J. Brenton Stearns is Associate Professor of Philosophy at the University of Winnipeg.
The following article appeared in Process Studies, pp. 237-248, Vol. 6, Number 4, Winter, 1976. Process Studies is published quarterly by the Center for Process Studies, 1325 N. College Ave., Claremont, CA 91711. Used by permission. This material was prepared for Religion Online by Ted and Winnie Brock.
Determinism conflicts with the common sense understanding of time and is to be rejected on that ground, but determinism is not absurd. However, the issue resolved is of little systematic importance to process metaphysics. Process philosophers can maintain their critique of classical substance, of absolute idealism, of materialism, and of classical theology without having it rest on the rather flimsy structure of a “refutation” of determinism.
A doctrine that appears often in the process philosophies of this century holds that the causal determinist is contradicting himself when he talks about a fully determinate future. The idea seems to have originated with Henri Bergson. It also appears clearly in the context of a refutation of determinism in the writings of Milic Capek, Charles Hartshorne, and Paul Weiss, and at least the germs of it are in the work of A. N. Whitehead and William James. The low initial plausibility of the doctrine combined with the high importance of it in current process metaphysics makes an intriguing problem. Is this argumentative strategy effective against modern determinists?
The causal determinist maintains that the future relative to any moment is fully determinate at that moment and is predictable on the ground of natural regularities by any perfect knower of the initial conditions and relevant laws. That looks harmless enough with respect to a philosophy of time. But the theory I am to examine holds that if the determinist is really serious about what he is saying, he is committed to the view that tensed language must be dispensed with. If past, present, and future are equally definite (closed to alternative possibilities), then it makes no sense to say that an event is “now” definite, as distinguishable from “was” or “will be” definite. The determinist may reply that he is not at all sure about tenses, but he can surely distinguish between the earlier and later relative to any temporal moment. The effect still comes after the cause, even if both cause and effect are timelessly definite. That kind of reply, however, simply puzzles the process thinker, for if events do not really “happen”, no unique temporal sense can be given to “earlier” and “later.” Time is spatialized, and “earlier” and “later” become odd spatial directions, like “left” and “right.” Time is odd as a spatial dimension, because it is supposed to display features of asymmetry not dependent on the orientation of the observer, but the process thinkers cannot figure out in what the asymmetry is supposed to consist if determinism is true. The past seems like the future in all respects if the determinist is right. But then the temporal talk is about nothing at all. If time does not add anything to the definiteness of reality, time does nothing, and hence time is nothing. A determinate future is an actual future, an example of the absurdities that result when the determinist introduces temporal language. The point of all this for the process thinker is to show that determinism conflicts with a common sense understanding of time and is to be rejected on that ground.
Since there already exists an impressive body of literature on the relation between determinism and becoming, I want to provide a preliminary sketch of what I hope to contribute to it. I will reject the idea found in the process writings that there is a conceptual link between causal determinism and the “block universe” where consciousness, to borrow an apt phase from Richard Gale, sneaks up on (LT 235) and illumines what is tenselessly there. That may not seem to be a terribly exciting thesis to readers of Adolf Grünbaum, who stresses the independence of the determinism issue from the question of a metaphysics of becoming. However, what Grünbaum is primarily intent upon is showing that causal indeterminism is perfectly compatible with the theory that becoming is mind-dependent. Since the question of whether causal determinism requires his metaphysics of time is not at all a threat In his metaphysics, that issue does not attract his attention very much. When the process writers allege an entailment relation between causal determinism and the block universe, they do so not to defend their own metaphysics of time but to show that causal determinism is false. That issue is not in the forefront of Grünbaum’s mind. I think the process writers have some interesting arguments for their alleged entailment relation. I want to show wherein they go wrong and to show that other aspects of process metaphysics can be held independently of the attack on determinism. I shall assume that the process thinkers are right in holding that an adequate metaphysics of time must recognize dynamic becoming in nature as well as in human consciousness. The question then is whether time as the process philosophers think of it is incompatible with causal determinism. Recently Frederick Ferré discussed what forms of determinism Grünbaum’s metaphysics of time entails (1). Now let us begin from the other end and examine the allegation that causal determinism entails Grünbaum’s theory of time.
