Forms of Non-Reductive Physicalism

---

Davidson's "Mental Events"

• Davidson identifies three principles which he appear to be inconsistent with each other
  1. Principle of Causal Interaction: some mental events interact causally with some physical events.
  2. Principle of the Nomological Character of Causality: events related as cause and effect fall under strict deterministic laws.
  3. Anomalism of the mental: there are no strict deterministic laws on the basis of which mental events can be predicted and explained.
• Davidson will now argue
  1. The three principles can be mutually consistent
  2. There are no psychophysical laws
  3. These two findings allow for a kind of identity theory.
• 1. The three principles can be mutually consistent
  • First, we need to determine what the mental and physical are
    • Mental events are events that are described by an expression that requires at least one mental term. (This assumes irreducibility! But moving along....)
    • Physical events are events that are described by an expression that requires at least one physical term.
  • Lawlike relations between physical and mental phenomena are presumably relations between physical kinds and mental kinds.
  • Davidson identifies four kinds of positions
    
    

    	Monism	Dualism
    Psycho-P Laws	Nomological Monism (e.g., reductive physicalism)	Nomological Dualism (e.g., interactivism, epiphenomenalism, paralellism)
    No Psycho-P Laws	Anomalous Monism (Davidson's view)	Anomalous Dualism (e.g., claim: Descartes)

    
    
    (I find it perplexing that parallelism is an example of nomological dualism. Parallelism is by definition the view that there are no psycho-physical laws but that things work out OK anyway. You might argue that with a Humean account of laws parallelism gives you psychophysical laws, but then it's not parallelism. For Descartes, his commitment to the freedom and immediacy of the mental can be read as saying the mental obeys no laws; and then you get no psychophysical laws. But if you read him as an interactivist open to the possibility of mental laws, he is a nomological dualist.)
    
    
  • Anomalous monism is the view that some events described with physical vocabulary are not fully described; sometimes it is appropriate also to use mental vocabulary.
  • Mental descriptions somehow elude law. Davidson argues this is analogous to the truth predicate: a logical system cannot contain a predicate expressed only in the syntax of that system that describes all and only true sentences of the system. This is just an analogy! But he's claiming that "mental" is like "true" -- it will exceed the physical theory.
  • The three principles can be consistent because the second principle refers to kinds, and the first principle refers to particulars.
• 2. There are no psychophysical laws
  • Psychophysical laws will be laws relating the mental and the physical
  • Davidson claims there are no psychophysical laws.
  • He will motivate this claim with a discussion of philosophical behaviorism
    • We reject philosophical behaviorism because we see there is a systematic kind of failure in the attempted reduction
    • Claim: when we try to define some mental content in terms of behavior, we find we must add endless provisos which are themselves mental.
    • This is Davidson's Holism about the mental: he claims we can only have beliefs in relation to other beliefs, and all mental states must exist in a vast network.
  • Lawlike relations are between the right kinds of things: we recognize beforehand what kinds of things can be so related.
  • Mental and physical predicates are not related such that lawlike relations fit between them.
  • Here's the punchline: "Just as we cannot intelligibly assign a length to any objecdts unless a comprehensive theory holds of objecdts of that sort, we cannot intelligibly attribute any propositional attitude to an agent except within the framework of a viable theory of his beliefs, desires, intentions, and decisions."
  • But "we make sense of particular beliefs only as they cohere with other beliefs, with preferences, with intentions, hopes, fears, expectations, and the rest."
  • Davidson is not giving us an argument, though this pretends to be one; rather he is describing his holism, and showing how it may entail a kind of anomalism about the mental.
• 3. These two findings allow for a kind of identity theory
  • Each particular mental event is a physical event
  • But: there are "no strict laws at all on the basis of which we can predict and explain mental phenomena."
  • Physical theory determines what physical events occur; all mental events are physical events; but physical theory does not determine what mental events there are.
  • Davidson avoids stating it, but this surely means that various kinds of physical events might instantiate at different times a particular mental kind.

Some observations

• Davidson mentions supervenience. Roughly, x supervenes on y means that there can be no change in x without some change in y. This is usually coupled with an implicit assumption that x and y are not identical. Thus, there may be changes in y without changes in x. If the mental supervenes on the physical, every change in the mental requires a change in the physical, but not every change in the physical results in a change in the physical.
• Davidson's interest is really propositional attitudes. He tends to treat mind and language ability as equivalent. Keep this in mind. His arguments really target beliefs and other PAs, and not things like perception.
• How can it be that mental events are physical, but there are no mental laws? Davidson is cagey, but elsewhere seems to grant that this is a multiple realizability argument. Here is an example from his paper "The Material Mind":

  What I have supposed is that for any particular, dated psychological event we can give a description in purely physical terms; and so for any given, finite class of events, we can set up a correlation between psychological and physical descriptions. But although this can be done, it does not follow that such psychological predicates as "x desires his neighbor's wife," or "x wants a kaffee mit schlag" or "x believes that Beethoven died in Vienna" or "x signed a cheque for $20" which determine, if not infinite classes, at least potentially infinite ones;-- it does not imply that such predicates have any nomologically corresponding physical predicates. Of course, if a certain class of psychological events is finite, and each psychological event has, as we are assuming, a physical description, then it follows trivially that there is a physical predicate that determines the same class as each psychological predicate. But this fact is in itself of no interest to science. Science is interested in nomological connections, connections that are supported by instances, whether or not the instances happen to exhaust the cases.

• Question: is this kind of multiple realizability plausible?
  
  
  Revised 2004.10.01