In the classic Star Trek episode “City on the Edge of Forever”, Captain Kirk and Mr. Spock must travel back in time to reverse changes to the timeline that have wiped out the Federation, and with it the USS Enterprise. In Terminator 2, the heroes try to prevent the creation of an artificial intelligence that will be responsible the deaths of more than 3 billion people in a future nuclear war. And in X-men: Days of Future Past, Wolverine is psychically projected decades into the past to disrupt a chain of events that will ultimately result in the enslavement of humanity and the genocide of superpowered mutants.
What is the ontological status of the bad futures the heroes are striving to avert in these stories? In each case, at least some of the heroes have experienced the bad future first hand. In Days of Future Past, one characters even experienced multiple bad futures first hand, all of which involved of his comrades dying horribly at the hands of mutant-hunting robots. Even when these futures are in some sense averted, it seems there must be some other sense in which the events of those futures are still real, even if they can no longer be related temporally to the world of the present in the familiar way. To express this idea in another way, if God were to list every event, the list would include events from the bad future, even though these events could not truly be said to be in the future (or past) of any other events. We might say that the events of the bad future are absolutely real (i.e. real in an absolute sense) but relatively unreal (i.e. unreal relative to some set of events). Absolute reality is a necessary but not a sufficient condition for relative reality. When Kirk and Spock, or the T-800, or Wolverine travel into the past to prevent the bad future from coming about, the most they can hope to do is render the events of the bad future relatively unreal. An event that is absolutely real cannot be made absolutely unreal.
If bad futures cannot be made absolutely unreal, then our picture of the stakes in stories like Terminator 2 changes dramatically. Sarah, John and “Uncle Bob” can’t really be trying to prevent the nuclear war, at least not in the same sense that I might prevent a ball from rolling into the street by reaching down to pick it up before it reaches the curb. No matter what they do, the list of all events will still include events of suffering and death on an apocalyptic scale. But nor are they only rendering the events of the bad future relatively unreal. They are, in Sarah Connor’s words, “making up history” — bringing new events into absolute reality.
The upshot is that stories of this type only make sense within a consequentialist framework. Nothing can truly prevent or ameliorate the horrors of the bad future. The inhabitants of those worlds are lost beyond all possibility of help; between us there can be no moral relationship founded on virtue or duty. The most a time traveler can do is try to make the world better, and the only way to make the world better is to bring a better future into existence. Only this way can there be more good events than bad. And in all of these stories, bringing about a better feature is treated as having the exact same kind of urgency (morally and prudentially) as preventing bad events from ever coming about.
It does not follow from this that consequentialist reasoning has global validity — that it provides the appropriate framework for dealing with all kinds of moral situations. Virtue ethical and deontological frameworks can offer no practical guidance for time travellers from bad futures, but consequentialism faces equally serious problems dealing with normally more pressing ethical challenges like divided loyalties. It’s also not clear that time travel of the kind depicted in any of these stories is even logically possible, let alone permitted by the laws of physics. If the argument is sound, I think the most interesting thing about it is that it shows how the content of a final theory of normative ethics might depend on facts about what the laws of physics allow; if the laws of physics allow time travel, then a final theory of normative ethics must be consistent with the validity of consequentialism in at least some situations.
Our hero giving the official salute of consequentialism