things, objects
properties, relations
events, states, processes, facts
If we divide properties (and relations) into psychological and physical (keeping open the possibility that these categories overlap), we can say that a physical event is the instantiation of a physical property, and a psychological event is the instantiation of a psychological property.
Basic idea: no difference in psychological properties without a difference in physical properties.
Mind-Body Supervenience I: things that are alike in all physical properties cannot differ with respect to mental properties.
Mind-Body Supervenience II ("strong supervenience"): if anything x has a mental property M, then there is a physical property P such that x has P, and necessarily any object that has P has M.
Mind-Body Supervenience III ("global supervenience"): possible worlds that are alike in all physical respects are alike in all mental respects.
Clearly II entails both I and III. Whether I and/or III also entails II is trickier (and we won't worry about it).
Ontological Physicalism: the world contains nothing other than "bits of matter and aggregate structures composed exclusively of bits of matter" (Kim, 13).
possible problems: (a) maybe physics countenances things that aren't "bits of matter": what about gravitational fields? Black holes? Are they "bits of matter"? (b) on the other hand, maybe it's compatible with physics that "bits of matter" aren't actually things at all! It has been suggested that there is only one "thing," spacetime, and everything else is just properties of spacetime.
In view of these issues, a more neutral definition of physicalism might be better, something like this: The world contains nothing but the things physics talks about. Or better yet, the things a perfected physics would talk about.
Nonreductive Physicalism: the psychological properties of a system are distinct from, and irreducible to, its physical properties. However, psychological properties supervene on physical properties. [A closely related view: Property Dualism accepts that psychological properties are irreducible to physical ones, but denies that they supervene on physical properties. Kim distinguishes the two views only in a footnote, which could be confusing.]
(What exactly does "reducible to" mean? We'll talk about this in some detail later on.)
Reductive Physicalism (= Type Physicalism): psychological properties are reducible to, and reductively identifiable with, physical properties.
Varieties of Mental Phenomena
Main varieties:
1. sensations, sensory qualities, properties with a phenomenal character, raw feels, qualia
2. propositional attitudes, intentional mental states, content-bearing states, states attributed by embedded that-clauses
"propositional attitude": a proposition is what is expressed by a declarative sentence, for example that grass is green or that my cat loves me. The kind of "attitude" in question includes believing, hoping, fearing, etc. Either the proposition or the attitude can vary independently of the other: I can believe that grass is green, I can believe that my cat loves me, etc.; and I can believe that my cat loves me, hope that my cat loves me, etc.
Other, perhaps derivative varieties of mental phenomena:
feelings and emotions
volitions
character traits, habits, abilities, etc.
Candidates for the "Mark of the Mental"
Goal: necessary and sufficient conditions (C is a necessary condition for P iff, if P, then C; C is a sufficient condition for P iff, if C, then P)
1. direct or immediate knowledge (I know about my own mental states without needing to infer this from evidence). Arguably not sufficient: It seems that I can know I'm looking at a tree without inferring this from any evidence beyond my immediate experience.
2. privacy, or first-person knowledge (I have access to my mental states in a way no one else does). This takes care of the looking-at-a-tree example, but still appears not to be sufficient: I have proprioceptive knowledge of the positions of my limbs in a way no one else does, but they aren't mental.
3. infallibility and transparency
knowledge of m is infallible iff [if and only if], if I believe that I have m, then I do have m
m is transparent iff, if I have m, then I believe that I have m
Transparency doesn't seem to be necessary (since unconscious mental states are not transparent).
4. nonspatiality (Descartes thought that this was a mark of the mental, but I think it seems a bit odd to most people now)
5. intentionality
Philosophers use the term "intentionality" in a somewhat odd way: it is synonymous with "representationality." That is, a state is an intentional state if and only if it represents or stands for something else.
Of course, there are representational states that aren't mental (so this isn't a sufficient condition): e.g. sentences in books. But perhaps intrinsic intentionality (that is, intentionality not derived from anything else) provides a necessary condition for a certain type of mental state.
One way to try to put most of these criteria together to get necessary and sufficient conditions for being a mental state:
A state M of a subject S is an experiential mental state iff S can directly and infallibly know that S is in M, and no one else can directly and infallibly know this.
M is a contentful mental state iff M has intrinsic intentionality.
M is a mental state iff either M is an experiential mental state or M is a contentful mental state.
The disjunctive character of this attempt seems a bit unsatisfying. Why do we group experiential and contentful mental states together into the same category if they don't have some essential characteristic in common?