1. Actions, Reasons, and Causes

Davidson's (1963) “Actions, Reasons, and Causes” offered a defense of the thesis that an agent's acting on or for a particular reason – as distinct from the agent's merely acting in accord with a reason – requires that the reason contribute causally to the production of the action.

What is the relation between a reason and an action when the reason explains the action by giving the agent's reasons for doing what he did? We may call such explanations rationalizations, and say that the reason rationalizes the action…I want to defend the ancient—and commonsense—position that rationalization is a species of causal explanation.¹

In so arguing, Davidson was setting his sights on a philosophical establishment steeped in attitudes encouraged by Wittgenstein. Davidson cites Gilbert Ryle, Elizabeth Anscombe, Stuart Hampshire, H.L.A. Hart and A.M. Honoré, William Dray, Anthony Kenny, and A.I. Melden as prominent philosophers who had argued that reason explanations are not a species of causal explanation.²

One line of anticausal reasoning began with a comparison of ordinary explanations of actions and causal explanations found in the sciences. Causal explanations invoke natural laws. You explain why a particular natural event occurred by showing that its occurring then and there is an instance of a law. This might mean deducing the occurrence from a statement of the circumstances under which the event occurred together with a law. Explanations of actions appear to be dramatically different. You explain Betty's suddenly leaving the room by reference to Betty's beliefs (that her car is parked at a meter that is about to expire) and desires (that she not be ticketed for illegal parking). It is most doubtful that there is a natural law governing such occurrences. Even if there were, your explanation of Betty's action appears not to rely on laws in any way. Reasons apparently explain, not by playing a causal role in the production of actions, but by illuminating, or making sense of actions performed. To think of reasons as causes is to confuse intentional explanation, which aims at an understanding of agents' perspectives on their own actions, with mechanistic explanation of the sort appropriate in the nonintentional sphere.

In response, Davidson notes that you could have a reason, a good reason, to perform a given action and perform the action, but not for that reason. As Kant observed ruefully, actions can accord with more than one of an agent's reasons, some commendable, some not. When agents act, the question is, on what reason did they act, what is the reason for which they acted?

A person can have a reason for an action, and perform the action, and yet this reason not be the reason why he did it. Central to the relation between a reason and an action it explains is the idea that the agent performed the action because he had the reason

(Davidson 1970: 9).

The force of “because” here, Davidson contends, is causal. Reasons rationalize an action when they motivate the action, and motivating is a causal notion.

One consideration thought to bear on the question whether reasons were causes appealed to the manifest contingency of causal relations. Reasons explanations could not be a species of causal explanation because reasons bear logically on actions they rationalize. Part of what it is to want to A is to A unless something prevents it. Causal relations, in contrast, hold between logically independent events. This is the “logical connection” argument developed most expressly by Melden (1961). Davidson (1970: 10) suggests that the argument might be put this way: “when we explain the action by giving the reason, we…redescribe the action; redescribing the action gives the action a place in a pattern, and in this way the action is explained.”

Davidson points out, however, that a reason (or the onset of a reason, an event) and the action it rationalizes are distinct events. Their being distinct does not preclude descriptions of the events being logically related. Think of a case not involving reasons. Suppose your pulling the rope caused the bell to ring. In that case, it would be true that what caused the bell to ring (namely, your pulling the rope) caused the bell to ring. A “logical connection” holds between descriptions of the events, not the events themselves, not the causal relata.

Although “anticausalist” conceptions of action have recently made a comeback (see, for instance, Dancy 2000; Lowe 2008), it is probably safe to say that most philosophers, even philosophers who have little sympathy for Davidson's views on other matters, are happy to accept his conception of reasons as causes, or at least accept that explanations of intentional actions are causal in nature. Now, however, a number of other problems must be faced. It would seem that psychological explanations generally, not merely reasons explanations of actions, are causal. But how is this supposed to work? Causality might be thought to require natural laws, and, although psychology has gradually produced a host of important lawlike generalizations, these fall short of the kinds of “strict” law you find in the “hard sciences.”

