Acts

1. The Identity Thesis

What is an act? One of the problems concerning the nature of acts is the problem of individuation. Suppose that John does each of the following things (all at the same time): (1) he moves his hand, (2) he frightens away a fly, (3) he moves his queen to king-knight-seven, (4) he checkmates his opponent, (5) he gives his opponent a heart attack, and (6) he wins his first chess game ever. Has John here performed six acts? Or has he only performed one act, of which six different descriptions have been given? Again, suppose that John (1) moves his finger, (2) pulls the trigger, (3) fires the gun, and (4) kills Smith. Are there four distinct acts that John has performed, or are all of these one and the same act? An answer to such questions will provide a partial answer to the question of the nature of acts.

A straightforward answer to these questions has been proposed by G. E. M. Anscombe and defended in some detail by Donald Davidson. On the Anscombe-Davidson view there is but one act that John has performed in each of the two cases. On their view John's moving his hand is identical with John's moving his queen to king-knight-seven; John's moving his hand is identical with John's checkmating his opponent; etc. There is but one act here, which has been described in a variety of ways. Miss Anscombe writes:

... a single action can have many different descriptions, e.g. "sawing a plank," "sawing oak," "sawing one of Smith's planks," "making a squeaky noise with the saw," "making a great deal of sawdust," and so on and so on. ... ... Are we to say that the man who (intentionally) moves his arm, operates the pump, replenishes the water supply, poisons the inhabitants, is performing four actions? Or only one? ... In short, the only distinct action of his that is in question is this one, A. For moving his arm up and down with his fingers round the pump handle is, in these circumstances, operating the pump; and, in these circumstances, it is replenishing the house water-supply; and, in these circumstances, it is poisoning the household.

So there is one action with four descriptions....

Similarly, Davidson writes:

I flip the switch, turn on the light, and illuminate the room. Unbeknownst to me I also alert a prowler to the fact that I am home. Here I do not do four things, but only one, of which four descriptions have been given.

And in another article Davidson writes:

But what is the relation between my pointing the gun and pulling the trigger, and my shooting the victim? The natural and, I think, correct answer is that the relation is that of identity.

I shall call this thesis the identity thesis. It has been espoused not only by Anscombe and Davidson, but also by D. S. Shwayder and Eric D'Arcy.

The identity thesis constitutes one clear and attractive way of individuating acts. But there are several difficulties it encounters. In general, if X and Y are identical, then X must have all and only the properties that Y has. We shall find, however, that some of the pairs of acts which are alleged to be identical do not share all the same properties.

Let us look, for example, at John's pulling the trigger and John's killing Smith, which were mentioned in one of the examples given above. According to the identity thesis, these acts are supposed to be one and the same act. But are they really identical? Consider the act of John's killing Smith and consider the event consisting in the gun's going off. Is it true to say that this act caused this event, that John's killing Smith caused the gun to fire? Surely not. It would be extremely odd to say that John's killing Smith caused the gun to go off. But now consider John's act of pulling the trigger. It is certainly true of this act that it caused the event in question, i.e., that it caused the gun to fire. Thus, John's pulling the trigger has the property of causing the gun to fire, whereas John's killing Smith does not have the property of causing the gun to fire. However, since one of these acts has a property which the other lacks, they cannot be one and the same act.

Let us take a slightly different example to illustrate the same point. Suppose that John is playing the piano, and that his playing causes Smith to fall asleep while also causing Brown, who was already asleep, to wake up. John has performed the following acts: (1) he has played the piano, (2) he has put Smith to sleep, and (3) he has awakened Brown. According to the identity thesis, John's playing the piano = John's putting Smith to sleep = John's awakening Brown. But are these genuine identities? Consider the following two events: (e1) Smith's falling asleep and (e2) Brown's waking up. Ex hypothesi, both of these events were caused by John's playing the piano. Now let us compare John's playing the piano with John's awakening Brown. Clearly, while John's playing the piano caused (e1), Smith's falling asleep, John's awakening Brown did not cause (e1). Similarly, compare John's playing the piano with John's putting Smith to sleep. John's playing the piano did cause (e2), Brown's waking up, while John's putting Smith to sleep did not cause (e2). We see, then, that John's act of playing the piano has a property which is lacked by John's act of putting Smith to sleep and has another property which is lacked by John's act of awakening Brown. Hence, John's playing the piano cannot be identical with John's putting Smith to sleep and cannot be identical with John's awakening Brown.

