and(2) The water is not turbulent at t4
that(3) It is not in my power to swim at t3
Suppose that we have empirically established not only (1) and (2), but also the following:(4) At t3I have the ability (know-how) to swim
(5) Conditions for the exercise of my swimming ability are “normal” at t2and t3 (meaning that I have not been bound, drugged, or hypnotized; I am in swimable waters, etc.)
If we use language in the ordinary ways, then the falsity of (3) follows from (4) and (5), and, since the above argument is valid and premises (1) and (2) are true, it follows that the only remaining premise, PF, is false. Thus we prove that, if language is used in its ordinary ways, which involves PF’s being a synthetic statement, then PF is false.
4. Departing from Ordinary Language, Taylor Treats PF as Analytic. But it is precisely Taylor’s purpose to argue that, in just such a situation as I have described, (3) must be true because PF is true. And it is because he clings to this position that I am forced to conclude that, unbeknownst to him, PF does not, for Taylor, have the status of a synthetic statement: he has, all unwittingly, departed from ordinary language by conferring upon PF the status of an analytic statement.
It seems that Taylor will not accept this description of his stand. He apparently maintains that PF is a synthetic statement. 7 Notice, however, how he argues for the truth of PF. He tells us that we have the same reason for accepting PF that we have for accepting PP. And he says regarding PP that “it is no more within my power to exercise ... [my] abilities in the absence of some past condition necessary for doing what they enable me to do, and thus to determine the past, than for the finest pianist to exercise his virtuosity in a room that contains no piano.”8 The alternative is that “it would be within my power to adjust the past to my liking, which it most certainly is not” (loc. cit.). “No one doubts that fatalism with respect to the past is true, i.e., that we have no power to make happen what did not in fact happen. My argument showed that we have the same reasons for saying it is true with respect to the future, given the usual interpretation of the law of excluded middle.”9 But it is analytic that it is not in the power of a pianist to play a piano without a piano. It is analytic that it is not in one’s power to adjust (change, alter) the past. [The latter is analytic on two counts: (a) it is analytic that it is not in one’s power to bring about (produce, cause) past events; (b) it is analytic that it is not in one’s power to change the past (or, for that matter, the future), when changing it is taken to entail (the nonsense) that a certain portion of the past (or the future) is Φ at tn at one time but non-Φ at tn at some other time.] Thus Taylor argues for the truth of PF in just the way that one argues for the analytic truth of a statement. Nor, from all that he has said, is there any reason to believe that Taylor would be willing to describe any logically possible situation whose existence would falsify PF. I can only conclude that he has unwittingly “reformed” ordinary language by making PF analytic. And it is this which is responsible for the strange (“fatalistic”) results at which he arrives.
5. Some Features of Ordinary Language that Taylor Would “Reform.” Let us now utilize once more our aquatic example in order to bring out some of the features of the ordinary use of ‘power’ (in the sense of an agent’s power to perform an act) and related terms and to note how Taylor would “reform” these features of our language.
5a. Free will and compulsion. Suppose that statements (1), (2), (4), and (5) of Section 3 are true. Suppose also that the following two statements are true:(6) At t2I decide not to swim at t3
(7) My deciding at t2 not to swim at t3 is a sufficient condition of my not swimming at t3
We so use our language that it follows from the truth of (4), (5), (6), and (7) that I refrain from swimming at t3 of my own free will: I could swim at t3 if I wished, but I decide not to. Taylor, however, is committed to saying that I do not refrain from swimming of my own free will, but am compelled not to swim by my own decision not to swim. For from (6) and (7) it follows that a necessary condition of my swimming at t3 (viz., my not deciding at t2 not to swim at t3) is lacking, and from this plus P it follows that it is not in my power to swim at t3. Note that it is the PP portion of P which is here responsible for the departure from ordinary language. Taylor assumes that everyone accepts PP, and he bases much of his case for PF upon this assumption. But, as the foregoing clearly demonstrates, we do not accept PP. The acceptance of PP would oblige us to allow that whenever our desires, deliberations, decisions, etc. are sufficient conditions of our subsequent acts, there are no other acts in our power to perform, and, hence, that the acts we do perform are not acts of our own free will. They are acts we are compelled to perform (or not perform as the case may be). On Taylor’s view, then, we arrive at the nonsense that our own decisions deprive us of free will.
