George Molnar sets out 5 key features of power properties: directedness, independence (they exist whether manifested or not), actuality, intrinsicality, and objectivity (mind-independence). His discussion of directedness is very interesting, as he argues that the directedness of powers has all the elements associated with intentionality. Therefore, the existence of (mind-independent) causal power properties means that physical intentionality is a ubiquitous feature of the world. (Please note the title of this post is my creation, and not a term used by Molnar).
The paradigm accounts of intentionality are those of mental intentionality due to Brentano and like-minded philosophers. In fact, Brentano took intentionality to be what distinguishes the mental from the physical. In philosophy of mind, a debate of the following sort has taken place: a dualist would make the case for the uniqueness and irreducibility of mental intentionality; a materialist would (rarely) claim that introspectively revealed intentional mental states are illusory, or would (more commonly) argue that physical things also exhibit intentionality and either take an instrumentalist rather than realist view of the whole phenomenon, or suggest a way to reduce it to non-intentional physical description. The dualist might counter that the examples of physical intentionality (a map, for instance) are all examples of derived intentionality with true mental intentionality being the origin of these seeming cases. (There is obviously immensely more to these debates than this caricature: see this SEP article).
Molnar turns this all around. He accepts the existence of mental intentionality and argues that “something very much like intentionality is a pervasive and ineliminable feature of the physical world.” (Emphasis original, Ch.3, p.61).
He presents these parallels between the directedness of powers and the intentionality of mental states:
1. Physical powers, like mental states, are directed towards something beyond themselves. In the case of the power, it is directed toward its manifestation.
2. In both cases, they can exist even if their intentional object does not exist. A power can remain unmanifested.
3. There can be indeterminacy about the intentional object. The power can be a propensity toward a manifestation.
4. Finally, Molnar describes a parallel between two semantic criteria for the intentional as applied to the two cases: the non-truth functionality of the intentional reference and referential opacity.
Molnar considers several possible objections to the notion of physical intentionality.
He looks at objections which try to point out other distinctive aspects of mental intentionality which counter the analogy. It seems there can be impossible intentional objects and absolutely unique intentional objects in the mental realm but not in the physical. He thinks there are merits to these cases of potential disanalogies, but in the first case, representations of impossible objects are a minor and atypical example of mental states, and in the second the uniqueness seems to depend on the experiential rather than the intentional nature of the state. These cases don’t make intentionality the demarcation between mental and the physical. So the analogy seems robust enough.
The most important possible objection has to do with the relationship between intentionality and meaning. Mental states are directed toward their intentional objects by representation, where the representation (whether pictorial, symbolic or in some other form) provides meaning. The states are “about” the objects. It seems we can’t extend this to the physical: power properties do not represent. Solubility is directed toward the dissolving of the solid, but doesn’t feature a representation of the event: it isn’t “about” it, in that way. Molnar’s strategy, here again, is not to deny the presence of semantic properties of mental states, but to loosen the tie between representation/aboutness and intentionality by pointing out paradigm cases of mental intentionality which do not include aboutness.
He points out that even in the case of perception, the perception cannot be wholly reduced to a representation. But he concedes it still has a great deal to do with representation. A much better example to use for a mental state without representation is in the realm of bodily sensation, and specifically pain. Pains are mental states which meet the four criteria for intentionality discussed above. But do pains have meaning above and beyond these intentional features? Are there representational features? Molnar says no: “…a headache does not represent my head hurting, it is my head hurting.” (Emphasis original, p. 77) The only sense it makes to say that the pain represents the bodily hurt involved is in a sense where you might say an effect represents its cause. And of course one could say this in cases of physical causation as well. If this is meaning, it is a form of natural, non-representational meaning. Molnar expands on the example to assert there are two kinds of mental intentionality, roughly corresponding to a traditional distinction between the rational and the sentient. The first can be analyzed in terms of representational content of the state, the second cannot. The second kind is the kind which is also true of the power properties of the physical world.
A last objection, of interest to me of course, is referred to by Molnar as “the threat of panpsychism” (p.70; why it is so darn “threatening”?). The argument is that the case for physical intentionality actually makes the case that the mental realm is co-extensive with what we think of as the physical realm. Molnar says he would not endorse this argument, because while he is ruling out using intentionality to demarcate the mental from the physical, there should be other ways of doing so, such as the “capacity for consciousness”. But, alas, this goes beyond the aims of the book. It is my view that the most elegant and compelling metaphysical account sees the human mind as a (particularly interesting) complex instance of a natural causal nexus which is ontologically grounded in the same way as the rest of the universe. No ontological demarcation is needed or wanted.