As A. N. Prior has pointed out, the necessity of the logical truth “X or not-X” is illicitly transferred from the entire disjunction to the individual disjuncts.
Returning now to Professor Taylor’s crucial deduction, it seems to me that Taylor performs just such an illicit transfer of modality in his first premise (If Q occurs then not-O is impossible and if not-Q occurs then O is impossible). Taylor’s equivocal P2-P5 allow him to shift from the innocuous logical truth “If O implies Q, and not-Q, then necessarily not-O” to the dubious premise of his argument: If not-Q occurs then O is impossible. If Taylor were right to make this shift, then the nonoccurrence of a sea battle tomorrow would indeed deprive the admiral today of his power to order it. But it would also follow that, since every event is a necessary and sufficient condition for itself, the nonoccurrence of an event would render that event impossible while the occurrence would render it necessary. Thus when logical and causal modalities are conflated, necessity becomes equivalent to truth and impossibility to falsehood. From this artificial and vacuous interpretation of modal expressions, fatalism follows as an obvious corollary. We are destined to conclude that fatalism is a consequence not of the law of excluded middle nor of the inefficacy of time, but of a misuse of modal language which it is clearly within our power not to perpetrate.
NOTES
1 R. Taylor, “Fatalism,” Philosophical Review, LXXI (1962), 56-66.
2 As Taylor indicates, P2 and P3 are simply definitions of necessary and sufficient conditions.
3 P5 is the real villain of this drama. According to P5, I am unable to drink more coffee than I shall in fact drink; I am powerless to spank my child more often than he will provoke me to spank him. According to Taylor, our present powers are as dependent on future events as on past and present conditions.
4 Op. cit., 57-58.
5 For ease of later reference, I have abbreviated Taylor’s argument and I have used “impossible” in place of “not within (my) power to do.” The argument is on page 61 of Taylor’s essay.
6 As Taylor employs these terms they are like logical modalities in being atemporal and like causal modalities in being nonanalytic. A. H. Prior employs a similarly hybrid modality (“S”), in his system of modal logic, following Lukasiewicz, but he recognizes its artificiality and handles it with care. Cf. A. N. Prior, Formal Logic (Oxford, 1955), p. 247.
7 Taylor’s own language, in his illustrations of P5, preserves the proper temporal direction for causality and thus belies his repudiation of temporal relevance. Viz., “I cannot, for example ... swim five miles without ever having been in water, or read a given page of print without having learned Russian” (my italics). Op. cit., 58.
8 Cf. Prior, op. cit., p. 210.
8
A NOTE ON FATALISM
RICHARD TAYLOR
MY QUALIFIED argument for fatalism drew some impassioned protest. 1 Such arguments have disturbed philosophers since the days of St. Augustine and Boethius; since, in fact, men first took seriously the idea of divine omniscience and, with this, the idea that all truth is timeless. Few nowadays consider divine omniscience the corner-stone of all that we cherish, but the suggestion that some propositions about the future may be as yet not true and as yet not false, but will in time be made true or made false by men’s acts, is generally received as though it were an attack upon reason itself.
The fundamental error made by all my critics consists of treating some or all the expressions “It is within my power to do E,” “It is possible that I shall do E,” and “I am able to do E” as synonymous. They clearly are not. It is often possible that I shall do something that it is not within my power to do—for example, inherit a fortune. It is sometimes possible that I shall do something that I am unable to do—for example, compose a good sonnet. And I am often able to do something which, in the absence of some condition necessary for doing it, it is not within my power to do—for example, I am able to play the piano, but it is not within my power to do so in a room that contains none. It is this last confusion that is at the root of all our troubles.
THE PHILOSOPHICAL REVIEW, VOL. 72, NO.4, 1963.
All three critics point out that one is often able to do something, even in the absence of some condition—past, present, or future—necessary for its accomplishment. Thus, one might be perfectly able to swim, even though he is not near a lake, or able to lift a weight, even though he does not exert the necessary effort, or able to administer more spanks to his child than he will be provoked to administer, and so on. This is all perfectly true. Indeed, to assert otherwise would be to involve oneself in a sort of absurdity that can be elicited as follows.
Let us suppose, to use my own example, that my issuing a certain order is, under prevailing conditions, sufficient for the occurrence of a naval battle tomorrow. In that case, the occurrence of that battle tomorrow is a necessary condition of my prior issuance of that order. But it cannot be a necessary condition of my being able to issue that order—for if it were, then it would follow that my mere ability to issue such an order would be sufficient for the occurrence of that battle, even without my actually issuing it! This would be an absurd thing to say. All one can conclude, in case no such battle occurs tomorrow, is that I did not issue that order—not that I was unable to.
This is but a variation of Boethius’ answer to fatalism,2 which has satisfied countless thinkers since then who have felt bound to uphold God’s omniscience or the law of excluded middle, or both. To see what is wrong with that answer, we need only note that the very same rejoinder can be made against the claim that we should be fatalists about the past—which claim, however, is correct.