The various strands of current process thinking converge on Bergson, whose analysis of the durée is closely tied to a defense of a kind of creativity which is incompatible with determinism. Bergson was worried about the interpretation of relativity theory proposed by Minkowski in which time is a fourth dimension of space and in which the experience of temporal relations is analogous to shining a flashlight through the fourth dimension and finding the things that are “already there” (timelessly there). Bergson saw, I think rightly, that the spatialization of time loses time, except as an appearance to consciousness. Some form of determinism, but not necessarily a causal determinism in which the future is predictable in principle, is the result of such thinking. The future relative to any present would be timelessly determinate or definite.
The more troubling problem is whether causal determinism entails the spatialization of time. Bergson thought it does. This comes out clearly in both Time and Free Will and Creative Evolution where he says that to be a determinist is to hold, with Spinoza, that the future is logically contained in the present. To predict is to produce before the predicted thing is produced (CE 6). Since the conclusion of a deductive argument is timelessly contained in the premises and since predictions follow deductively from premises stating initial conditions, time is spatialized. That is, the future exists side-by-side with the present, and there is no radical becoming. If there is to be temporal process at all, time must gnaw at reality; time must make a difference (CE 46). The thesis is argued in the same way for our own generation by Milic Capek, who recognizes both its crucial importance for Bergson and the controversy it rightly inspires (PICT, BMP). And Paul Weiss in his Nature and Man, a book strikingly like Bergson’s Creative Evolution in the territory it covers and the positions it takes, introduces us early to the Bergsonian theory: “Things come to be. They are therefore free” (NM xxii).
Has anything gone awry? True, the lawlike generalizations and the initial condition statements used in the premises of a prediction are all true at the time of the initial conditions, and the predictions are timelessly deducible from them. However, the lawlike statements contain within them temporal distinctions, the concept of events taking place at different times. The only way we can maintain that the predicted happenings are “contained in” the present state of the universe described in the premises is to read back into the premises the dubious thesis at issue, that all events that are definite now are also actual now.
Milic Capek maintains in Bergson’s defense that the incompatibility of temporally subsequent states of affairs proves the falsity of determinism. The determinist deduces the future from the present. Since the conclusion of a valid deductive argument is always logically compatible with the premises, the determinist must hold that the temporally subsequent state of affairs is compatible with the temporally prior one. But this is clearly a mistake, for the world we predict is one that cannot be actual at the same time as the initial conditions. Hence, the determinist is committed to an absurdity, that the predicted state of affairs is deducible from the initial conditions yet is logically inconsistent with them (BMP 114).
Once again, we must note on the other hand that the laws that sustain the predictions make use of the concept of temporal succession. True, we should be worried if we produced the concept of time in the conclusion of a deduction where it is not present in the premises. But that does not happen here. And, as we shall see later in greater detail, the Bergsonian argument proves too much, for it entails that we should never be able to speak truly now of a future event that is incompatible with the present state of the universe.
In the writings of Charles Hartshorne we do not find the picturesque Bergsonism that has time gnawing at reality. The argument is cast in a somewhat different form, but it turns out to be a different route to the same destination. Look deeply enough into anything and you will, it may be, find its ancestors and its history inscribed upon its nature. But will you equally find its posterity and its destiny? This is the question of indeterminism, in one of its forms” (DR 68). Events are related to past particulars, and that allows us to find the ancestors of an event by examining it carefully. Presumably one will not find out everything about the event’s ancestors, but one will find out the causally relevant facts about them. However, particular events are not related to future particulars.
The reason there can be no relation to future particulars is that what has relation to X has X, for X is a constituent of relation-to-X. Now if the cause has the effect, by virtue of having relation to it, then in the cause the effect already is, and the whole time process is the illusion of new events, whereas all events were pre-contained in their predecessors. This reduces the idea of time to an absurdity. (DR 69)
But in what sense does the causal determinist find posterity inscribed upon the present? The minimum and important sense is that, if the relevant causal laws are known, a perfect examination of the present would yield perfect predictions. But according to the above passages, the causal determinist is supposed to believe not only that the future can be perfectly inferred from the present but also that the future already is present. There are two meanings of “inscribed” and the argument plays upon the ambiguity. In one sense of inscription, a perfect reading of the present provides a perfect reading of the future. The inferences one can validly make from a reading may be regarded as part of the reading. But in another sense of inscription, the future is already actualized in the present. Because the determinist can hold to the former sense of inscription and not the latter, the argument is incorrect.
