Fate, Time, and Language: An Essay on Free Will by David Foster Wallace
Such reasoning, it should be noted, is made possible by the fact that, for any two events or states of affairs A and B, if A is a necessary condition of B then B is a sufficient condition of A, and vice versa, and it matters not at all which of them occurs first in time.
“Being able” and “Knowing how.” Taylor has been criticized for equivocating in the use of the term “can.” John Turk Saunders, for example, suggests that Taylor confuses logical impossibility with “not having the power to.” The presupposition that no agent can perform any action in the absence of some condition necessary for its accomplishment expresses, according to Saunders, only a certain innocuous logical impossibility, and thus has nothing to do with what any agent is able to do. He argues thatMy knocking upon a thin wooden door with my fist is a sufficient condition for the door’s shaking. Hence the door’s shaking is a necessary condition for my knocking upon the door. But the door’s shaking is not a necessary condition for my ability to knock upon the door.4
Taylor, however, did not argue that no agent can know how to perform some act in the absence of some condition necessary for its accomplishment, and thus, in that sense, does not have the ability to perform it. His point was, rather, that no matter what an agent might know how to do, he still cannot even do what he knows how to do (and is in that sense able to do) if there is lacking some condition necessary for doing it.
For example, imagine an expert pole-vaulter locked in a room with an eight-foot ceiling. Both Taylor and Saunders might agree that an expert pole-vaulter has the know-how or technical expertise to pole-vault twelve feet. In this sense of the word “can” the pole-vaulter can pole-vault twelve feet. What Taylor is asserting is that, given the conditions of the locked room, the pole-vaulter does not have it within his power to pole-vault twelve feet. His know-how is constrained by circumstances that prevent him from exercising it.
Bruce Aune makes a similar point. Aune claims that Taylor’s presupposition that no agent can perform any given action if there is lacking some condition necessary for the accomplishment of that action, has “absurd consequences.” As an example he suggests the following:If a man should say that he can swim, or that he has the ability to swim, he would surely take it as a poor joke if someone replied, “No, you cannot swim: you lack the ability to do this because you are not now in a pool or lake.”5
To this we should reply that if a man should say that he can swim, or that he has the ability to swim, he would surely take it as a poor joke if someone said, “Well then, you can swim under any conditions. Let’s see you swim out of water.” The first of these “jokes” is not a joke at all, since, in one sense, it simply states an obvious truth, namely, that one needs water in order to swim, and it is precisely in this sense that Taylor utilizes it. The second “joke” is indeed a joke, since it assumes that when a man says that he can swim, he means that he can swim at a specific time even if conditions necessary for his swimming then are lacking. Of course, no man takes this to be the meaning of the sentence “I can swim.” What one means is that he is able to swim at any specific time if all other conditions necessary for his swimming are then present, and such conditions obviously include the presence of water. Again, Taylor’s use of “can” seems perfectly legitimate and is not a distortion of common usage.
The Simple Rejection of Fatalism. Some critics have, in effect, pointed out that Taylor’s arguments lead to fatalism.
Saunders, for example, dismisses it as strange that “my mere ability to knock upon the door will suffice to make it shake.” This, however, is simply part of the fatalist position. I cannot perform a given act if there is lacking a condition necessary for doing it, no matter what I might know how to do, and this does indeed imply, as Saunders points out, that if I can knock on the door then I shall. However strange this may seem, it is only because fatalism is strange, and it is hardly a criticism that Taylor’s argument, which purports to yield a fatalist conclusion, does yield such a conclusion.
Peter Makepeace’s comments are in some respects similar. Like Saunders, he appears to allow Taylor’s argument while disputing claims not made by Taylor. He agrees that “I cannot make something happen in the future if it is not going to happen.” But this is just Taylor’s conclusion. It is logically equivalent to saying that if it is true that a certain event E is not going to happen, then I cannot make it happen—to which we can add that, if it is false that the event in question is not going to happen, then I cannot prevent it from happening. And this is fatalism.
What, then, does Makepeace dispute? He introduces the following example:If conditions are such that a snowfall yesterday is a necessary condition for the lawn’s being snow-covered this morning, then, given that no snowfall occurred, we can conclude not only that the lawn is not snow-covered, but that it cannot be.6
He then claims that we ought not to speak of the lawn’s state of being snow-covered as not being “within its power,” and that it is “absurd” to add that this “is consistent with its being able to carry snow, having the ability not to melt it, and so on, and thus being able, in that sense.”
