The Psychology of a Time Traveller (I)

Making decisions in the past

Posted Dec 31, 2010

Time travel. We're all fascinated by the idea. Philosophers are interested in time travel into the past in part because they are interested in what our universe would need to be like for in order for anyone to travel backwards in time, and in part because they are interested in how time travellers would make decisions and in what sorts of decisions they might make. You might wonder why that is at all interesting, given that there is little prospect of there being any time travelling into the past any time soon, and indeed, it may be that the laws of nature in our world mean that time travel is physically impossible. The interest lies in the fact that thinking about how decisions might be made in the past draws our attention to various interesting features of how decisions are made at any time.

What do philosophers have in mind when they are considering time travelling scenarios? While there is debate, many philosophers think that we need to be very careful when we think about what sorts of scenarios make sense in terms of travelling into the past. They think, for instance, that many of the representations of time travel that we see in film and on television are misleading. Here is a common one. Billy travels backwards in time to the 1930s. Billy has done some history and is aware that Hitler's coming to power in Germany in the thirties resulted in terrible things happening. So Billy decides to change what happened in the 1930s. He arrives in the past, equipped with his modern pistol, he discovers a youthful Hitler before he has come to power and before he has much in the way of security, and Billy assassinates Hitler before the national socialists come to power under his sway. Thus the rest of history is changed: there is no holocaust, there is no second world war, and so on. When Billy travels forward in time to the location from which he left - his present - the time he returns to is changed in various ways as a result of the changes he made in the past.

Most philosophers think that this sort of time travel scenario is impossible: that is, the scenario is in fact inconsistent. There is no way our world could be, in which time travel like that occurs. What these philosophers object to is the idea that someone could travel into the past and change that time from being one way, to being some other way. This, they hold, simply makes no sense. Change is always change with respect to something: usually it's with respect to time (you and I change by being one way at one time, and a different way at some other time). But there is no dimension in which a moment in time can change.

If one accepts this widely held view about the past, then all sorts of interesting questions are raised about the nature of decision and agency. The first thought you are probably having is that if the past cannot be changed, then we can make no sense of time travel at all. After all, when a time traveller steps out of the time machine she is thereby changing the past: she is changing it from being a time when she was not there qua time traveller, to being a time at which she is there qua time traveller. Even stepping on an ant will amount to changing the past, even if only in a small way. 

Here is where philosophers want to distinguish between changing the past, and causally interacting with the past. Changing the past is altering some moment in time from being one in which x happened, to being one in which x did not happen. Causally interacting with the past is being in the past, and being part of the events that unfold, and causing various things to happen in the past. Why is the second not the same as the first?

Surely if I causally interact with the past, I thereby change it. No. Suppose that Billy travels back in time and meets the young Hitler and tries to assassinate him. Billy fails in his quest - we know he fails, because Hitler did, in fact, go on to become leader of Germany. This is a fact about the past, and as such is something that will never be changed. But Billy causally interacts with Hitler when he talks to him, and when he tries to kill him. Indeed, for all we know Billy tries to kill Hitler multiple times, and it may be that it is these repeated assassination attempts that significantly impair Hitler's psychology to the point that he becomes a sociopath and heads down the road that we all know he takes. If that were so, then Billy's time travelling would, as it turns out, be one of the causes of the second world war. Billy has not changed the past in this case, he has merely made it the case that the past is the way we now know it to have been. What philosophers call consistent time travel stories (time travel where the traveller does not change the past, but instead is part of the reason why the past is the way it is) can be found represented in popular culture by films like 12 monkeys and one of the Harry Potter films that features time travel.

The crucial thought is that when someone travels back in time, what they do in the past has just the same sort of causal effects as what you and I do in the present. If a time traveller drops a glass on a hard floor, it most likely shatters. It is just that the things that the traveller does makes the past the way that it is. If the time traveller shatters a glass, then it has always been the case, ever since that moment in time, that a glass was shattered at that moment in time. The past is not changed from being one in which no glass shatters, to being one in which the glass does shatter.

Here is where the whole time travel scenario becomes interesting from the point of view of thinking about decisions and agency. For the time traveller, unlike you and I, already knows a whole bunch of things about the past moment to which he travels, and also knows much about what will happen after that moment. Some of this knowledge might come from memories, and some from historical record.

Suppose that you remember a tragic day when you were 6 when you left the gate of your house open and your dog escaped and was lost forever. That was a terrible time for you, so you decide to travel back in time to that day. On the assumption that you cannot change what happened in the past, there is an interesting question about what the older wiser time travelling you ought to decide to do when you go back in time. You know that the gate was left open. So you know that there is no point deciding to go back in time in order to prevent yourself from leaving the gate open. That action is doomed to failure, since you clearly remember that the gate was left open. So it would be irrational of you to travel back in time with the intent of closing the gate. You also know that the dog escaped, never to be seen again. So it would be irrational to try and prevent the dog from having escaped. But you do not know what happened to your dog. You have always feared that something bad might have befallen the dog. So here is a perfectly rational plan. You travel back in time and wait until your younger self leaves the gate open and the dog escapes. Then you grab the dog as it decamps from the yard, and take it to live with a nice family in the next town.