What about the idea that a relation of a particular to a past particular is possible but to a future particular is impossible? By parity of reasoning, the impossibility of relations of particulars to past particulars can be shown. Suppose that Y, a particular effect, is related to X, a particular cause. X then is a constituent of Y-related-to-X. Therefore, X, by virtue of having a relation to Y, is contemporaneously present to Y. This means that all causes have to be contemporary with their effects. So there is no novelty, and everything is present at once. So there is no time.
The difficulty here is in supposing that both constituents of a dyadic relation have to be copresent contemporaneously. Let us deny that assumption. Then one can readily conceive a relation between an effect and a prior cause. Furthermore, one can also then readily conceive a relation between a cause and a subsequent effect. True, in the latter case the causal relationship is not yet fully actualized. Must we go on and say the relationship does not yet exist even in part, since there can be no dyadic relationship where one of the members related does not exist? But that would make it impossible for us to talk about a causal relationship between a present effect and a past cause, under the assumption that the past cause is now no longer in existence. We may be mistaken in supposing there is a symmetry here, but it would seem that if we can talk of a causal relationship between a present effect and an already perished past cause, we can also talk of a causal relationship between a present cause and a yet to be actualized effect.
Hartshorne would probably deny the symmetry. Although the past event that is inscribed on the present has perished, its memory remains, and that is the memory of a particular. But why, then, can we not maintain that if there are memories of particulars inscribed on particulars there can also be anticipations of particulars inscribed on particulars? I think the only way to deny the symmetry is to beg the question, that is to suppose that future particulars are not yet determined. The position then amounts to this: Relations to past particulars are possible because the past is now determinate and carried into the present in memory. Relations to future particulars are impossible because, although there are some anticipations inscribed on the present, the future is not now in every detail determinate. But this position gives no argumentative support to indeterminism, because it assumes it in the first place.
Not only does the Bergsonian thesis appear in current process philosophies, but it plays a central role in the process systems. Determinism is considered to be one of several if not ‘the chief metaphysical pitfall which must be ruled out for the process system to succeed. The philosophers I have cited do not simply believe determinists are wrong, that in view of the unpredictability of subatomic “particles” we must apply Occam’s razor and not admit determinism as a metaphysical doctrine. If that is all the objection were, determinists could be regarded as holding an implausible, unconfirmable, and scientifically unnecessary position. But the problem runs deeper than that for these process metaphysicians. For them, determinism must be soundly refuted, for if determinism is true, process metaphysics logically must be false. The Bergsonian argument is thus a two-edged sword for the process metaphysicians. On the one hand it is introduced as a powerful ploy against the determinists, provided the opponents want to regard the difference between “now” and “will be” as objective. But once this move against the determinists is made, the stakes become very high indeed. For if determinism should turn out to be true, hardly anything could be salvaged from process metaphysics. Every new advance in successful prediction would become falsifying evidence with respect to the entire process system of thought.
I have now commented critically on some presentations of the Bergsonian doctrine. Next I shall speak to some of the systematic considerations that bear on the attraction the doctrine holds for many thinkers. I see part of the difficulty as a confusion regarding what is involved in the “existence is not a property” idea. What Bergson, Weiss, and Hartshorne want to do is to deny that a given event can be determined or definite now but is to be actualized later. What are we to suppose that the actualization adds to the already determined and definite event? It does not add anything, because existence is not a property. Since actualization adds nothing to the determinate event, we are told we must suppose that an event’s actuality comes right along with its definiteness. Therefore, if the future is determinate now, it is also actualized now, and there is no time.
What difference does time make? If the determinist believed existence is a property, he might reply that time makes the difference between the existence and nonexistence of definite states of affairs. The future is fully determinate, but it does not exist yet, and that is an important “but.” An existing $100 is more than a nonexisting $100; otherwise why should we worry about bringing it into existence? And what does it matter whether the bringing of the $100 into existence is predictable in principle? One is wealthier if he has the $100 than if he does not have it, regardless of the truth or falsity of determinism.