In ordinary language it is analytic that some past necessary conditions of an act are such that their absence renders one powerless to perform the act, whereas the absence of others does not. As we have seen, my not deciding at t2 not to swim at t3 is a necessary condition of my swimming at t3 whose absence does not render me powerless to swim at t3. If, on the other hand, I had been securely bound, hand and foot, at t2 or had assimilated a potent knock-out drug at t2, etc., then I would indeed have been rendered powerless to swim at t3 by the absence of past necessary conditions of swimming at t3. But notice that the absence of these conditions falsifies (5): conditions for the exercise of my swimming ability are, on this supposition, not normal at t2 and t3.
5b. Ability. The latter fact reveals another respect in which Taylor departs from ordinary usage. The term ‘ability’ (in the sense of an agent’s knowing how to perform acts of a certain kind) is so used in ordinary language that, by definition, if, for example, one has the ability to swim, then nothing but the absence of a normal condition for the exercise of swimming ability renders one powerless to swim. Thus, being drugged, bound, etc. destroy my power to swim, although deciding not to swim does not, nor does the subsequent nonturbulence of the water. In accepting P, Taylor is obliged to say that the absence of any necessary condition whatever of my swimming at t3 deprives me of the power to do so, whether it be the absence of water turbulence at t4 or my decision at t2 not to swim at t3 (i.e., the absence of my not deciding at t2 not to swim at t3). But what meaning remains to ‘ability’ when we allow that one is able to swim who is nevertheless very often powerless to do so under perfectly normal conditions for the exercise of this ability? I am not sure what ‘ability’ then means, but that this is not the ordinary sense of ‘ability’ is obvious.
5c. Taylor’s reasons for accepting PP and PF. We have seen both that Taylor is mistaken in thinking that we all accept PP and that its acceptance would deviate from our ordinary linguistic principles. He maintains that the reasons that oblige us to accept PP also oblige us to accept PF. What are these reasons? That it is impossible to bring about (produce, cause) a past event which does not occur. And, indeed, this is impossible simply because our language is such that (i) ‘bringing about a past event’ makes no sense, and (ii) ‘bringing about a nonoccurrent event’—and in this way changing (altering, adjusting) the past (or any stretch of time)—also falls short of sense (cf. Section 4). But though it is true that this is impossible, it does not follow that PP is true: for (in Section 5a) we have found it to be false. It does not follow, because to deny PP—and thus say that someone has the power to perform an act a past necessary condition of which is lacking—is not to say that he has the power to bring about a past event and is not to say that he has the power to bring about a nonoccurrent past event. It is simply to say (a) that he has the power to perform an act sufficient for the occurrence of a past event, and (b) that the past event does not occur. (From this we may deduce that he does not perform the act, but not that it is not in his power to perform it.) Nor do reasons of the sort which Taylor provides for PP oblige us to accept PF. It does, of course, make sense to speak of bringing about a future event
. And it does not make sense to speak of bringing about a nonoccurrent future event (and in this way changing the future). But it does not follow that PF is true; indeed, we have seen reason (in Sections 3 and 5b) to believe that it is false. It does not follow because to deny PF and, thus, to say that someone has the power to perform an act a future necessary condition of which is lacking, is not to say that he has the power to bring about a nonoccurrent future event. It is simply to say (a) that he has the power to perform an act sufficient for the occurrence of a future event, and (b) that the future event does not occur. (From this we may deduce that he does not perform the act, not that it is not in his power to perform it.)