More seriously, it would seem that psychological explanations are inherently provisional. Intelligent agents are, whatever else they are, physical systems that operate on physical principles. When an agent acts on a reason, the agent's body moves in particular ways. Bodily motions are apparently brought about by purely physical occurrences. Suppose Betty leaves the room because she wants to top up a parking meter. Betty's leaving the room includes her body's moving in particular ways. Surely there is a purely physical, nonpsychological, explanation for these bodily motions, a “mechanistic” explanation that appeals only to purely physical events (see Malcolm 1968). Were that not so, there would be physical events – all the physical events constituting actions – that lacked complete physical explanations. We regard the physical world as “causally closed,” however: every physical event that has a cause has a proximate physical cause.

You might be skeptical about such closure principles, but even if you are, you might want some account of how physical and mental causes could coexist peacefully. Attention now shifts from the question, whether reasons cause actions, to, how reasons could cause actions. This brings us to “Mental Events.”

2. Mental Causes

In “Mental Events,” Davidson (1970: 208) defends the compatibility of three theses (or “principles”) that bring these questions into focus:

(1) Principle of Causal Interaction: “at least some mental events interact causally with physical events.”

(2) Principle of the Nomological Character of Causality: “where there is causality, there must be a law: events related as cause and effect will 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.”

You might find one or more of these theses objectionable, but if you do only, or primarily, because you regard them as inconsistent, you would do well to hear Davidson out. Each thesis has a measure of plausibility quite independently of questions about the mind–body problem and mental causation. An attempt at reconciliation, if successful, would afford a satisfying depiction of mind–body interaction, a depiction not requiring controversial trade-offs or ad hoc adjustment to otherwise plausible doctrines.

I shall begin by spelling out Davidson's position as it is most often spelled out. The position, which has spawned countless discussions and responses, will be familiar to many readers. I will then back up and describe what I understand as Davidson's actual position. If I am right, then much of what has been written about Davidson's approach to the mind–body problem will be revealed as irrelevant. This is important, I think, because the view standardly attributed to Davidson has been hugely influential in motivating “nonreductive physicalism,” today's orthodoxy in the philosophy of mind and philosophy of science. Getting Davidson right exposes the shaky foundations of this orthodoxy.

Start with mental and physical events. Physical events, Davidson says, fall under “strict deterministic laws,” but there are no such laws governing mental events. This would seem immediately to exclude the possibility of either mental–mental or mental–physical causal interaction. Davidson insists, however, that “at least some mental events interact causally with physical events.” How could this be?

What if every mental event were a physical event, what if mental and physical events were “token identical”? That would allow for mental–mental and mental–physical causation, but it would also seem to allow for “strict deterministic” psychophysical laws, a possibility Davidson (1970: 212–213) denies. And what if, while accepting mental–physical “token identity” (every mental event is a physical event), you rejected mental–physical “type identity” – mental types are not identifiable with or reducible to physical types? Every event falling under a mental type, falls under some physical type, even though mental types are not “reducible to” physical types.

So what are “types”? Davidson speaks of mental and physical “terms” and “descriptions.” Presumably, terms or descriptions apply to events in virtue of those events' characteristics or properties. “Types,” then, must be properties, where properties are taken to be universals. So, mental properties are distinct from physical properties.

Now we have the thesis that every (particular) mental event is a physical event, but mental properties are distinct from physical properties. An event is mental when it “falls under a mental description” and physical when it “falls under a physical description” (Davidson 1970: 210). Descriptions apply to events owing to those events' properties, so we have the thesis that every mental event, every event with a mental property, is identical with some physical event, some event with a physical property. Jaegwon Kim summarizes.

Imagine a Davidsonian universe of events: all of these events are physical, and some of them are also mental. That is to say, all events have physical properties, and some of them have mental properties as well. Such is Davidson's celebrated “anomalous monism.”

(Kim 1989: 269)

Mental properties are not reducible to, or identifiable with, physical properties. This, however, is consistent with a kind of dependence of mental properties on physical properties.

Although the position I describe denies that there are psychophysical laws, it is consistent with the view that mental characteristics are in some sense dependent, or supervenient, on physical characteristics. Such supervenience might be taken to mean that there cannot be two events alike in all physical respects but differing in some mental respect, or that an object cannot alter in some mental respect without altering in some physical respect.

(Davidson 1970: 214)

Mental properties “supervene” on physical properties. Given the intimate connection between what goes on in our brains and our mental lives, this is to be expected. Supervenience as characterized by Davidson is consistent with reducibility, consistent with “type identity,” consistent with the possibility that mental properties are physical properties. However, this possibility is excluded in the course of reconciling the three theses with which the argument began.