At this point the disciple of Bergson can respond that the determinist’s case is really not aided one bit by taking refuge in the view that existence is a property. Let us try that view out. Suppose the future is fully determinate or definite and existence is a property. In specifying the future possibilities that will be actualized, one produces a list of properties. If the list is to be complete and if existence is a property, existence must be on the list. Then, the definiteness of future states of affairs that will be actualized must include existence. Suppose a determinist were to describe a future conflict between two nations which he believes is already causally definite. Since existence is a property, existence can and should be included in the list of properties which specify the already definite possibility. The already definite future conflict will be, among other things, an existing future conflict. But that is a contradiction in terms. If it is existing, it is not future; if it is future, it is not existing. So if determinism is true, the course of history is indeed timeless even under the hypothesis that existence is a property.
The dilemma is this: If existence is not a property and if the future is determinate, there is no difference between a future definite state of affairs and a present definite state of affairs, and time is unreal. If, on the other hand, existence is a property and the future is determinate, the definiteness of the future includes existence along with its other properties, the future is now, and time is unreal. This is the dilemma that gives force to the Bergsonian position. I think the determinist is best advised to grasp the first horn of the dilemma.
Although “existence” is in some contexts a grammatical predicate, I would say Kant is right in thinking it is not a determining predicate that can be used in specifying possible states of affairs, for that would play havoc with the distinction between actualized and unactualized possibilities. A determinist could agree here and still hold that there is a difference between a determinate future and a determinate present. In what can the difference consist if the determinate future possibility and the determinate present actuality are identical with respect to every constituting property? A number of considerations can be urged in reply.
1. The point of the “existence is not a property” doctrine is to deny the suggestion that the existence of a conceived entity can increase or decrease the conceived entity in character, quality, value, or perfection. The point is not to say that existence makes no difference, that it does not matter whether or not one has the $100. The boundaries of existence are still important in specifying temporal order, temporal relation, causality, etc. The boundaries of existence make the difference between being thankful and hopeful. They lend to human life its moral concern, its joy, and its tragedy.
2. The argument that the determinist cannot tell us the difference between a definite future and a definite present appears symmetrical with regard to the past. The past is now determinate, hence the past is now actual. The indeterminist may want to say that the past is completely held in memory by the present and therefore the past is indeed actual. But then, in what does its pastness consist? Is not this just as absurd as what the Bergsonian accuses the determinist of believing? If the concept of the fully determinate future is logically forbidden, so must the concept of the fully determinate past be also.
3. Some indeterminists would want to say that some future events are now determinate. But if the Bergsonian argument were taken seriously, that position would be inconsistent (2). If there is present determinacy, there is also present actuality, and to speak of the events as future is illegitimate.
4. A position that would be more likely to be accepted by Weiss and probably by Hartshorne is that the future is determinate often in broad outline, but not in its every detail. If an apple is thrown from the window five minutes from now, it is now definite that it will fall down to the pavement and that it will be damaged upon impact. But exactly how it will fall and how it will land and how it will be bruised or smashed is now indefinite. But if that is the position, one must now hold that the predictable aspects of the apple’s flight and end are timeless, whereas the aspects that are indefinite prior to the fall are temporal. The result would be a bifurcation between the predictable and unpredictable aspects of reality.
Another systematic commitment which lends appeal to the Bergsonian position is a use of “possibility” such as to preclude antecedently determinate states of affairs from being possible states. I do not recall A. N. Whitehead’s presenting the argument in the Bergson-Weiss form. However, there are in Whitehead some clues that he accepts the idea in a modified way. At any rate, there is an oddity in Whitehead’s technical terminology that is grist for the mill of the Bergsonians.
