TLS: The future's not what it was

by Barry Dainton
Review of THE LABYRINTH OF TIME. By Michael Lockwood. 405pp. Oxford University Press.

We all know from the nursery rhyme how the loss of a nail can lead to the loss of an entire kingdom (via the intervening losses of shoe, horse, rider and battle), so it's not too difficult to appreciate how a small alteration to the remote past could cause a cascade of changes which would lead to the creation of a very different present - perhaps a better present, if the right alterations were made. Nor is it difficult to see how time-travel technology might be of use to future governments: if a policy decision turns out to have catastrophic consequences, the damage can be reversed by sending a "Don't do it!" message back in time. This plot device is employed in Orson Scott Card's novel Pastwatch, where a group of time travellers deliberately set out to alter history in a way which is guaranteed to erase their own timeline. The twist is that they do so with the explicit permission of the millions of people they will be annihilating: faced with global environmental catastrophe, the inhabitants of this post-apocalyptic society nobly opt to sacrifice themselves in order to give humankind a second chance. Could travelling to the past really pose a threat to the present? Probably not.
For a start there is the problem of self-defeating causal loops. If the changes you inflict on the past produce a future in which you no longer exist, how can you go back and inflict the changes in the first place? But alterations which aren't self- defeating are problematic too. Last night I cut my finger while chopping an onion; the pain wasn't severe, but I certainly felt it. Now consider: once this particular experience has come into existence, can anything make it so that it never happened in the first place? As soon as any event occurs it is part of the sum total of everything that ever happens, and once an event has this status how can it lose it? Consider again Scott Card's scenario: can billions of people live out their lives and also not live out their lives? Although some philosophers have held that reality can harbour contradictions, it is difficult to believe that this is one of them. On reflection, the idea that the past can be changed in even the smallest of ways looks dubious.
Should we conclude that time travel is impossible? No, but we must accept that if people successfully travel back in time, the past won't be in the least bit changed by their so doing. Hence the arrival of the time tourists and all their actions (and all the consequences of their actions) must already be part of history before they set off. Rather than picturing the past happening twice over -once without any time travellers, once with -we should picture it as happening just once, with any time travellers already in situ.
Recognizing the inviolability of the past solves one problem but it leads to others. What is to prevent a time traveller doing something they know didn't happen? Suppose you travel back to 1968 on an afternoon when you know you participate in a particular Parisian demonstration: you've seen the pictures in the history books which prove the fact. What's to stop you spending the whole afternoon wandering along the banks of the Seine instead? What's to stop you getting back in your machine and travelling to another historical epoch? In one sense nothing at all. You aren't confined to a prison cell or bound and gagged - to all outward appearances you are as free to act as you choose as anybody else.
But if we assume history cannot be changed and that the history books are accurate, then irrespective of what you try to do, you will end up doing just what the history books say: some chain of events (one that you cannot as yet foresee) will lead to you being where you're supposed to be. Why does this fortuitous chain of events take place? Because it did; that's just how history worked itself out.
You may feel perfectly free as you contemplate ways in which you might escape your destiny, but this is an illusion. Your immediate future (like our past) is set in stone.
That we find it perturbing to think that our actions could be constrained in this sort of way is (in part) due to how we normally think of time itself. If asked to characterize the differences between the past, present and future many of us would say something like this: "Although there is a fact of the matter as to what happened in the past, the past isn't real in the same way as the present: things are only really alive in the present -though of course things are only present for an instant before becoming past as time marches on. As for the future, it doesn't exist at all, and what it will bring is as yet not settled". Our futures may be open in this way, but travellers into the past must accustom themselves to the fact that their immediate futures are not.
But if this conception of time explains why the predicament of time travellers seems so strange, it also makes it hard to understand how backward time travel could be possible at all. How could someone arrive in the present from the future if the future doesn't exist? Non-existent people can't build time machines, nor can they step out of them!
Happily for would-be travellers all is still not lost. The view of time just outlined is often called the "tensed conception" - since it assumes that there are significant differences between the past, present and future. Its main competitor is the "tenseless conception". As the label suggests, on this view there are no significant differences between past, present and future. For tenseless theorists all events are equally real, irrespective of when they occur. It follows that the present time is in no way privileged: all times and events are present as and when they occur, just as any place is "here" for a person who happens to be located there. The tenseless conception may not correspond with the common-sense picture of time, but if the future is real, the notion that people might set out from there with the intention of arriving in the past is no longer problematic. Time travel seems to require tenseless time. For reasons that will emerge, the possibility or otherwise of time travel plays a prominent role in Michael Lockwood's The Labyrinth of Time. Lockwood's main aim is to assess the competing merits of the tensed and tenseless conceptions. Philosophers who favour the tenseless view often do so because they think the tensed view can be shown to be incoherent. Lockwood, however, finding these arguments to be flawed, swiftly moves on to more promising territory: the implications for the nature of time of contemporary physics.
In broad brush strokes the story runs as follows. The doctrine that only presently occurring events are fully real is put under severe pressure by Einstein's special theory of relativity. Common sense suggests that the present is a fleeting phenomenon but not a localized one: if one's current experiences are present, so too are all the other events that are happening at the same time, irrespective of where they are located. The special theory does not dispense with this notion of simultaneity, but it does relativize it: people who are in motion with respect to one another will legitimately reach different verdicts on which events are present. Obviously, if simultaneity is relative in this way, the tensed theorists' contention that the present is a universe-wide frontier separating a real past from an unreal future is in serious trouble.