These two objections to the identity thesis make reference to the effects or consequences of two allegedly identical actions. Further objections can be raised to the identity thesis by examining the causes or causal factors of actions. If A and A' are one and the same action, then they are one and the same event. And if they are the same event, one would expect them, if they are caused at all, to be caused by the same set of events or states of affairs. If we find, to the contrary, that A and A' have somewhat different sets of causes or causal factors, that would give us reason to conclude that A and A' are not the same after all,

John answers the phone and says "hello." He says "hello" because he wishes to greet the caller. But John has just been quarreling with his wife and is in a tense emotional state. As a result, he says "hello" very loudly. He doesn't intend to shout over the phone; it just comes out that way. Now consider the following acts: John's saying "hello" and John's saying "hello" loudly. According to the identity thesis, these acts are one and the same. However, they appear to have different sets of causal factors. John's act of saying "hello" loudly is an effect, at least in part, of his being in a tense emotional state. But John's act of saying "hello" (simpliciter) is not at all an effect of this emotional state, since John would have said "hello" whether or not he had been angry or tense. Thus, there is a causal factor of John's saying "hello" loudly that is not a causal factor of John's saying "hello."

Let us turn next to a slightly different example concerning the causes, or causal conditions, of actions. Suppose there is a light bulb missing from a certain socket and that George fetches a bulb and screws it into the socket. A moment later John comes along, flips the switch and thereby turns on the light. Now consider John's act of flipping the switch and John's act of turning on the light. If we tried to list all the causes or relevant causal factors of John's turning on the light, we would certainly include George's screwing the bulb into the socket. If George had not screwed the bulb into the socket, then John would not have succeeded in turning on the light. He would still have flipped the switch, perhaps, but this would not have resulted in the light's going on. Thus, George's screwing the bulb into the socket is a cause, or causal condition, of John's turning on the light. On the other hand, George's action is not at all a cause, or causal factor, of John's act of flipping the switch. For, as I am imagining the case, John would have performed this act whether or not George had screwed the bulb into the socket.

If the reader finds it a bit odd to say that George's screwing the bulb into the socket is a "cause" of John's turning on the light, we might say instead that John's act of turning on the light was "enabled," or "made possible," by George's screwing the bulb into the socket. We may then add that John's act of flipping the switch was not "enabled," or "made possible," by George's activity, since John's success in flipping the switch was quite independent of George's contribution. Thus, John's act of turning on the light has at least one property that is lacked by John's act of flipping the switch.

I turn now to a rather different sort of case in which there is a failure of two putatively identical acts to coexemplify a given property. In this case the property in question is the property of being supererogatory. Suppose that I owe Smith two dollars. Seeing him on the street, I reach into my pocket for some cash, and discover two single dollar bills and one two-dollar bill. I like to collect two-dollar bills myself, but I recall that Smith simply goes wild over them. Bearing this in mind, I pay Smith the money with the two-dollar bill. Now consider my act of giving Smith the two-dollar bill and my act of repaying Smith two dollars. The former is supererogatory, while the latter is not supererogatory in the least. Acccording to the identity thesis, these acts are identical. But since one of them has a property which the other lacks, they cannot be one and the same act.

We have seen that many pairs of acts which are alleged to be identical fail to satisfy the principle of the indiscernibility of identicals: they fail to have all properties in common. A further difficulty confronting the identity thesis stems from a certain relationship that holds between many of these acts which are said to be identical. I wish to turn now to this relationship and to its consequences for the identity thesis.

We often say of a person that he performs one act "by" performing another. We say, for example, that John turns on the light "by" flipping the switch, or that he checkmates his opponent "by" moving his queen to king-knight-seven. As used in these contexts, the term "by" expresses a relationship that holds between acts, between John's act of flipping the switch and his act of turning on the light, and between John's act of moving his queen to king-knight-seven and his act of checkmating his opponent. The relationship in question might be expressed by saying that the one act is a "way" or "method" by which the other act is performed. Typically, when act A is the "way" by which act A' is performed, we can explain how act A' has been performed by citing act A. For example, if John checkmates his opponent by moving his queen to king-knight-seven, we can explain how he checkmated his opponent by referring to his act of moving his queen to king-knight-seven.