An event is mental, then, if it has a mental property and physical if it has a physical property. In light of supervenience, every mental event, every event with a mental property, is a physical event, an event with a physical property. There is no difficulty in the idea that “some mental events interact causally with physical events”; the mental events in question are (also) physical events, and, in consequence, the interactions “will fall under strict deterministic laws.” Mental properties, although dependent on, are distinct from, physical properties. This is why “there are no strict deterministic laws on the basis of which mental events can be predicted and explained.” When a mental event falls under a law, it does so owing to its physical properties. Think of a law as having the form, the Fs necessitate the Gs: N(F,G). The Fs and Gs are physical properties.

Whenever an event has a mental property, it has some physical property, some property apt to figure in a law. This is the supervenience of the mental on the physical. The denial of reducibility means that you can know of any mental event that it is physical, know that it has some physical property, without having a clue as to what that physical property is. Events' causes or effects, however, depend on their physical properties: events fall under laws in virtue of their physical properties. So long as you identify an event by its mental property, you are in no position to say with certainty what caused it or what it might cause. The event falls under a law, but to know which law, you would need to know its physical properties, a practical impossibility in most cases.³ Mentality is “anomalous” with respect to the physical.

Compare the moral case. Suppose moral properties supervene on nonmoral, natural properties (Hare 1952). If you have two physically indiscernible agents, they will be morally indiscernible: if one is good, the other will be good as well. But there are endless ways of being good, so no prospect of reducing moral properties to nonmoral properties, no prospect of reformulating moral claims in a naturalistic vocabulary. In this regard, morality is anomalous with respect to the natural.

3. Difficulties With This Account

Many philosophers have found fault with Davidson's argument. Jerry Fodor, for instance, takes issue with Davidson's contention that causality requires strict laws. Fodor accepts, indeed insists on, the irreducibility of psychology, but argues that psychological kinds can figure in “hedged,” ceteris paribus laws (see, e.g., Fodor 1974). This, he thinks, is enough to underwrite respectable causal relations among mental and physical events.

Other philosophers have argued at length that Davidson's anomalous monism in fact undercuts the possibility that reasons could be causes or, more generally, that mental properties could have a hand in the production of physical or mental effects.⁴ An analogy might help. Suppose your placing a red billiard ball on a cushion creates a depression in the cushion. The ball is red. A red ball creates a depression in the cushion. But the ball's being red has nothing to do with this effect. The ball has a number of properties, among these a particular density and mass. These properties are responsible for the ball's effect on the cushion. Its color is irrelevant.

The worry in the case of mental causation is that mental events could have particular effects, but those events' having these effects has nothing to do with those events' being mental. Given Davidson's insistence that it is only in virtue of possessing physical properties that they fall under strict laws, any effects a mental event might have would depend exclusively on the event's physical properties. A given mental event could have a particular physical effect, then, but not because it is mental, not qua mental.

Suppose your acquiring a reason to leave the room is your brain's going into a particular state. These events are “token identical.” Given Davidson's conception of causation (“events related as cause and effect will fall under strict deterministic laws”), and given his denial that mental properties figure in causal laws (“there are no strict deterministic laws on the basis of which mental events can be predicted and explained”), your acquiring a reason to leave the room could cause your leaving the room, but the physical nature of this event is what is responsible for your departure, not its being the onset of a reason, not qua reason. (Compare: the red ball depresses the cushion in a particular way, but not because it is red, not qua red.) By insisting on a physical basis for causality and on the irreducibility of mental properties, Davidson has rendered the mental “causally irrelevant” epiphenomenal.

All of this can be depicted by means of diagrams (inspired by Kim) that have become standard fare in discussions of mental causation. Suppose a particular event, c, causes an event e. The cause, c, possesses a physical property, P₁, and e possesses a physical property, P₂. Suppose, further, that c causes e by virtue of this sequence's instantiating a strict, deterministic law of the form, P₁-type events necessitate P₂-type events. The events fall under this law in virtue of being P₁ and P₂. Finally, suppose that M₁ and M₂ are mental properties supervenient on P₁ and P₂, respectively. By Davidson's lights, c, a mental event, an event possessing a mental property M₁, causes a physical event, e, an event possessing a physical property P₂. Because both events are mental and physical, we could say, as well that a mental event causes a mental event (the event possessing M₁ causes the event possessing M₂), that a physical event causes another physical event (an event possessing P₁ causes an event possessing P₂), and that a physical event causes a mental event (an event possessing P₁ causes an event possessing M₂).

