We’ve now seen a couple of different ways that words like Materialism and Physicalism, on the one side, and Dualism, on the other side, get used.
We began by using these labels to discuss disagreements about how many different kinds of substances there were.
As we’ve noted a few times, some philosophers use “physicalism” more specifically, to mean the kind of type-identity view that Smart proposes. Here for example is Block in the reading we’ll be looking at next in the class (p. 95):
By “physicalism” I mean the doctrine that pain, for example, is identical to a physical (or physiological) state [type]. As many philosophers have argued… if functionalism is true, physicalism is false. The point is at its clearest with regard to Turing-machine versions of functionalism. Any given abstract Turing machine can be realized by a wide variety of physical devices… Therefore, if pain is a functional state, it cannot, for example, be a brain state, because creatures without brains can realize the same Turing machine as creatures with brains.
But in a footnote, Block acknowledges that we could also use “physicalism” in a somewhat broader sense, where functionalists could count as physicalists. They don’t have to be: one could call oneself a functionalist, and think that (functionally-defined) mental states are or could be realized in souls. But one could be a functionalist who insists that (functionally-defined) mental states have to have physical realizations. There’s a sense in which such a view is also naturally counted as physicalist. But it rejects the kind of type-identity theory that Block was talking about.
The natural sense in which this view still counts as physicalist is that it accepts that mental facts and properties (and mental event-types) depend on how the world is arranged physically. There couldn’t be two situations in which subjects had all the same physical properties, but they had different mental properties, or one of them had mental properties and the other had none.
For many contemporary philosophers, this is the core of what they understand “physicalism” to amount to. It’s more specific than just denying the existence of souls. You could deny that there are souls, but also deny that the physical facts determine or settle all of the mental facts. Such philosophers call themselves property dualists. Physicalists in the sense we’re now considering go further, and accept the supervenience claim.
At the same time, though, it’s more general than just the Smart-type view that Block and some other authors have used the label “physicalism” to talk about. “Role functionalists” will also accept supervenience, even though they don’t accept the type-identities that Block was referring to. Davidson accepts supervenience, even though he doesn’t accept any kind of identities between mental and physical types.
As we mentioned before, some authors expect that if one’s going to be a physicalist/materialist, then if one doesn’t accept type identities, one will at least have to accept token identities. But this is not straightforward. It depends on what other background views one holds. In the context of some background views, it may be true; in the context of other background views, not. And it doesn’t seem like the choice of whether to be a physicalist about the relation between mental and physical forces one’s choice about those background views.
Another idea that philosophers sometimes link with the idea of being a physicalist is whether one will think that there are at least laws connecting the mental and the physical, even if there aren’t identities. But we’ve seen Davidson arguing that that one can be a materialist without thinking there are any strict mental-physical laws. (Perhaps there can be ceteris paribus laws.)
On the other hand, dualists (whether substance dualists like Descartes, or contemporary property dualists) may be happy to accept that there are mental-physical laws. They can accept those while denying that the mental facts supervene on the physical facts. (This combination of views makes sense because supervenience talks about what has to follow from the physical facts with the strongest kind of necessity, whereas the laws only talk about a weaker, causal sort of necessity.)
Even if a physicalist accepts some identities between mental types and physical types, they don’t have to think that those identities are part of the meaning of our mental concepts. We’ve seen some physicalist theories who think they are: “logical/analytical” behaviorists and “analytical” functionalists. But we’ve also seen some physicalist theories who don’t think that: Smart’s form of type-identity theory, and “scientific/psycho-functionalism.”
All of these are extra choices that one can make, inside the broad category of physicalism. The way we’re now understanding physicalism, what defines that view is whether one accepts that the mental facts supervene on the physical facts. Not these additional choices.
We’ve come across the term “reduction” a couple of times in our reading. What would it mean to say that mental facts “reduce to” physical facts?
Sadly, this is one of the issues where philosophers have been especially messy. It seems that this can mean a variety of different things, and in different conversations the meanings may not be the same.
We saw a paradigm of a Reductionist view when Broad was describing the “Mechanistic” view he opposed. That view said that the chemical and biological and psychological facts could be deduced from the physical facts and a small number of general laws. Many times “reductionism” has this kind of epistemic component to it: it has to do with what people could deduce or infer a priori (without doing further scientific investigation) if they were informed of the relevant physical facts.
Understood in that sense, Smart’s kind of physicalism would not count as a reductionist view. But the logical/analytical forms of behaviorism and fucntionalism would.
Other times, though, philosophers work with a broader understanding of “reduction,” where any kind of type-identity view would be enough to count as having a reductionist view, even if those identities are only knowable a posteriori (or even not knowable at all).
A view like Davidson’s though, or any kind of physicalist who endorsed supervenience but denied that there were type identities, would generally not be counted as a reductionist. These views are sometimes called non-reductive materialists.