A possibility for Whitehead is something which may or may not be actualized. That might appear unexceptionable, were it not for the fact the “may or may not be actualized” says more than that the state of affairs in question is logically possible. Whitehead writes, “An actual event . . , is divested of all possibility” (PNK (61). Why should that be? After all, an actual event is a logically consistent state of affairs. Should not an actual event be considered a logical possibility? Maybe so, but what Whitehead means by saying an actual event is divested of all possibility is that it is restricted to one definite kind of thing. What then does the determinist believe? The determinist says that the future has the same kind of restrictedness. The future is closed to alternatives. The future is not the area of what may or may not be. Instead, what will be will be, and what will not be will not be. The past cannot be the area of possibility either, because that too is restricted to being exactly the way it was. For the determinist then there are no possibilities in the requisite sense. It follows that there would not be any unfulfilled possibilities. The distinction between actualized and unactualized possibilities is collapsed. The only distinction the determinist can draw is between actuality and impossibility. The predicted future for the determinist is not impossible. So we must conclude that it is for him actual. Hence the philosophy of creative process requires the refutation, repudiation, or dismissal of determinism. An actual future is no future at all.
Milic Capek accepts the Bergsonian argument as well as the Whiteheadian use of “possibility”. Capek also cites William James in this connection. Time is incompatible with the preexistence of the future, and the preexistence of the future is what Capek thinks the determinist is committed to. A real future must have the status of possibility (PICP 336). But if the future is determinate, it is divested of possibility and thus is not future at all. In this way Capek accuses the determinist of believing that everything conceivable must be either impossible or actual.
This terminological oddity in Whitehead and James obscures a distinction which the determinist wants to make and which I think is reasonable. That distinction is between being a possibility and being closed to alternative possibilities. For the determinist, past, present, and future are equally closed to alternative possibilities. However, what has happened, what is happening, and what will happen are all possibilities. They are logical possibilities and causal possibilities. A more precise way of putting it is that with regard to every future logical possibility the causal probability of occurrence is either 1 or 0. Because our abilities in prediction are not perfected, we may for practical purposes have to put up with probabilities somewhere between, as we do when we predict what electrons will do. But with respect to every logically possible future state of affairs, it either will be or it will not be. Although we cannot distinguish between past, present, and future on the ground of being open or closed to alternative possibilities, we can distinguish between them on the ground of whether or not the possibilities have been actualized and on the basis of the temporal order of the actualization. There is no reason to suppose that a possibility’s being closed to alternative possibilities entails that it is not a possibility at all.
Given any conceivable future state of affairs, the causal determinist believes either it will be or it will not be. Some thinkers believe that admission in itself is sufficient to rule out transiency or becoming in nature. It is true that the causal determinist is committed to the view that all events can be tenselessly described. “World War II (tenselessly) ends in 1945″ would for the causal determinist be true before, after, and during 1945. So would any true description, even those of allegedly free actions that were not decided” until the time they occurred. However, the causal determinist is free to adopt a compatibility thesis. The timeless truth of “World War II (tenselessly) ends in 1945” is compatible, he could say, with the tensed truth of “World War II ended in 1945” when uttered in 1946 and the tensed falsity of “World War II will end in 1945” when uttered in 1946. Under the compatibility thesis, tensed descriptions say everything the corresponding tenseless description says and more. Corresponding to every true tensed description, indeed constituting every true tensed description as one of its components, there is a timeless meaning core which is tenselessly expressible and which is true both before and after the occurrence of the event in question. However, there is also in every true tensed description a tensed component which is not timelessly true. It is necessary but not sufficient for the truth of the tensed description that the timeless meaning core be true. Consequently, there is no logical route from the thesis that “what will be will be” to the further thesis that the future tense in the old adage is illegitimate.
There is one other important aspect of Bergson’s challenge to the determinist which I need to comment on, and that is that the determinist cannot find any asymmetries in nature upon which the objective distinction between before and after can be based. This makes the attack on determinism even stronger, because here determinism is said to be incompatible with any understanding of time, not just with process philosophy’s understanding of it. The claim here is that only the indeterminate’s becoming determinate can establish any ‘temporal distinctions or direction whatsoever. Bergson would say that he at least is giving a philosophical account of time, whereas the determinist is silent on the subject. For the Bergsonians, time is a process of becoming determinate, whereas for the determinists time, if it is said to be a process at all, can be nothing more than a process of becoming — mere becoming. But what is it to say that tune is a process of becoming? It is simply to say that time is time. The determinist is stuck with a symmetry between prediction and retrodiction such that he cannot specify the difference between saying “It will happen” and “It happened.” Well, one refers to the future and the other to the past, but what is that distinction all about? And even if the determinist is willing to give up tenses, he is still in trouble, because he is unable to specify the difference between the relative earlier and later.