M₁		M₂
⇑	→	⇑
P₁		P₂

But really this is misleading. Event c causes event e in virtue of c's being P₁. M₁ has nothing to do with it. Event c possesses M₁, but M₁ is in no way responsible for c's causing e. Here is a more accurate depiction of the situation:

M₁		M₂
⇑		⇑
P₁	→	P₂

The “causally relevant” properties are P₁ and P₂. M₁ and M₂ are along for the ride. M₁ and M₂ appear epiphenomenal. This is what Brian McLaughlin calls “type epiphenomenalism”: mental types, mental properties, add nothing by way of causal efficacy to events possessing them (McLaughlin 1989).

Davidson's problem has evolved into a problem for “nonreductive physicalism” generally. Suppose, inspired by Davidson, you thought that all the properties distinctive of the “special sciences” that figure in causal explanations supervene on more basic physical properties. That is, not just psychological properties, but biological properties, meteorological properties, geological properties, not to mention sociological properties, and economic properties, depend on, but are distinct from, fundamental physical properties. It looks as though all such “higher-level,” supervenient properties would be preempted or “screened off” in causal interactions.

This is unacceptable. It seems clear that mental properties play a role in what we do and think, a causal role. Even if you were prepared to concede the inefficacy of mental properties, you are much less likely to doubt that biological properties, or meteorological properties, or geological properties “make a causal difference.” Special science explanations are causal through and through.

One option might be to weaken Davidson's requirements for causation. Perhaps causal sequences need not be backed by strict, exceptionless laws, but only by “hedged” ceteris paribus laws (Fodor 1974). These laws would preside over causal relations within restricted domains. Laws of physics are exceptionless because there is nothing “outside” physics to disrupt their operation. Physics is unique in this regard. Psychological laws, in contrast, are subject to nonpsychological interventions.

To see the point, imagine that it is a psychological law that, if you want something and believe that some action, A, would enable you to get what you wanted, you will form an intention to A, and subsequently A. Now imagine that you have the requisite desire and belief, but, before you have a chance to form the intention, you are distracted by a loud noise and, as a result, the intention is never formed. Alternatively, you might form the intention, but fail to act because you are struck on the head by a falling tree branch and lose consciousness.

This is how it would be for any special science law. Consider biology. The dinosaurs were on track to evolve in particular ways, but their biological milieu was disrupted by a meteor's striking the Earth and changing the atmosphere in ways that proved catastrophic. Biological laws must be “hedged.” No one doubts that biologists identify genuine biological causes, however, no one doubts that specifically biological properties can be “causally relevant.”

The situation in fundamental physics is completely different. When the behavior of a particle is affected by other particles, the whole system of particles is governed by exceptionless laws. The particles are not subject to outside interference because, at the fundamental level, there is no “outside.”

Maybe Fodor is right, all we need are “hedged,” ceteris paribus laws to underwrite causal interactions. The trouble with this suggestion is that it threatens to introduce a kind of systematic causal “overdetermination.” The point can be appreciated by considering again c's causing e, where c is P₁, e is P₂, and M₁ and M₂ supervene on P₁ and P₂. Suppose you thought that c's being P₁ caused e's being P₂ and that c's being M₁ caused e's being M₂.

M₁	→	M₂
⇑		⇑
P₁	→	P₂

Presumably, the law governing P₁ → P₂ is strict, and the law governing M₁ → M₂ is a “hedged,” ceteris paribus law. But now ask yourself why M₂ is on the scene. M₂, recall, supervenes on P₂, so P₂ suffices for the presence of M₂. In fact, it looks as though the only way M₁ could influence the production of M₂ is by bringing about P₂.

M₁		M₂
⇑		⇑
P₁	→	P₂

This suggests that P₂ has not one cause, but two: P₁ and M₁. That, however, flies in the face of the thought that P₁ → P₂ is itself a “strict deterministic law.” The “strictness” of this law makes it hard to see how M₁ could have a role in producing P₂, hence how M₁ could be thought responsible for M₂. It appears that we must choose between “systematic overdetermination” and “type epiphenomenalism.”