This, however, is not absurd; it is only an odd choice of words. Makepeace is rightly reluctant to use “within its power” in connection with inanimate objects. But this is a minor point. If he wishes to change the examples from animate to inanimate objects, Taylor can change the expression “within his power” to “within its capability.” Now, we find Makepeace repeating Saunders’s error, by disputing what Taylor has not claimed. Taylor has not claimed that the lawn does not possess the capability to hold snow (i.e., to carry snow, not to melt it, and so on), any more than he claims that the pole-vaulter does not have the know-how to pole-vault twelve feet in a room with an eight-foot high ceiling.
What Taylor can rightly claim is that, given the absence of a necessary condition, the lawn does not have it within its capability to be snow-covered, just as the pole-vaulter does not have it within his power to pole-vault twelve feet. And this, it would appear, is just what Makepeace admits when he concludes “I cannot make something happen in the future if it is not going to happen.” Certainly, if I can’t do it, neither can a lawn.
Fatalism and Linguistic Reform. In a subsequent criticism Saunders accuses Taylor of redefining “within one’s power” while still employing it in its usual contexts. It is this “linguistic reform,” he claims, that accounts for the seeming fatalistic conclusion of Taylor’s argument. Taylor, according to Saunders, treats it as analytic that “the only events which it is within one’s power to produce are those which occur.”
Taylor does not treat this statement as analytic. It does, however, follow from his argument, and it leads to the conclusion that the only actions one is able to perform are those which he does perform—which is, again, the conclusion of fatalism.
Does this, however, amount to a linguistic reform? It seems not. Consider a violinist, for instance, who has forgotten to bring his violin to his recital and is unable to obtain another before the time of the recital. What Taylor is not asserting is that this violinist could not play the violin at his recital even if he had a violin in his hands. Such an assertion would be patently false. What he is asserting is that if at the time of the recital the violinist does not have a violin to play, then he cannot at that time present a recital, for he cannot play an imaginary violin. This statement, in contrast to the previous one, is obviously true, and in a perfectly ordinary sense of “cannot.”
Taylor admits that there is another sense to the word “can” which he does not utilize. This is the notion of know-how. There is a sense of the word “can” such that it is true that the violinist without his violin can still play the violin, since he knows how to. Taylor does not use this sense of “can,” however, since if this sense were to be utilized, fatalism with respect to the past would also be shown to be false.
Assume, for instance, that a sufficient condition of my having gone to a lecture yesterday is my having my own notes from it. Suppose that yesterda
What Taylor has done is to disregard that meaning of “can” with respect to the future also. He would claim that, if it is true that I will not go to the lecture today, then I cannot perform any act sufficient for my attending it, and this is consistent with my knowing how to walk to the lecture hall, find a seat there, and so on. Saunders’s seemingly plausible claim that one can sometimes do something sufficient for the future occurrence of what is not going to happen is in fact no more reasonable than the absurd claim that one can sometimes do something sufficient for the past occurrence of what did not happen.
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.
In a reply Taylor suggests that if Saunders’s argument does indeed refute fatalism in respect to the future, then it also refutes fatalism in respect to the past.7 Saunders denies this and asserts:... if the non-occurrence of an event in the future does not entail my lack of power to bring about that event, then neither does the non-occurrence of an event in the past entail my lack of power to bring about that event ... but it is not due to the non-occurrence of an event in the past that I lack the power to bring about that event. I have no such power because we so use our language that it is false or nonsense to say that one has the power to bring about any event whatever in the past.8
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.
Fatalism and Causation. It has also been suggested that Taylor confuses causally necessary conditions with logical necessity. Raziel Abelson, for example, suggests that, if the states of affairs described in Taylor’s argument are not logically related, they must be causally related. This is not quite correct, however, since Taylor expresses his argument entirely in terms of necessary and sufficient conditions which, unlike causal conditions, involve no temporal relations at all. If, for instance, the presence of oxygen is a necessary condition of a certain man’s being alive over a given period of time, then that man’s continuing to live over that period is a sufficient condition for there being oxygen present. But neither of these is logically necessary or sufficient for the other, nor is either the cause of the other. The presence of oxygen may be a causal condition of that man’s continuing to live, but certainly his living is no causal condition for the presence of oxygen—even though it is a sufficient condition for the presence of oxygen.
Aune, on the other hand, criticizes Taylor for excluding logically necessary and sufficient conditions from his examples and for formulating his argument entirely in terms of what Aune labels “physically” necessary and sufficient conditions. He points out that logical necessity implies physical necessity and that the introduction of logical necessity into the argument has a damaging effect upon it.