Thus, to use my other example, suppose that the occurrence of a naval battle yesterday is, under prevailing circumstances, a necessary condition of my reading a certain kind of headline today. In that case, my reading such a headline today is sufficient for the prior occurrence of such a battle. But that battle cannot be a necessary condition of my being able to read such a headline—for if it were, then it would follow that my mere ability to read such a headline would guarantee the prior occurrence of that battle, whether I actually read it or not. And this, too, would be absurd.
But now note that, while all this is true, it is beside the point, for it does not refute fatalism with respect to the past. Although I may be ever so skilled at reading headlines, and thus have the ability to do so in that usual sense of ability, that fact does not render it within my power to read a headline that never exists and the nonexistence of which is ensured by what has occurred. If it were otherwise, then it would be within my power to adjust the past to my liking, which it most certainly is not. Similarly, though I may be ever so skilled at issuing orders, and thus have the ability to do so in that usual sense of ability, that fact does not render it within my power to issue an order that never exists and the nonexistence of which is ensured by what will occur. Here our hesitancy does not concern the legitimacy of this illation, which is identical to the plainly correct one that precedes it, but rather the legitimacy of applying the law of excluded middle to the future in exactly the way we apply it to the past (tertium non datur).
Whatever may be my abilities, it is no more within my power to exercise those 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. And surely we have the same reason for saying that, whatever may be my abilities, it is certainly not within my power to exercise those abilities in the absence of some future condition necessary for doing what they enable me to do, and thus to determine the future. We can avoid the fatalism to which that thought points only by denying the truth of such a future contingency, unless and until it becomes true—or, better, until it is made true, in case it ever is, by what someone does in the meantime.
Remarks upon what one may or may not have the ability to do, in the usual skill sense of ability, have no relevance t
o this problem at all. Not one of my critics has seen this. Nor have they seen that the very refutations they give of my fatalism about the future would work just as well to prove that we should not be fatalists about the past. I described a fatalist, however, simply as a man who looks upon the future the way we all look upon the past, so far as concerns what it is and what it is not within his power to do. If anyone wanted to show that we should not be fatalists about the past, that it is to some extent now up to us what happens yesterday, and so on, he could find all his arguments in the remarks of my critics, needing only to change a few tenses.
NOTES
1 Bruce Aune, “Fatalism and Professor Taylor,” Philosophical Review, LXXI (1962), 512-519; John Turk Saunders, “Professor Taylor on Fatalism,” Analysis, XXIII (1962), 1-3; Raziel Abelson, “Taylor’s Fatal Fallacy,” Philosophical Review, LXXII (1963), 93-96.
2 The Consolation of Philosophy, Book V.
9
TAUTOLOGY AND FATALISM
RICHARD SHARVY
RICHARD TAYLOR has given an argument for fatalism1 which I shall summarize briefly. (a) No agent has the power to perform an act if a necessary condition for the performance of that act is lacking; (b) if X is a sufficient condition for Y, then Y is a necessary condition for X, and vice versa. Then, on the hypothesis that my performing an act O is a sufficient condition for the occurrence of an event Q, and that my performing an act O′ is a sufficient condition for the nonoccurrence of this event (call this nonoccurrence the event Q′), Taylor argues as follows: (c) the occurrence of Q is a necessary condition for my performing the act O, and the occurrence of Q′ is a necessary condition for my performing the act O′; (d) either Q or Q′ must fail to occur, and thus (e) either a necessary condition for my performing the act O is lacking, or a necessary condition for my performing the act O′ is lacking, and hence (f) either act O is not within my power or act O′ is not within my power.
Now the first objection to this argument that comes to mind is that Taylor seems to suppose that a necessary condition for an agent’s performing an act is a necessary condition for his having the power to perform that act.2 This would mean that an agent’s merely having the power to perform an act would be sufficient for the occurrence of that act. Another way of putting this objection is Bruce Aune’s observation 3 that Taylor’s uncritical use of premises a and b “leads to the abolishment of all modal distinctions”; i.e., that on Taylor’s use of these premises... the following statements must all be logically equivalent: (i) he performs A; (ii) he can perform A, or it is physically possible for him to perform A; and (iii) he has to, or physically must, perform A.
THE JOURNAL OF PHILOSOPHY, VOL. 61, NO. 19, 1964.