M₁		M₂
⇑		⇑
P₁	→	P₂

If neither option appeals to you, you might want to give up on the claim that mental properties – or any other “higher-level,” supervenient properties you regard as causally efficacious – are irreducible. Kim argues that in many cases in the special sciences, this is the most promising option (Kim 2005). Where reduction is not feasible, we must accept epiphenomenalism. At the present time, the philosophical community is divided on the matter. Some accept Kim-style reduction, others embrace epiphenomenalism, and still others are happy to allow widespread systematic “overdetermination.”

4. Another Look at Davidson

The preceding discussion presumes a particular reading of Davidson, a reading not obviously supported by the text and implausible on the face of it. In this section I shall provide an alternative reading that has at least two advantages: first, it fits with what Davidson actually says in “Mental Events”; second, it avoids the kinds of difficulty I have been discussing.

On the received reading, Davidson accepts the “token identity” of mental and physical events (every mental event is identical with some physical event), but rejects mental–physical “type identity,” where types are taken to be properties. This commits Davidson to property dualism: mental properties are distinct from physical properties. The supervenience of the mental on the physical yields the dependence of mental properties on physical properties. Distinctness and dependence, coupled with the idea that causation requires “strict deterministic laws” and the denial that there are any “strict deterministic laws on the basis of which mental events can be predicted or explained” seems to guarantee the causal irrelevance of mental properties.

The received reading is fundamentally mistaken, however. The key to understanding Davidson is to recognize that, for Davidson, the mental–physical distinction is not ontologically deep. The mental–physical distinction is one of conception only, not a “real distinction,” not a distinction in reality. Davidson says repeatedly that an event is mental or physical “only as described.” Mental–physical supervenience is the doctrine that any event that can be given a description in a mental vocabulary could be described in a physical vocabulary. This ensures token identity: if an event answers to a mental description, there is some physical description to which it, the very same event, would answer.

Davidson admits that it is not easy to say what makes a description mental.

We may call those verbs mental that express propositional attitudes like believing, intending, desiring, hoping, knowing, perceiving, noticing, remembering. Such verbs are characterized by the fact that they sometimes feature in sentences with subjects that refer to persons, and are completed by embedded sentences in which the usual rules of substitution appear to break down.

(Davidson 1970: 210)

You might think that this characterization is wholly inadequate. This would be worrisome had Davidson set out to give an analysis of the mental. But this is not his aim. Let mental descriptions be whatever you like. Make a list of mental terms and use the list as a basis for an account of what makes a description mental. Davidson's idea is that whenever an event is identified by means of a term on the list, the event would answer to a physical description.

So what makes a description physical? Let a physical description be a description couched in terms of fundamental physics. Now the claim would be that whenever an event can be picked out using a mental vocabulary, that very event could just as well be picked out using predicates at home in fundamental physics. Thus, token identity: every mental event, every event describable using a mental vocabulary could, as well, be described in terms of fundamental physics.

Although Davidson does not identify the physical vocabulary as I have done, I believe that invoking fundamental physics captures the spirit of Davidson's approach. In addition, this makes it easier to understand his claim that causal sequences are governed by “strict, deterministic laws.” Such laws are what you get in fundamental physics. Thus, whenever events are related causally, there is some description of the events in fundamental physical terms that includes the application of a strict, exceptionless, deterministic law. Although there are typically many ways to describe a causal sequence, fundamental physics takes us to the causal nexus.

Davidson says that while every event is physical, every event answers to a physical description, and some events answer to mental descriptions. This suggests that some events, indeed most events, are not mentally describable, hence would not count as mental events. You might worry that Davidson's technique for identifying an event as mental would lead to the exclusion of an important class of mental events: experiences of pains and afterimages, sensations and feelings of dread, what E.C. Tolman called “raw feels” (Tolman 1932: 426–7).

This might be a concern were Davidson aiming to provide an analysis of mental concepts. But this is not on the agenda.

The criterion actually covers not only the havings of pains and after-images, but much more besides. Take some event one would intuitively accept as physical, let's say the collision of two stars in distant space. There must be a purely physical predicate “Px” true of this collision and of others, but true of only this one at the time it occurred. This particular time, though, may be pinpointed as the same time that Jones notices that a pencil starts to roll across his desk. The collision has now been picked out by a mental description and must be counted as a mental event.

