and (iv) the match catching on fire. There's one event, and it causes the other event to happen. Dualists are perfectly happy to allow that what physically goes on in our bodies causes us to have certain mental states. (Though ¥¥as we'll see later, they may have difficulties explaining how these causal relations work.)

Contrast the different set of relations that obtain between (v) being enrolled at Harvard College, and (vi) being a college student. Being enrolled at Harvard doesn't cause you to be a college student; rather, that is just one way of being a college student. There are other ways to be a college student, too; you could be enrolled at some other college. But being a college student isn't some further, optional extra to being enrolled at Harvard College. Once you're enrolled at Harvard College, you're then automatically and already a college student. (Assume for the sake of this example that Harvard couldn't cease to be a college--it couldn't, for example, turn into a music conservatory or a trade school--and still be Harvard College.)

The same goes for the relation between some piece of software and the physical states of the computer it's running on. There are certain electrical changes going on in my computer. Also my computer is running a Solitaire game. But the running of Solitaire is nothing over and above those electrical changes. Their taking place doesn't cause my computer to run Solitaire, in the way the baseball caused my window to break. Rather, their taking place is what it is for my computer to be running Solitaire. It would be impossible for my computer to go through those changes and for Solitaire not to be running.

It wouldn't be impossible, on the other hand, for you to throw the ball against the window and the window fail to break. In principle, it's possible that the window might fail to break. It may even be physically possible (for example, if someone installed special dampers on the window). But it would certainly be logically or metaphysically possible for the ball to hit the window, and to bounce away harmlessly. Similarly, it wouldn't be impossible to strike a match and have it fail to light. So this is an important difference.

Let's introduce a new piece of technical vocabulary to help capture this difference.

We say that one set of facts supervenes on another more basic set of facts when the more basic facts "fix" or "determine" that the first set of facts has to be true. To be precise, when it's metaphysically impossible to have the basic facts settled in a certain way without also having the first set of facts settled, too. So in our example, whether you're a college student supervenes on what institutions you're enrolled at. It's metaphysically impossible to be enrolled at Harvard College but fail to be a college student. And if you and your friend Teresa are both enrolled at the same institutions, then either you're both college students or neither is.

Similarly, whether my computer is running Solitaire supervenes on what electrical patterns are there inside the computer. If I'm running Solitaire, and your computer has exactly the same electrical patterns, then your computer must also be running Solitaire. It's metaphysically impossible for it to have those patterns but not be running Solitaire.

Here's another example. Alex the Aesthetician says that the beauty of a painting supervenes on the facts about how the paint is distributed on the canvas. Alex may be right, and he may be wrong. We don't care about that. What we do care about is what his view says. What it says is that, once you've distributed the pain on the canvas in a certain way, it's metaphysically settled how beautiful the painting is. Any painting which has its paint distributed in exactly the same way would have to be equally beautiful. It's metaphysically impossible for there to be two paintings that are the same in how their paint is distributed, but different in respect of how beautiful they are.

Comment: This may not be the correct view about paintings. Some people say that forgeries are less beautiful than original paintings--even if they have their paint distributed in exactly the same way. That is a controversial question. But as I said, we don't care now about whether Alex's view is correct. We're just trying to understand what it says.

Note that when A supervenes on B, it does not automatically follow that B supervenes on A. There might be two paintings of equal beauty which nonetheless have their paint distributed differently on the canvas. That shows that it's sometimes possible to change how the paint is distributed, while leaving the painting just as beautiful as it was before. The claim that the painting's beauty supervenes on the distribution of paint is compatible with that. It only implies that the reverse sort of change is impossible. If the beauty is fixed by the distribution of paint, then you can't change how beautiful the painting is while leaving the distribution of paint exactly the same.

How does all that help us understand the debate between property dualists and materialists?

The materialist thinks that what mental properties you have is wholly determined by your physical properties. In other words, all the mental facts about you supervene on the physical facts about you. There could not possibly be someone who is physically just the same as you, without his also being mentally just the same.

The property dualist, on the other hand, says that whether or not you have a given mental property is in principle independent of the physical facts about you. Certain physical facts about you may typically cause you to be in given mental states. If I hit you on the head, that will typically cause you to have a headache. But, on the dualist's view, there is nothing metaphysically impossible about there being someone who is just like you physically, but who has different mental properties than you, or perhaps no mental properties at all. (The substance dualist thinks that too, because she thinks there could be someone who is just like you physically, but who has a different kind of soul, or no soul at all.)


Last Modified: 
Copyright © The President and Fellows of Harvard College