That action is consistent with everything you know about the past, so it is a perfectly rational plan for you to formulate before you get into the time machine. Indeed, let us suppose that you travel back in time and succeed in your endeavours. You now feel much better, knowing what became of the dog. Moreover, your actions as a time travel explain why the dog was never seen again: it was living with a new family a town away. You did not change the past by your visit, you were responsible for the disappearance of the dog.

What is interesting about this case is that only some decisions you might make about what you will do in the past are ones that are rational. It would be silly of you to decide to prevent the dog from having disappeared, given that you know that the dog did disappear. You have good reason to think that the dog did disappear, because not only do you remember this happening, but so does everyone else in your family. But now consider a case in which your knowledge is rather less certain. The moa were large flightless birds that lived in New Zealand and were hunted to extinction by the Maori. That there are no more moa in New Zealand is beyond doubt, and that there were hunted by the Maori is beyond debate. But suppose you are a huge fan of these flightless birds, and you have a time machine. What can you do about their plight?

 Certainly you cannot travel in back in time and explain to the Maori people that continued hunting will result in the birds becoming extinct. You cannot bring it about that the moa do not vanish from New Zealand, because they do vanish from New Zealand. But no one really knows what happened the last of the moa. It is consistent with everything we know about the past, that someone collected a reasonable number of moa and left New Zealand with those birds. So there is a way of saving the moa from extinction: you travel back in time, collect many moa and bring them back to our own time. Nothing we know rules this out, so it would be perfectly rational for someone to decide to save the moa from extinction by acting in this manner.

In effect, what makes these cases interesting is that our knowledge of what did happen, changes the sorts of things that it is rational to attempt to do when in the past. The usual way of save a bird from extinction is to change its environment to one that is more conducive to its reproducing. The usual way to prevent the trauma of the loss of a dog is to make sure the gate is closed. But neither of these strategies would be rational ones, given what the time traveller in question knows about what happened in the past.

Paradoxically, the less a potential time traveller knows about some time in the past, the less she is constrained in terms of which projects it would be rational for her to plan to carry out in the past. If a time traveller is incredibly well informed and knows everything about the time to which she travels, then she knows, among other things, what she did do at that time in the past.  That is, she comes to know things that are in her own subjective future (things she will do once she travels back) but things which are in the past and thus have already, in some sense, been done. Once a time traveller knows what she will do, her decisions about what to do seem to be entirely constrained. There is, after all, no point in a time traveller deliberating about whether to try and gather a bunch of moa together and bring them back to the present with her, if she already knows that in fact that is not what she will when she travels to the past.

If one accepts that once a time traveller knows everything about the past, she therefore knows what she will do in the past, and therefore she no longer has the capacity to deliberate about what she will do, then there are senses in which the time traveller ceases to be an agent. This is the sense in which it is crucial to agency that agents consider alternative choices and deliberate about which option to choose. The fully informed time traveller will not engage in such deliberation.

That, in turn, is interesting since it raises all sorts of questions about our own agency in the present. Many philosophers think that there is no significant difference between the time traveller and you and I, because there is no significant difference between locations that are in the past, the present or the future. This is a widespread belief, because it accords with the standard physical models of our universe as a block of space-time that includes all locations that did, do, and will exist. On this model, all of the temporal locations in our universe are equally real. The dinosaurs exist somewhere in space-time, as do any future robots. There is nothing special about the present moment - it just happens to be the moment that you are I are located at. For the robots, the moment they are located at is the present, and likewise for the dinosaurs. If that is right, then the only difference between the time traveller and you and I is that by being a time traveller, the traveller gets to know what did happen at some past location, and thus comes to know what he or she will do in her subjective future. Once she knows these things, her ability to deliberate qua agent is diminished or extinguished. If you and I are like the time traveller, then our ability to deliberate qua agents is only the result of the fact that we do not know what will happen in out subjective futures, in locations that is, that occur later than our current location. Were we to come to know what will happen in the future, the way a time traveller knows what will happen in the future (in virtue of it being in her past) we too would cease to be able to deliberate as agents.

One thing that one might say about all this is that it shows that there is a mere illusion agency brought about by the fact that individuals do not usually know all of the relevant facts about what will occur. In a perfect state of knowledge, however, there would be no agents. Alternatively, one might say that there are agents and agency is no mere illusion, but that agency just is the thing that you get when you put things like us under conditions of imperfect knowledge. It is true that agency would disappear were we to come to know what will happen in the future, but that is no reason to conclude that agency is an illusion.

More on this in the psychology of the time traveller part II.