(Davidson 1970: 211)

Davidson goes on to note that

This strategy will probably work to show every event to be mental; we have obviously failed to capture the intuitive concept of the mental. It would be instructive to try to mend this trouble, but it is not necessary for present purposes. We can afford Spinozistic extravagance with the mental since accidental inclusions can only strengthen the hypothesis that all mental events are identical with physical events. What would matter would be failure to include bona fide mental events, but of this there seems to be no danger.

(Davidson 1970: 212)

Do those philosophers who think Davidson distinguishes mental and physical events on the basis of distinctive mental and physical properties really think that Davidson would be sanguine about the possibility that every event has a mental property? What these passages illustrate is Davidson's rejection of the idea that the mental–physical distinction is ontologically deep. The distinction, rather, is not for Davidson, as it was not for Spinoza but was for Descartes, a real distinction. The distinction is one of conception only.

What of “type diversity”? What of Davidson's robust antireductionism? There is no question of mental properties reducing, or failing to reduce, to physical properties. The idea, rather, is that categories and taxonomies of psychology and physics are orthogonal. This means, in part, that there is no hope of expressing application conditions for psychological terms in fundamental physical terms, no hope of eliminating psychological categories and replacing them with categories of fundamental physics.

Many philosophers have pointed out that what Davidson says about the autonomy of psychology extends to all the special sciences. Take biology. Any biological event, any event describable in biological terms must be describable as well by means of categories of fundamental physics. Biological categories, no less than psychological categories, will be orthogonal to these. Here is how Jerry Fodor puts it.

Physics develops the taxonomy of its subject-matter which best suits its purposes: the formulation of exceptionless laws which are basic. … But this is not the only taxonomy which may be required if the purposes of science in general are to be served: e.g., if we are to state such true, counterfactual supporting generalizations as there are to state. So, there are special sciences, with their specialized taxonomies, in the business of stating some of these generalizations. If science is to be unified, then all such taxonomies must apply to the same things. If physics is to be basic science, then each of these things had better be a physical thing. But it is not further required that the taxonomies which the special sciences employ must themselves reduce to the taxonomy of physics. It is not required, and it is probably not true.

(Fodor 1974: 114)

Some readers will be surprised to find Fodor lumped with Davidson. But both Fodor and Davidson are antireductionists, and both regard reduction as a matter of replacing categories (or predicates, or taxonomies) in one science with those in another, presumably more basic, science.

Philosophers sometimes talk as though reduction is a relation among properties or families of properties. But what would it mean to reduce one property to another? The mistake underlying “nonreductive physicalism” is to start with warranted premises concerning predicates and categories, ways of representing real divisions in the world, and draw from these unwarranted ontological conclusions. The fact that biological or psychological categories are not reducible to, or replaceable with, categories belonging to fundamental physics, does not imply that there are realms of “irreducible” biological and psychological properties. What follows is that there are truths about the world expressible in biological or psychological vocabularies that have no systematic counterparts in physics.

Plato described science as being in the business of carving the world at its joints. The mistake is to imagine that there is just one correct way of doing this. On the contrary, the world includes endless joints, endless divisions, and endless opportunities for classifying. We have little interest in most of these. Some joints, however, are salient, some reflect important causal and qualitative similarities. These form the basis of our various taxonomies. Now think of fundamental physics as endeavoring to discover joints that themselves have no joints.

Here is one way to read Davidson. Psychological truths, truths expressible in a psychological vocabulary, are made true by ways the world is that could be given an exhaustive description in the language of fundamental physics. Physics, fundamental physics, gives us the deep story about the truthmakers for every truth that has a worldly truthmaker. Because psychological (biological, meteorological, geological, &c) predicates are not replaceable by predicates of fundamental physics, reduction is not in the cards.

Unthinking reliance on Quine's “criterion of ontological commitment” is responsible for much of the confusion here.⁵ Linguistic, classificatory differences are turned into ontological differences. In fact, what we get from Quine is not an ontological inventory, but an accounting of all the truths to which we are committed. It is an open question what the truthmakers are for these truths.

How could you know the truths without knowing the truthmakers? The truthmaker for “snow is white” is snow's being white. So, when you know that “snow is white” is true, do you not know the truthmaker? It is one thing to be able to say what the truthmaker is, one thing to identify the truthmaker, however, and another matter altogether to know its nature, to know the deep story, for instance, about snow and snow's whiteness.

