CONTENTS

1   INTRODUCTION

2   TECHNICAL DEVELOPMENTS

3   B-THEORY: EXPOSITION

4   A-THEORY: EXPOSITION AND PARADOX

5   ARGUMENTS FOR A-THEORY

6    A-THEORY: REVISED EXPOSITION

7    TIME AND ASSERTION

8    THE PAST-FUTURE ASYMMETRY

9    CONCLUSIONS

8   THE PAST-FUTURE ASYMMETRY

Time is not just any radical multiplicity, it has a linear-ordering structure, with a clear asymmetry between past and future. This asymmetry is closely bound up with the phenomenon of passage, and inherits from it its unsayable character. We want to say that the passage of time turns the future progressively into the past, or, conversely that it takes us (and the rest of the world) from the past to the future. But these models of passage, invoking the idea of change or motion, point to a crucial truth about passage but always fail to express it in language as a literal truth.

We should also resist the temptation to say that in A-theory the future is open, indeterminate, while the past is closed, determined. It is essential to A-theory that future tensed statements have determinate truth-value just like present and past tensed statements, however hard the future truth-values may be to know. The truth about the future that we try (in vain) to express by saying that each future moment will, in due course, be present entails the determinate nature of the truth-value of future tensed sentences. It is the very truth-value that the corresponding present tensed sentences will, in the fullness of time, have. This ontological result indicates the way out of epistemological difficulties with the future. There is a perfectly good method of finding out about the future, one which is quite unavailable for the past. It is called the wait-and-see method. All you have to do is stay alive long enough to find out was happens.

None of this entails the world should be deterministic in the causal sense, whereby the future is determined by the present. The truth-value of future tensed sentences is determinate, because it is determined by the way things will be. It may be that we can not know now what the future will be because these things are presently strictly unknowable. The whole present world may not “know” the future - which is a way of saying that the laws of motion of the world are not deterministic - in which case no person within that world can know the future. But we still know that there is some state to be ascribed, using the future tense, to the world in the future (unless sense can be made of the possibility that time might stop, beyond which point there is no future). 

Similarly the logical determinacy of the future does not preclude the possibility of free action. It does not entail a logical fatalism. The tautology “Que será, será” still leaves open the possibility that what will be will have been the result of my free choice and action. Even if there were an omniscient God who knew what I would choose long before I even existed (I take it that the quantification in “omniscient” extends over all time - a strongly B-theoretic notion), when it comes to the point of choosing and causing the future state in question, the choice and the cause will be mine. God, as omniscient, would have known that all along.

Conversely although the past seems fixed to us (and is logically determinate, in a past tensed way), it can become epistemologically detached from us. It is quite possible that all physical records of large parts of the past have simply been, either in practice or indeed in principle, destroyed. It may never be possible to know the colour of dinosaurs or the what Jesus did as a child. With the lost past, le temps perdu, we do not even have the wait-and-see method available for the future. (The question of the retention or erasure of records of the past turns out to have important causal significance in quantum mechanics.)

There is certainly an epistemological asymmetry between past and future which creates the impression of closed past and open future. We are much better at remembering the past than we are at anticipating the future. Certainly, our access to the past is much enhanced now that we have a long tradition of keeping records, and have developed the sciences of interpreting physical records. After three millennia of continuous literacy, half a millennium of printing, two centuries of systematic palaeontology and archaeology, and now living in an age saturated with means of recording information, we perhaps underestimate how difficult memory is to bare unaccommodated man. Nevertheless we have made no such great strides in the technology of anticipation. Much of the immediate future is opaque to us while the immediate past is documented in minute detail.

I conjecture that this asymmetry derives from the fundamental asymmetry between Self and Other. The Self is small and the Other is large. My knowledge of past things derives from the influence the Other has on me. To know about the Other I have to have received messages from it - this includes the limiting case of that very particular Other, my past Self. Conversely I have knowledge of the future precisely to the extent I can influence things. Suppose I predict that in one minute’s time, the cup on my desk is going to move into the kitchen. One minute later I carry the cup into the kitchen, and claim a successful prediction. I am likely to be accused of cheating, because I was the one who moved the cup, and at the time of making the prediction I fully intended to move it. Instead of cheating, I am merely demonstrating the circumstances under which I have reliable knowledge of the future. Because the rest of the world is bigger than me, and influences me more than I can influence it, I have a bigger store of this reliable knowledge about the past than I have about the future.