This is claimed to be erroneous, and the fallacy of Taylor’s proof of fatalism
Now I will agree with Taylor’s critics that Taylor is collapsing modalities in such a way that “He performs A” becomes logically equivalent to “He has the power to perform A.” However, I am not convinced that this observation can serve as a refutation of Taylor’s argument, since asserting that the above equivalence does not hold is, in a way, begging the question. After all, a fatalist is just a person who says that we do all and only those things which are within our power. For the fatalist, it is a contradiction to say “He could have done it, but he didn’t”; so, for the fatalist, “He performs A” is equivalent to “He has the power to perform A.” Thus, to say that the fatalist abolishes modal distinctions is simply to say what a fatalist is—he is a person who abolishes modal distinctions. Furthermore, these criticisms of Taylor’s article do not seem to have convinced Taylor that he is in error.4
I do think, however, that Taylor’s fatalistic conclusions can be analyzed away without rejecting his use of modal language, but by accepting this and seeing where it leads. First of all, I would like to make a suggestion about the motivation of the particular form of the argument that Taylor presents. It is not very satisfying to the fatalist to be able to say “John did not have the power to do X” only after John has in fact failed to do X. It would be more satisfying to the fatalist if he could give a statement about a future event which he could know, in the present, to be true. Thus Taylor takes as one of his premises a statement “about the future” which can be known in the present—a tautology(!). The statement is d above, that either Q or Q′ must fail to occur. His conclusion is f, that either I do not have the power to perform O, or I do not have the power to perform O′. An equivalent way of saying this is that I do not have both the power to perform O and the power to perform O′.
Now at first glance, this conclusion may indeed appear to be fatalistic. However, because of the collapse of modal distinctions that is built into fatalism, this conclusion turns out to be trivial and not at all fatalistic. The reason for this is that, with the “abolishment of modal distinctions,” the distinction we ordinarily make between (1) “I have both the power to do X and the power to do Y” and (2) “I have the power to do both X and Y” collapses. 1 and 2 are equivalent, because 1 is equivalent to “I do X and I do Y,” which is identical in meaning with “I do both X and Y,” which is equivalent to 2.
But this means that Taylor’s conclusion f is equivalent to “I do not have the power to do both O and O′,” and there is nothing fatalistic about this, since by hypothesis O and O′ have incompatible consequences. There is certainly nothing fatalistic about a statement like “I necessarily lack the power to (effectively) order a sea battle both to occur and not to occur.” This may be the reason Taylor thinks that freedom and the law of the excluded middle are incompatible, but I do not think that this tautologous sort of fatalism, i.e., the assertion that we lack the power to perform logically incompatible acts, can seriously be considered to be any limitation of our freedom.
NOTES
1 “Fatalism,” Philosophical Review, 71 (1962): 56-66.
2 Cf. John Turk Saunders, “Professor Taylor on Fatalism,” Analysis, 23, 1 (October, 1962).
3 “Fatalism and Professor Taylor,” Philosophical Review, 71 (1962): 517.
4 Cf. his reply to his critics, Philosophic Review, 72 (1963): 497.
10
FATALISTIC ARGUMENTS
STEVEN CAHN
OVER THE past several years much controversial literature has appeared on the subject of fatalism. Richard Taylor’s fatalistic arguments, 1 in particular, have drawn critical comment from many quarters,2 and A. J. Ayer in his latest book has added new fuel to the controversy.3
In what follows I shall concentrate on the crucial points in Taylor’s argument and on replies that have been made to it. In part I I shall show that these replies are inconclusive. Indeed, some of them amount to nothing more than pointing out that Taylor’s premises have fatalistic implications, which was precisely what Taylor was suggesting. In part II I shall set forth a demonstrative argument to show that, contrary to what Taylor and many others have supposed, these fatalistic implications cannot be avoided, even by modifying the law of excluded middle.
It should be emphasized at the outset, however, that Taylor’s article is not an argument for fatalism as such. Rather, it claims that certain assumptions adopted almost universally in contemporary philosophy yield a proof of fatalism. Taylor leaves it an open question whether to accept fatalism or make adjustments in these assumptions. He himself suggests modifying some of these assumptions, and, in particular, the law of excluded middle.
THE JOURNAL OF PHILOSOPHY, VOL. 61, NO. 10, 1964.
[I]
The Crucial Assumptions. One of Taylor’s crucial assumptions is that any proposition, in whatever tense, is either true or, if not true, then false—the law of excluded middle. Another is that no agent can perform any action in the absence of some conditions necessary for its accomplishment. By “a necessary condition for X” Taylor means some event or state of affairs which is logically unrelated to X but which is nevertheless such that X cannot occur without it. It is in this sense, for instance, that oxygen is necessary for human life. No ma
n can live without oxygen, though it is not logically impossible to do so.
Thus, given that it is true that a certain event E has occurred, then it is not now within the power of any agent to perform any action A which would, if performed, be sufficient for the nonoccurrence of E—for a necessary condition for the performance of A is the prior nonoccurrence of E. It is this idea that is expressed in the remark, “the past is unalterable,” and it seems quite uncontroversial. But Taylor applies the same reasoning to the future. Thus, given that it is true that a certain event E will occur, then, he says in effect, it is not now within the power of any agent to perform any action A which would, if performed, be sufficient for the nonoccurrence of E—for a necessary condition for the performance of A is the subsequent nonoccurrence of E. This is fatalism, and it is controversial indeed.
Fate, Time, and Language Page 9