What is snow? What is snow's whiteness? You can know that this is snow and that it is white without knowing what makes this snow or what makes snow white. Snow could be an arrangement of corpuscles, a fluctuation in the quantum field, a particular kind of thickening in space–time. The ontology of snow is not revealed by our ways of talking about snow. To know snow's ontology, you will need to know the physics and chemistry of snow.

5. Causation and Mental Causation

Suppose I am right about how to read Davidson. In that case, problems associated with Davidson's position, most particularly the problem of “causal relevance,” recede. Consider the following combination of theses.

(1) Mental events are events possessing mental properties and physical events are events possessing physical properties.

(2) Although mental properties depend on physical properties, mental and physical properties are distinct.

(3) Every mental event (every event possessing a mental property) is identical with a physical event (an event possessing a physical property).

(4) Events' causally interacting is a matter of those events falling under some strict law.

(5) Only physical properties figure in strict laws, so events fall under strict laws solely in virtue of their physical properties.

This combination of theses, mistakenly ascribed to Davidson, is thought to generate the problem of “causal relevance.” Mental events can have physical effects, but not in virtue of being mental, not qua mental.

I take (1)–(3) to be central tenets of “nonreductive physicalism.” If you accept, as well, (4) and (5), if you accept (1)–(5), you are going to have trouble explaining how mental properties could be “causally relevant” in the production of physical effects – and, given the dependence of the mental on the physical, on the production of mental effects as well. This is why nonreductivists are so often willing to jettison (4) and (5). Some, Fodor, for instance, accept (5) but reject (4), holding that strict laws are not required to back causal relations. Others, Sydney Shoemaker, for instance, have advanced intricate accounts of the mental–physical dependence relation in hopes of defusing the problem (Shoemaker 1998, 2001, 2007, 2011).

What all of these attempts have in common is a commitment to a strong ontological construal of reduction according to which the nonreducibility of the mental to the physical is explicated, as in (2), by reference mental and physical properties. This places property dualism at the heart of antireductionism. Dualism is the source of the problem of causal relevance, but it is nonnegotiable.

Suppose, however, Davidson – my Davidson – is right, suppose the mental–physical distinction is not a real distinction, but only a distinction in conception. In that case, it would make no sense to wonder whether an event caused a particular effect qua mental or qua physical. An event is mental or physical “only as described,” and how you choose to describe an event has nothing to do with what it causes.

Does this toss the baby out with the bathwater? Does it merely pretend to solve the problem of “causal relevance” while in reality eliminating or downgrading the mental? You would think that only if you thought that psychological truths required distinctively mental properties as truthmakers. But why think that? Davidson allows all the mental truths – and all the biological truths, all the meteorological truths, and all the sociological truths. Truthmakers for these will be particular ways the world is. The world really is those ways. If our aim is to provide the deep story of the nature of the truthmakers, we must turn to fundamental physics. This is rarely our aim, however; we are rarely interested in the deep story concerning the ontology of the world as we find it.

One ongoing source of confusion stems from the fact that, in some philosophical idiolects, “property” functions as a synonym for “predicate.” In that case, to say that a possesses the property of being F, is just to say that a is F. If that sounds strange, consider that what makes it true that a is F might not be the possession, by a, of a property, F. This billiard ball is red. Does this mean that there is a property, being red, possessed by the billiard ball? The billiard ball is a determinate shade of red, crimson, say. Pretend that this determinate shade of red is a property. Does the billiard ball possess two properties, the property of being crimson and, in addition, the property of being red? Or is it rather that it is true to describe the billiard ball as red (or, for that matter, as colored) because it is a determinate shade of red? You would think the former only if you thought that distinct truths require distinct truthmakers. But this is just what Davidson denies. Every mental truth is made true by some way the world is, a way that also makes true some truth expressible in a very different vocabulary, the vocabulary of fundamental physics.

There is no crime in taking a relaxed view of properties, no crime in equating properties and predicates. Mistakes arise, however, when premises invoking properties in this relaxed sense are used to support conclusions concerning properties in an ontologically serious sense. This, arguably, is the mistake at the heart of “nonreductive physicalism.” But it is not Davidson's mistake.

1 Davidson (1963: 3). References to this paper and to Davidson (1970) are to the more easily accessible reprints in Davidson (1980) and Davidson (2001).

2 See Anscombe (1957), Dray (1957), Hampshire (1959), Hart and Honoré (1959), Kenny (1963), Melden (1961), and Ryle (1949).