He is quite right in noting that Taylor has chosen to deal only with “physically” necessary or sufficient conditions which are not also logically necessary or sufficient conditions, but not right in suggesting that Taylor denies that logical necessity implies “physical” necessity. Taylor takes no stand on this, which is, in fact, not relevant to his argument. Likewise irrelevant is the claim that his argument, if expressed in terms of logically necessary or sufficient conditions, leads to the abolishment of all modal distinctions. It is no more to the point to criticize Taylor’s arguments for ignoring logically necessary or sufficient conditions than to criticize it for ignoring causal conditions, for the question is not whether other, more or less similar arguments yield a fatalistic conclusion, but whether Taylor’s argument does.
The nonefficacy of time. Abelson also claims that Taylor’s assumption that time is not efficacious is ambiguous, since, he says, time is logically efficacious. Here Abelson seems simply to misunderstand Taylor’s notion of the efficacy of time. Taylor explains this by noting that the mere passage of time does not augment or diminish the powers or capacities of anything. Abelson, however, seems to equate the sentence “Time is logically efficacious” with the sentence “Time often has a lot to do with the truth of what we say.” But these two sentences are entirely different. For instance, the sentence “It is now raining” may be true today and false tomorrow. Quite obviously, time has a lot to do with the truth of the sentence. But it is not time which augmented the power of the clouds to produce rain. Certain meteorological conditions did that. Time in this sense is not efficacious.
Aune also criticizes this assumption, but somewhat differently. He notes that time cannot pass without something changing. This is doubtless true, but it has nothing to do with Taylor’s assumption, which says only that the passage of time “has no causal effect upon anything.” Perhaps something must change during any period of time, but it is not time which causes such change. A lake, for example, is dried up, not by time, but by certain meteorological conditions or by emptying the lake. This happens in time, to be sure, but time by itself is no cause of it.
The Scope of Modal Concepts. Several critics have suggested that Taylor has simply misplaced certain modal concepts, which is a fairly common fallacy. Taylor’s crucial assumption, for example, is that no agent can perform any action in the absence of some condition necessary for its accomplishment. But, according to these critics, all this really means is that it (logically) cannot be the case that an agent does perform an action in the absence of some condition necessary for its accomplishment—which is perfectly compatible with saying that he can perform such an action.
Thus Abelson accuses Taylor of committing a fallacy which “lies at the root of the famous paradox of Chrysippus: a man necessarily does X or does not do X (excluded middle). Therefore either he necessarily does X or he necessarily does not do X.” In his argument the necessity of the logical truth of the first statement is illicitly transferred from the entire disjunction to the individual disjuncts.
However, it appears that Abelson has committed a similar error in reverse. Taylor’s argument can be interpreted, with certain qualifications, as saying that if A, then necessarily B, and if ~B, then necessarily ~A. Abelson transfers these individual necessities to the necessity of the entire proposition “If A implies B, and ~B, then ~A.” This proposition is logically true, as is the first statement of the Chrysippus paradox. But whereas that paradox asserts a second statement which does not logically follow from the first, Abelson denies a second statement which, in fact, is the premise from which his first statement is deduced. What Abelson does is to transfer the necessity of two individual implications to the necessity of a logical truth which follows from these two individual implications. He is not logically in error in doing so, but then he refuses to acknowledge the necessity of the two individual implications, since they do n
Aune also criticizes Taylor’s use of modal concepts, but somewhat differently. He says that if Taylor’s crucial assumption, to the effect that no agent can perform any action in the absence of some condition necessary for its accomplishment, is “taken as a necessary truth” or “a result of logical analysis,” then the statements (i) “he performs A,” (ii) “he can perform A,” and (iii) “he has to perform A” are all logically equivalent. All that follows from Taylor’s assumption, however is that these statements are extensionally equivalent, not that they are logically equivalent. To assert their extensional equivalence, however, is only to assert fatalism, which does indeed follow from Taylor’s argument. If I can perform A, then I do, in fact, perform A, and, moreover, I must. This is a strange conclusion only if one happens to reject fatalism. To point out that it is strange is only to reject the conclusion; it is not to refute it.
[II]
Taylor, like many others, has suggested that one could avoid fatalism by rejecting the law of excluded middle in regard to certain statements about the future and allowing that some of these might be neither true nor false.
I shall present a slightly altered version of Taylor’s argument which does not utilize such statements and which is, therefore, unaffected by any modification of the law of excluded middle. This variation of his argument presupposes the same assumptions that he makes, and can be set forth in terms of the same example, namely, that of a naval commander (NC) about to issue one of two orders, one of which (O), will ensure a naval battle the following day, and the other of which (O′) will ensure that no naval battle occurs the following day.