Since the only remotely obvious way of making sense of the situation is to hold that all events are real, it is not surprising to find that Einstein himself favoured the tenseless conception. So much for the special theory, but how do things stand from the vantage point of Einstein's general theory of relativity? According to the latter, gravitational effects are due to mass-induced distortions in the spacetime continuum, rather than the action-at-a-distance force posited by Newton. One might well think that if space and time can be contorted by the presence of matter (the more matter the greater the contortions), then the tensed theorists' universe-wide tide of becoming is in even worse trouble than previously. But it is not quite that straightforward.
Surprisingly, it turns out that it is possible to define a universal "now" in cosmological models permitted by general relativity. The definition proceeds via three-dimensional surfaces of "mean constant curvature" (which Lockwood explains very clearly). However, this putative present is only available in spacetimes that are "globally hyperbolic". Simplifying somewhat, these are spacetimes which don't include closed time-like curves, i.e. paths through spacetime which, if followed, would lead one back in time. Hence the importance of time travel. Does general relativity permit such journeys? It does. Closed time-like curves can be found in rotating black holes. Potentially of more use to the would-be time traveller, the equations also permit wormholes -in effect, tunnel-like short cuts through space and time. It is hard to see how the future could be anything other than real if it can be connected to the present by a tunnel.
So if we take relativity as our guide to the universe we need to face up to the possibility that the tenseless conception is true. While some tenseless theorists play down the implications of this for how we should think of ourselves and our lives, Lockwood does the opposite. I think he is right to do so.
To focus on just one issue: most of us take it for granted that, as we deliberate about what to do next, what we will do is not yet settled. If the openness of the future is an illusion, this assumption is untenable: our lives are embedded in the unchangeable past of a future that is as real as the present.
Hence it could easily be that we are all in a position analogous to that of a traveller to the past labouring under the delusion that they are free to do as they like. The only difference is that, since we don't have access to the relevant historical records, we are not in a position even to try to rewrite history.
This is on the assumption that general relativity is true. Einstein tried to reconcile the general theory with quantum mechanics but failed, and we are still looking for a viable quantum theory of gravity. Intriguingly, from the picture Lockwood paints, some of the leading candidates for such a theory may offer a few glimmers of hope to the tensed theorists. String theories are formulated against a backdrop of a flat spacetime, albeit of the special-relativistic variety, and there are approaches falling under the "canonical quantum gravity" rubric which construe the universe as succession of three-dimensional momentary spaces.
(Unfortunately talk of "succession" here needs to be taken with a large dose of salt: the canonical approach faces the difficulty that time doesn't feature at all in its equations, and there is no consensus as to how it can be retrieved.)
Rather than dwell on these highly speculative matters I will close with a brief mention of the best-confirmed part of contemporary physics: quantum theory. Readers familiar with Lockwood's previous work will be aware that he favours a version of the "many-worlds" interpretation. Quantum theory says that any given physical system can evolve in any number of ways from any given point in time, and the Schrodinger equation assigns specific probabilities to these various possibilities; what this equation does not do is tell us which possibility will be realized. The standard (Copenhagen) doctrine is that a system only comes to be in a definite state when it is observed. But there are obvious problems. How is it that observing a system has such profound consequences? What condition is the system in prior to its being observed? (Hence the notorious problem of Schrodinger's cat, which seems condemned to be both alive and dead before anyone peers into its box.) The many-worlds interpretation sidesteps these difficulties by holding that all the possibilities licensed by Schrodinger's equation are realized, albeit in different branches of reality. We are accustomed to objects being in different conditions at different times; if we accept the many-worlds interpretation, we need to accept that a single object can be in different conditions at a time. As a way of making sense of this, Lockwood suggests that the familiar twosome of time and space needs to be extended to a threesome of time, space and (what he calls) actuality. This additional dimension houses all the multitude of branches of the universe, each of which has its own space and time.
The many-worlds interpretation is by no means the only game in town, but it is taken seriously by physicists -especially those working in cosmology. It also has implications which go beyond its ontological profligacy. Would-be travellers through time needn't abandon all hope of changing the past. You can travel back and murder your grandfather -or at least, you can travel to another branch of actuality at a time when your grandfather lives, and where (thanks to your actions) you will never be born. But while the multiverse offers new freedoms to time-travellers, deleting entire strands of history is not one of them. You may be able to travel back and prevent a disastrous policy decision being made in one branch of actuality, but the suffering that this policy produced in the branch you abandon will remain.
Also, although Lockwood doesn't dwell on this, there are implications for our freedom here too. We normally assume that the ability freely to choose between two courses of action goes hand in hand with the ability to pursue one of these courses and not the other. This assumption is menaced by the many-worlds view.
Let us suppose that after some deliberation you end up choosing to get up rather than stay in bed. In another branch of actuality you end up choosing to stay in bed.
How can this be? Because your decisions are the outcome of micro-physical processes in your brain, and these processes can (and do) unfold in indefinitely many different ways from any given point in time.
Under the combination of tenseless time and classical physics only a single course of action is a real possibility for us, despite appearances to the contrary. The implications of the many-worlds view are very different but just as disturbing: the many courses of action which seem to be possible for us really are possible (probably), but many are also actualized in one or more of the vast number of branches making up our tree-shaped lives. We are thus deluded if we think we have the option of not pursuing many of the possibilities open to us. It could easily be that we are condemned to pursuing all the courses of action we ever consider pursuing. If so, the implications are evident: no longer must we think of ourselves as saints or sinners (or something in between), we are all saints and sinners (and everything in between). I expect most of us will think this more of a burden than a boon, and, given the choice, would opt for linear rather than branching lives. Alas, the nature of time is one thing that is certainly beyond our control.