My first response is a tu quoque. Is our understanding of time really enhanced by believing time is the process of becoming determinate? The very idea of “becoming determinate” includes the idea of becoming, a temporal idea. It is not possible to know what becoming anything is unless one has some concept of time already. To say that time is a process of becoming determinate is to presuppose the very concept we are out to clarify. A tu quoque argument is reasonable in this context, because the claim we are examining is that only the indeterminist can give an account of what temporal relations are. What is time? Augustine and Whitehead both saw mysteries and paradoxes they were unable to unravel, though Whitehead moved much further toward a solution than did Augustine. Bergson thought time is duration, which would not be very enlightening if one is puzzled by the concept to begin with. Quite possibly temporal words are like moral words in that any attempt to define them either presupposes or betrays the concept in question.
If the determinist is asked to move on to a positive proposal of what time consists in, he might try to provide the requisite asymmetries Outside of nature, in logic or consciousness for example. Storrs McCall proposes in behalf of the determinist that the difference between past and future is logical, rather than physical or metaphysical in character. The determinacy of the future is a logically contingent matter, whereas the determinacy of the past is logically necessary (3:279). I think McCall means that, even if determinism is true, it would not do violence to our language to consider the possibility that the future or some aspects thereof is indeterminate, whereas we cannot even coherently think about the past’s being indeterminate. Process would then be the logically contingent’s becoming logically necessary. The “now” would be the moment of the change, and temporal order would be the succession of changes from contingency to necessity. The process would be irreversible, because logical necessity is not a characteristic a proposition can lose once it gets it. Apparently McCall does not have in mind arguing for determinism from the premise that the law of the excluded middle applies to all untensed propositions, for then the determinacy of the future would be just as logically necessary as the determinacy of the past. Apparently he would argue for determinism from the observation that causal regularity is a contingent feature of human experience.
However, I do not see that it matters whether determinism is an alleged logical or a contingent truth. The logical contingency of the future’s determinacy does not in one whit alter its determinacy. We are not reassured to learn that the timelessness of determinacy is logically contingent. Exactly what is it that is alleged to happen in becoming? A logical feature of propositions is said to change. Nature does not change, but propositions do and hence also the minds that entertain them. Then we wonder whether the future of consciousness is determinate. If it is, changes cannot take place there either. If it is not, then we must admit that the only way one can introduce tensed language is under the veiled assumption that at least some aspect of reality, i.e., propositions and/or the entertaining of them, is not subject to causal determination. Besides, if becoming is only a feature of logic, language, or consciousness, we get a serious mind-body bifurcation and an unbridgeable gulf between appearance and reality. And so, I think we must ask, if process is not a matter of the indeterminate’s becoming determinate, what then is its physical or metaphysical basis?
There is the suggestion that the asymmetry of temporal process can be linked to increasing entropy. Capek rejects that idea on the ground that time is more basic to reality than the contingent feature of increasing entropy in some systems. Furthermore, the recognition that there are or may be systems of decreasing entropy would involve contradictory ways of talking, because then relatively later stages at which entropy is decreased would have to be regarded as earlier (PICP 349). More simply, there is absurdity in talking of time’s going ahead backwards. Finally, to define time in terms of increasing entropy seems circular, because we are defining time in terms of entropy that increases with time.
I do not see absurdities here. We can compare two states of a system with respect to energy distributions. We then note that psychological time goes in the direction of greater entropy. In Bergsonian terms, the élan vital fights against this tendency, but matter finally wins out and there is death. We can then talk about systems of decreasing entropy, but only relative to systems of increasing entropy (including psychological time). Citizens of systems of decreasing entropy would not experience time in reverse; they would not experience time at all. Time would not be a feature of their system, for there would be no traces, no memory, no anticipation. I am speaking figuratively about there being citizens of such a system, for there could be no languages or persons under such conditions. So I see no problem in linking temporal asymmetry to entropy increase.