The important point to notice about this relationship is that it is both asymmetric and irreflexive. Consider first the matter of asymmetry. If agent S does act A' "by" doing act A, then he does not, do A "by" doing A'. John turns on the light by flipping the switch, but he does not flip the switch by turning on the light. He checkmates his opponent by moving his queen to king-knight-seven, but he does not move his queen to king-knight-seven by checkmating his opponent. We can explain how John turned on the light by indicating that he flipped the switch, and we can explain how John checkmated his opponent by saying that he moved his queen to king-knight-seven. But we cannot explain how John flipped the switch by saying that he turned on the light; nor can we explain how John moved his queen to king-knight-seven by saying that he checkmated his opponent.

The irreflexivity of the relationship can be seen in the same examples. We would not say that John turned on the light by turning on the light, nor that John checkmated his opponent by checkmating his opponent. We cannot explain how John flipped the switch by indicating that he flipped the switch; nor can we explain how John moved his queen to king-knight-seven by saying that he moved his queen to king-knight-seven.

The fact that the relationship in question is asymmetric and irreflexive has important consequences for the identity thesis. If A and A' are identical, there can be no asymmetric or irreflexive relation which one bears to the other. If A and A' are genuinely identical, then if a relation R holds of the ordered pair (A, A') it must also hold of the ordered pair (A', A). And if R holds of the ordered pair (A, A'), it must also hold of the ordered pairs (A, A) and (A', A'). But we have seen that there is a relation that holds between the ordered pair (John's moving his queen to king-knight-seven, John's checkmating his opponent) which does not hold of the ordered pair (John's checkmating his opponent, John's moving his queen to king-knight-seven); nor does it hold of the ordered pair (John's moving his queen to king-knight-seven, John's moving his queen to king-knight-seven) or of the ordered pair (John's checkmating his opponent, John's checkmating his opponent).

My final criticism of the identity thesis concerns the contrast between basic actions and non-basic actions. Along with many other philosophers, I am inclined to think that some of our actions are basic actions and that other actions of ours are not basic actions. Moving my hand is a basic action, whereas checkmating my opponent and turning on the light are not basic actions. Rather, they are actions I perform by performing some basic actions. Now if the identity thesis is correct, then the distinction between basic actions and non-basic actions must be abandoned. For if John's moving his hand is a basic action, and if it is identical with John's checkmating his opponent, then his checkmating his opponent is also a basic action. This consequence of the identity thesis has been acknowledged by Davidson, who does not find it unacceptable. But I, for one, would regard this as an undesirable consequence.

Having suggested a variety of difficulties for the identity thesis, let us consider how a proponent of the identity thesis might reply to these difficulties.

One possible reply would involve an appeal to referential opacity (or intensionality). It is well known that certain phrases, such as "desire," "believe," "necessarily," etc. create contexts in which expected substitutions cannot be made salva veritate. The truth of "John believes that Cicero denounced Catiline" does not ensure the truth of "John believes that Tully denounced Catiline," even though Tully is Cicero. And the truth of "Necessarily nine is greater than seven" does not ensure the truth of "Necessarily the number of planets is greater than seven," although nine is the number of the planets. A proponent of the identity thesis might contend that the phrases used in posing our objections also create referentially opaque contexts, viz. such phrases as "is the cause of," "is an effect of," "is supererogatory," etc. For example, it might be suggested that the context "... is the cause of the firing of the gun" is referentially opaque, and because of this the co-referential phrases "John's pulling the trigger" and "John's killing Smith" cannot be substituted salva veritate. The non-substitutivity, according to the identity theorist, is not to be accounted for by denying the identity of John's pulling the trigger and John's killing Smith, but rather by observing that "... is the cause of the firing of the gun" is an opaque context.

The first point I want to make is that there is a danger here of proliferating opaque contexts. Antecedently, it does not seem very plausible that all of the contexts, "... is the cause of X," "... is an effect of Y," "... is supererogatory," and "... is done by —," are referentially opaque. Thus, the conclusion that they are all referentially opaque should not be drawn unless this opacity can really be proved.

But can the proponent of the identity thesis prove referential opacity in these cases? I think not. The proof of referential opacity could only be accomplished by begging the question. In order to prove the opacity of the belief context in the Cicero-Tully case, it must be granted antecedently that Cicero is identical with Tully. In this case, the identity presents no difficulty, of course. But unless this identity is granted, it cannot be proved that the context in question precludes the substitutivity of identicals. Similarly, in order to prove the opacity of such contexts as "... is the cause of the firing of the gun," we must have antecedent agreement on the question of whether John's pulling the trigger is identical with John's killing Smith. But here there is a difficulty, for whether or not these are identical is the very point at issue. To presuppose that they are identical is to beg the question.