3 This will be true for ordinary physical events as well. Exceptionless laws govern only the fundamental things. Ordinary physical objects supervene on distributions of the fundamental things. Events involving ordinary physical objects figure in causal laws only to the extent that they are decomposable into events involving the fundamental things, fundamental events.

4 Philosophers who made this kind of argument early on include Honderich (1982), Kim (1984) , Sosa (1984), and Stoutland (1980). Many more followed.

5 You are “committed to the existence” of whatever you ineliminably quantify over in those theories you accept (Quine 1948).

References

Anscombe, G.E.M. (1957). Intention, 2nd ed. Oxford: Basil Blackwell., p. 1963.

Dancy, J. (2000). Practical Reality. Oxford: Oxford University Press.

Davidson, D. (1963). Actions, reasons, and causes. The Journal of Philosophy 60:685–700. Reprinted in Davidson (1980): 3–19.

———. (1970). Mental events. In Experience and Theory, L. Foster and J. Swanson (eds). Amherst: University of Massachusetts Press, pp. 79–101. Reprinted in Davidson (1980/2001): 207–225 (citations are to the latter).

———. (1980). Essays on Actions and Events. Oxford: Clarendon Press. Reprinted in 2001.

Dray, W. (1957). Laws and Explanation in History. Oxford: Clarendon Press.

Fodor, J.A. (1974). Special sciences (or: the disunity of science as a working hypothesis). Synthese 28:97–115.

Foster, L. and Swanson, J. (eds). (1970). Experience and Theory. Amherst: University of Massachusetts Press.

Gillett, C. and Loewer, B. (eds). (2001). Physicalism and Its Discontents. Cambridge, UK: Cambridge University Press.

Hampshire, S. (1959). Thought and Action. New York: Viking Press.

Hare, R.M. (1952). The Language of Morals. Oxford: Oxford University Press.

Hart, H.L.A. and Honoré, A.M. (1959). Causation and the Law. Oxford: Clarendon Press.

Honderich, T. (1982). The argument for anomalous monism. Analysis 42:59–64.

Kenny, A. (1963). Action, Emotion, and Will. London: Routledge and Kegan Paul.

Kim, J. (1984). Epiphenomenal and supervenient causation. Midwest Studies in Philosophy 9:257–270. Reprinted in Kim (1993): 92–108.

———. (1989). The myth of non-reductive materialism. Proceedings of the American Philosophical Association 63:31–47. Reprinted in Kim (1993): 265–84 (citations are to the latter).

———. (1993). Supervenience and Mind: Selected Philosophical Essays. Cambridge, UK: Cambridge University Press.

———. (2005). Physicalism, or Something Near Enough. Princeton, NJ: Princeton University Press.

Lowe, E.J. (2008). Personal Agency: The Metaphysics of Mind and Action. Oxford: Oxford University Press.

Malcolm, N. (1968). The conceivability of mechanism. Philosophical Review 77:45–72.

McLaughlin, B.P. (1989). Type epiphenomenalism, type dualism, and the causal priority of the physical. Philosophical Perspectives 3:109–135.

Melden, A.I. (1961). Free Action. London: Routledge and Kegan Paul.

Quine, W.V.O. (1948). On what there is. Review of Metaphysics 5:21–38. Reprinted in Quine (1961): 1–19.

———. (1961). From a Logical Point of View, 2nd ed. Cambridge, MA: Harvard University Press. Reprinted in 1963 by Harper and Row.

Ryle, G. (1949). The Concept of Mind. London: Hutchinson.

Shoemaker, S. (1998). Causal and metaphysical necessity. Pacific Philosophical Quarterly 79:59–77.

———. (2001). Realization and mental causation. In Gillett and Loewer (2001): 74–98.

———. (2007). Physical Realization. Oxford: Oxford University Press.

———. (2011). Realization, powers, and property identity. The Monist 94:3–18.

Sosa, E. (1984). Mind-body interaction and supervenient causation. Midwest Studies in Philosophy 9:271–281.

Stoutland, F. (1980). Oblique causation and reasons for action. Synthese 43:351–367.

Tolman, E.C. (1932). Purposive Behavior in Animals and Men. New York: The Century Co.

1. Actions, Reasons, and Causes

2. Mental Causes

3. Difficulties With This Account

4. Another Look at Davidson

5. Causation and Mental Causation

6. Concluding Remarks