This attempt to explain the epistemological asymmetry between past and future points us towards the underlying asymmetry, that with respect to causation and agency. When the world influences me, for example in a way which gives me knowledge about it, then its past creates my future. When I influence the world my past intentions create its future. I can not now bring about the past because it is intrinsic to the notion of “bring about” that what is being brought about is a future state. We can not say what “happening” is, or what the past-future distinction is, but we can know that the two are different aspects of the same thing. While the future is not open in the logical sense, it is logical in the sense that, to the (strictly limited) extent I can influence the world, I can choose the way I want the future to be.

There is much talk among physicists, most of whom, if they think about the issue at all, are incorrigible and unreflecting B-theorists, about the “Arrow of Time”. The valid and important question of physics is whether physical processes are reversible. Unfortunately this question is given the confusing name of “time-reversal symmetry”, which suggests that what is at issue is whether it is possible to travel backwards in time or not.

If we speak in slightly less tendentious terms of T-reversal, then the symmetry claim in deterministic physics says that if a process A®B is possible, then so is B®A. In probabilistic physics the statement is that the two processes have the same probability. If the symmetry does not hold, at least some changes in the world will be irreversible. In the deterministic case, once we go from A to B we can not get back. But in all cases, whether the symmetry principle holds or not, the processes go from past to future - as argued above, this is intrinsic to the notion of “process” and the past-future distinction. Indeed, without a distinction between past and future prior to any mention of the contents of time, the two processes would be identical (i.e. there would actually be only one process) and T-reversal symmetry would no longer be a substantive law of physics.

Put another way, if the processes in my body were T-reversed, in forty years time I would be a child again. I would be in the same state I was in 1958, but the date would be 2038. This is quite different from travelling back in time to 1958 - a notion, at least within A-theory, of dubious coherence. Of course if the clocks were likewise T-reversed, at the end of the process they would read “1958”, but, according to A-theory, this very T-reversal would render them invalid as clocks.

In B-theory by contrast there is no intrinsic direction to the temporal manifold. We then have to look to the contents of the manifold to give it an extrinsic direction. If for example we have an asymmetry in the way thermodynamic processes change the entropy of an isolated system, we can use this as an extrinsic “arrow of time” and say that the higher entropy states are later than the lower entropy ones. In this case the principle of entropy increase is not a law of physics, but rather a definition of the time-ordering word “increase”. Of course in the world of our experience, away from the fancies of B-theory, we start with a past-future asymmetry, and then notice that the higher entropy states are always later, as a contingent law of physics.

9          CONCLUSION

A combination of factors and arguments have come together in the twentieth century to support the belief that time is a static manifold of co-existing events, and that our supposed experiences of presentness, of the going away of the past and the coming to pass of the future are at best purely subjective or at worst illusory. In either case the supposed phenomenon of temporal passage is of no relevance to the physicist, the student of objectivity.

The strongest argument to this conclusion is that of McTaggart. When we try to express the things which we experience to be true of temporal passage, we encounter contradictions. McTaggart concluded that time, which he identified with passage must be unreal.

The problem with McTaggart’s argument is that it proves to much. If valid, it proves that even the illusion of passage is impossible. The sensation of passage, at least as an illusion, is one of the most fundamental and intimate parts of our experience. If this were unreliable as a source of knowledge then we could rely on nothing. Therefore there must be a way around McTaggart’s argument.

The conclusion argued for here is that what McTaggart has in fact shown is that there exists a fundamental limitation on our ability to express the known truths about time in language. This result is quite as startling as McTaggart’s own conclusion that time is unreal. I conjecture that time is an example of a “radical multiplicity”, the existence of which prevents us using the normal assertoric mechanisms of language. We know the facts about passage as well as we can know anything, and can point to them using words, but we can not use words to express this knowledge as articulated understanding.

The other argument against the physical reality of passage is that from Special Relativity. The argument from experience shows that there must be a way to reconcile our experience with relativistic reality, but producing such a reconciliation is a task left to another essay.

Click here to return to the start of the essay.