There are some phenomenological clues to this. The closest we get phenomenologically to a system of decreasing entropy is in our moments of creativity. Our creative moments are those when we come closest to escaping temporality. Suppose our awareness of temporality, our memory and our anticipation, bring us anxiety. The best remedy for temporal anxiety is creative endeavor. The causal traces of the past become least determinate upon us, and anticipation of all but the immediate future becomes irrelevant as every moment is endowed with an integrity of its own.
As a corollary to these observations, I suggest that causal indeterminacy is far from being the ground of temporal asymmetry, but is instead a lurking threat to time. If causal indeterminacy were complete, there would be no traces, no memory, no temporal order. If the present depended causally on nothing, there would be no sense of the past. That is the truth in the causal theory of time, and that truth is not endangered should our successes in retrodiction and in prediction match each other, for still causal dependency is an asymmetrical relation. Whitehead’s prehensive relationship that the present has to its past is one of causal dependency. The relationship is asymmetrical in even those very stable societies where our abilities in prediction are highest. The prehensive relationship remains asymmetrical even in those systems of lawfulness that permit both retrodiction and prediction.
But quite apart from the possibility of establishing a physical direction of time, process philosophers have as a matter of fact in their metaphysical systems recognized temporal asymmetries that have nothing to do with an indeterminate future. For example, Bergson himself metaphorically compared the temporal process to a rolling and accumulating snow ball. The accumulative nature of the process has, so far as I can see, nothing whatever to do with the absence of predictability of the future.
Whitehead adopts an asymmetry of memory and anticipation, quite similar to Bergson’s rolling snow ball. The present grasps the past within it. The present is dependent on the past for its very constitutive reality. Although the present anticipates and makes choices which bear on the future, the present is independent of the future in the sense that the reality of the present is not in one whit diminished or altered by the failure of the anticipated events to occur. Further, the status as future of anticipated states of affairs is not diminished by the assurance that they will occur. The asymmetry is independent of Whitehead’s doctrine of indeterminacy. I see no reason why the determinist must blur the distinction between memory and anticipation, and some of the process metaphysicians themselves supply all the conceptual apparatus the determinists need to handle this distinction.
If I am right, process metaphysicians should be able to take a more relaxed attitude toward determinism. Logical and empirical arguments appear indecisive with respect to whether there is universal and complete predictability. Determinism is subject to internal paradoxes and requires us to say at a good many places that what we think are choices between indeterminate alternatives are really not what they seem. Both Sartre and Whitehead in providing us with conceptual schemes for handling self-determination have shown us that indeterminism is not a confused idea. Perhaps too the indeterminist’s recognizing not only causality but also chance and self-causation produces a more comprehensive and somewhat less dogmatic position. On the other hand, determinism remains a possibly true theory which does have its attractions. I hope to have shown that how this issue is resolved is of little systematic importance to process metaphysics. Process philosophers can maintain their critique of classical substance, of absolute idealism, of materialism, and of classical theology without having it rest on the rather flimsy structure of a “refutation” of determinism. Determinist theory may be wrong, but it is not absurd. A now determinate future can yet be authentically future.
BMP — Bergson, Henry. Bergson and Modern Physics. Dordrecht-Holland: D. Reidel Publishing Company, 1971.
CE — Bergson, Henry. Creative Evolution. New York: Henry Holt and Company, 1911.
DR — Hartshorne, Charles. The Divine Relativity. New Haven: Yale University Press, 1948.
LT — Gale, Richard. The Language of Time. New York; Humanities Press, 1968.
NM — Weiss, Paul. Nature and Man. Carbondale: Southern Illinois University Press, 1965.
PICT — Capek, Milic. The Philosophical Impact of Contemporary Physics.
Princeton: D. Van Nostrand Company, 1961.
1. Frederick Ferré, “Transiency, Fate and the Future,” The Philosophical Forum 2/3 (1971), 384-95.
2. Paul Fitzgerald, “Is the Future Partly Unreal?” The Review of Metaphysics 21/3 (1968), 421-46.
3. Storrs McCall, “Temporal Flux,” American Philosophical Quarterly 3/4 (1966), 270-81.