One can easily imagine Taylor (and Cahn) arguing that I simply beg the question regarding PF by assuming that mine, and not theirs, is the ordinary sense of ‘power’ (or ‘can’). Witness Cahn:Taylor has not engaged in linguistic reform. Rather, he has utilized one sense of ‘can’ which, in regard to the past, is consistent with everyone’s use of that word. What he has tried to show is that this sense ought to be just as consistent with everyone’s use of the word in regard to the future, though this is not the case, since people are not aware of their limitations with respect to the future but are aware of these limitations with respect to the past.10
But we have seen that people do not use Taylor’s sense of ‘power’ (or ‘can’) with respect to the past, i.e., that we do not accept PP. Further, to admit, as Cahn does here, that “people are not aware of their limitations with respect to the future” and so (it seems tacitly granted) do not use Taylor’s sense of ‘power’ with respect to the future, i.e., do not accept PF, is to admit what Cahn began by denying in the first quoted sentence: Taylor has indeed engaged in linguistic reform. Surely, in calling our attention to our “limitations with respect to the future,” Taylor has not played the scientist and provided us with empirical evidence for the hypothesis PF. Rather he has called our attention to the empirical fact of a similarity between situations to which we do, and those to which we don’t, ordinarily apply ‘powerless,’ and has urged us to apply ‘powerless’ to all of them in light of the similarity. That similarity, of course, is the absence of necessary conditions of nonperformed acts: in light of the fact that a necessary condition of my swimming at t3 is absent both when I assimilate a knock-out drug at t2 and when the water is not turbulent at t4. Taylor invites us to say that either one of these alone renders me powerless to swim at t3.
5d. The direction of causality. Indeed, as the latter sentence reveals, it is not the opponents of Taylor, but Taylor himself, who is committed to the thesis that later situations may bring about (cause, produce) earlier situations. Since Taylor wishes to say that the calmness of the water at t4 renders me powerless to swim at t3, he presumably must say that the calmness of the water at t4 stops (prevents) my swimming at t3, that it causes me not to swim at t3. Here, again, Taylor departs from our ordinary modes of speech. For in ordinary language it is nonsense to say that a later situation brings about an earlier one.
A fortiori, it is nonsense to say that one has the power to bring about a past situation. Thus when we allow that a man has the power to perform an act sufficient for the occurrence of an earlier situation, we never say that he has the power to bring about that situation: if, for example, I swim today, and a necessary condition of my swimming today is that I have eaten in the past, then it is in my power to swim today, and thus it is in my power to do something today (swim) which is sufficient for the occurrence of an earlier situation (my having eaten); but we do not, on this account, say that it is in my power to bring about my previous eating. Taylor, however, is committed to saying of our example that it is in my power to bring about my previous eating. He is committed, in general, to saying that he who has the power to perform an act sufficient for the occurrence of an earlier situation has the power to bring about (cause, produce, make happen) that past situation. For, as we have seen, he maintains that PF is true for the same reason that PP is true: he charges that (B) if PP were false then (absurdly) someone would have the power to bring about a past situation that does not occur, and that the same absurdity militates against the falsity of PF. Thus we see that, because of the foregoing commitment, as well as that of the preceding paragraph, Taylor is obliged to modify our ordinary language by accepting a bidirectional causality, by speaking not only of earlier situations as causes of later ones, but also of later situations as causes of earlier ones.
In attempting to counter Taylor’s charge B, I have said that it is not because of the nonoccurrence of an event in the past that I lack the power to bring about that event, but because we so use our language that it makes no sense to say that one has the power to bring about any past event whatever (although it does make sense to say that one has the power to bring about a future event).11 Cahn replies:But this does not at all seem to answer Taylor’s charge. An expression possesses what meaning is conferred by its use. The question why it is used as it is still remains. Has it an arbitrary use? Or is there some actual difference between the past and the future which would account for making this distinction? If Saunders wishes to answer Taylor’s charge he must point out such a difference, for it is the denial of such a difference upon which Taylor’s argument essentially rests (300-301).
This I take to be a tacit admission that we do use our causal language in the way I have described and that Taylor is opting for a change in this language. I imagine that Cahn and Taylor are so dissatisfied with this sector of ordinary language that anything I might say in an effort to point out “some actual difference between the past and the future” that accounts for its being meaningful to talk of bringing about future, but not past, situations, would be met with charges of “superficiality” or “question-begging.” Nevertheless, I shall offer the following considerations.
We are not indifferent to the course of nature: we want some things to happen (and these we try to do or bring about); we want some things not to happen (and these we try not to do or to prevent). Suppose that one necessary condition of my swimming at t3 is my getting into the water at t2, and another is the water’s being turbulent at t4. Suppose, too, that I recognize that these are necessary conditions of my swimming at t3 and that at t1 I wish to swim at t3. Of course, then, I wish to satisfy both of these conditions, for both are necessary to that which I desire. But I realize that there is no point in waiting until t4, and then perhaps flaying the water with a stick, in order to satisfy the condition of the water’s being turbulent at t4. For my swimming at t3 is sufficient for the water’s being turbulent at t4, and it is the former I desire. The intelligent course is to concentrate on satisfying those necessary conditions of my swimming at t3 whose temporal status is prior to t3. Those necessary conditions of my swimming at t3 whose temporal status is posterior to t3 will be taken care of by my swimming at t3, if I do then swim, and so the best way to satisfy them is to satisfy those which are prior to t3. Ordinary language emphasizes these facts, and thus delineates the utilities of the situation; for we say: “Swimming at t3 brings about the water’s being turbulent at t4; the water’s being nonturbulent at t4 does not bring about my not swimming at t3.” For rather similar reasons, if at t1, I wish not to swim at t3, it would be intelligent to stay out of the water at t2 rather than wait until t4 and attempt then to calm the water. For it is swimming at t3 that I wish to avoid, and, since swimming at t3 is sufficient for turbulent water at t4, I shall not succeed in calming the water at t4 unless I avoid swimming at t3. The intelligent course is to concentrate on violating those necessary conditions of my swimming at t3 whose temporal status is prior to t3. Our ordinary ways of speech reflect these facts; for we say: “My not getting into the water at t2 brings it about that I do not swim at t3; the water’s being nonturbulent at t4 does not bring it about that I do not swim at t2.”
And what, it might be asked, if I wish at t6 to swim (or to have swum) at t3? Suppose that a necessary condition of my swimming at t3 is that the water is turbulent at t7 and th
at a sufficient condition of my swimming at t3 is that I am wet at t7. There is no point in trying to satisfy these conditions; if I swam at t3 the former condition will be satisfied (and it is swimming at t3 that I desire), and if I did not swim at t3 I will not succeed in satisfying the latter condition. In either case there is no intelligent course to pursue. Our ordinary language marks these truths; for we say: “Swimming at t3 brings it about that the water is turbulent at t7; the water’s being nonturbulent at t7 does not bring about my not swimming at t3; not swimming at t3 brings it about that I am not wet at t7; being wet at t7 does not bring it about that I swim at t3.”
In sum, it is useful to speak of earlier situations causing later ones, and not vice versa, because when we desire a situation to obtain in the future it is useful to concentrate upon satisfying those of its conditions (necessary and/or sufficient) which precede it, not those of its conditions which follow it in time; whereas it is not useful to concentrate upon satisfying any of its conditions when we desire that a situation obtain in the past. The very idea of an intelligent or useful thing to do in pursuit of a goal is the idea of earlier events being useful and later events being nonuseful with respect to a given goal. This difference with respect to utilities is an “actual difference between the past and the future” which our language is geared to reflect. In reflecting this difference our language serves our needs, and it is this service, I would imagine, which, in part at least, accounts for its being meaningful to talk of bringing about future, but not past, situations.
5e. Causality and power. Similar considerations indicate the utility of using ‘power’ in the ordinary fashion and the negative utility of using ‘power’ as Taylor would have us do. Returning to the foregoing aquatic example, if you wish at t1 to stop me from swimming at t3, there is no point in waiting until t4 and attempting then to calm the water. For if I swim at t3, you will not succeed in calming the water at t4. There is, however, point to your tying me up, or drugging me, at t2, given that these violate necessary conditions of my swimming at t3; for this will bring it about that I do not swim: indeed, this will bring it about in such a way that you will have stopped me from swimming. We find these facts reflected in our ordinary use of ‘power’; for we say: “Binding or drugging me at t2 destroys my power to swim at t3”; “The calmness of the water at t4 does not destroy my power to swim at t3.”