The future past problem
Why the plot dies the moment the machine turns on — a walk through the three doors, from killing Hitler to a universe that won't let you.
Time travel is one of the most used tropes in media, and it’s always killing the plot. I personally can’t stand it — the premise is faulty from the start, and you can’t build a good universe or a good story on it. Sure, it looks entertaining, but that stops the moment you start thinking about the implications. And unfortunately for me, I do that immediately: the second time travel gets brought up, I already see the plot as dead.
We dream about going back and fixing our own mistakes, or others’, for the sake of the present and the future. It’s human nature to regret and try to correct — it’s hard to move on. Or we dream about going forward, skipping the wait to see what happens. It’s an old wish, and a stubborn one: we’ve been dramatizing it for over two thousand years — and getting it wrong for just as long. We first went to the future around 400–700 BCE, when King Kakudmi travels to the heavens to meet the creator and returns to Earth to find that millennia have passed. We went back in time in 1640, when Sun Wukong travels to the World of the Ancients through a tower of magical mirrors. And we built the first device to travel through time in 1881, when Edward Page Mitchell wrote “The Clock That Went Backward” — a clock that ran the wrong way.
Even scientists thought about this for a while. Einstein cracked forward travel in 1905, through special relativity. If you think about it, we’re always moving forward through time — and Einstein showed that your speed changes the rate at which you do it: move fast enough, and your clock runs slow. Ten years later, general relativity added gravity to the same effect: sit close enough to a heavy enough mass, and time slows there too. But this real forward travel is inert. It slows your aging, not the world’s clock — you don’t skip ahead to a future, you live through the whole stretch of it slowly, one-way, with no way back to tell anyone what you saw. The kind of forward travel stories use is a different trick entirely: it drops you into a future that has to already be there for you to arrive at all — and if it’s already there, it’s already fixed, so there’s nothing for you to change — and then it sends you back with what you learned. That return trip isn’t relativity; it’s backward time travel in a forward coat, and it drags every problem of going back along with it. Which is the real subject anyway. The dream was always going back.
And here the equations got strange without ever saying no. If spacetime can bend, then in principle it can bend all the way back on itself — a path that loops around to its own past. Gödel found exactly such a solution in 1949 and handed it to Einstein for his birthday; forty years later Kip Thorne dressed the idea up as a wormhole you could, in principle, walk through. The math has never managed to forbid it outright. But every serious route to the past comes back holding the same bad news: you can build the door, yet the moment you ask what walking through it would actually mean, the universe hands you a rule — and every one of those rules is just a different way of saying the story you wanted to tell is already dead.
So how do movies actually do it? Every time-travel story, whether it admits it or not, has to answer one question: can you change the past? There are only three answers — and that’s the whole problem, because all three are dead ends. You can say yes, but only in a new branch; you can say yes flatly; or you can say no — the past is fixed. Three doors, and every one opens onto a wall. It doesn’t matter which one a film picks: the honest ones die slowly, and the ones that cheat — swapping answers whenever a scene needs tension — just die faster.
Let’s take the first answer: going into the past creates a new timeline. This seems the most plausible, but it falls apart the moment you think it through. Take a case. You’re in the present, at point 0, and with some machine or bit of magic you go back to point −1, before the war. In this hypothetical present the Nazis are in power, and you’ve been sent back to kill Hitler before WWII starts. So you arrive at a point before the war — and the instant you arrive, before you’ve done anything at all, the timeline already splits. It has to: in your original present there was no record of someone showing up from the future. Your mere arrival is a change. You go on with the mission, every action splitting off more timelines, and when you finally kill Hitler another one branches away — the one you wanted, where the war never happens.
Now you want to go home. But which present do you return to? The original one? The Nazis are still in power there; you changed nothing, and your mission was futile. The new one? No one is waiting for you there — in that timeline you were never sent back, because there was never any reason to go. So you’re stuck: a person from timeline A, stranded in a timeline that has nowhere to send you home, and no way even to check whether what you did worked.
The branch model fails because nothing you do ever reaches your own world. Not even you can reach it — or if you do, your trip was pointless. And that’s the deeper problem: since any decision at all spins off a new timeline anyway, a timeline where Hitler is already dead may exist already, without you ever lifting a finger. Your effort was pointless from the start.
The obvious fix is to forbid the branching: one timeline, and what you change in the past reaches your future. Same example, played again. You’re at point 0 and you travel to −1 to kill Hitler. You reach −1, you carry out the mission, you succeed — and then you go home? It doesn’t work. Because you killed Hitler, there are no Nazis in power at point 0, so you have no reason to travel back and kill him. So you don’t go. But if you don’t go, Hitler isn’t killed, so the Nazis are back in power, so you do have a reason to go, so you go and kill him — and now, once again, there’s no reason to go. It never settles. It just loops.
This kind of time travel always ends in a loop. You might think you can dodge it by doing nothing — go back, observe, touch nothing. The same loop occurs. You travel to −1 to see what really happened, so now, in the present, you know — so there’s no reason to go back to find out, so you don’t, so you don’t know, so you go back again. The loop is still there. It was never about changing anything: the trip itself is what erases its own reason.
So you forbid this as well, and you have the third answer: the timeline is fixed, and the universe simply will not permit anything that wasn’t always true. This is the physicists’ real escape hatch — Igor Novikov’s self-consistency principle. You can travel to the past. You just can’t change it, because whatever you do back there, you already did.
Take the same trip. You go to −1 to kill Hitler. But you never killed Hitler — history says he lived — so something stops you, every time. The gun jams. You shoot the wrong man. You arrive a day late. Or, worst of all, you are the reason: the strange foreigner who tried and missed becomes another line in the myth. There is no version where you succeed, because success is a contradiction, and the universe does not allow contradictions. Your bullet was always going to miss. It missed before you were born.
And this way there is nothing oscillating, nothing to branch; the timeline is, and always will be, perfectly stable. It is also perfectly pointless. You had no choice at any moment; every decision you thought you were making was already written into the history you came from. You are not an agent in the past. You are the same person you read about in the history books — the one who couldn’t do anything to change it. Here time travel is completely real and completely inert. The first two answers broke the story; this one keeps the story intact and quietly removes the only thing a story needs — the possibility that things could have gone otherwise.
But this does not make for a great movie. And to be honest, none of them do, once you think about them a bit more.
The same trap is waiting in the future. Imagine this: at point 0 you receive information from point 2 about an action you need to take at point 1, and you take it. Now point 2 has no reason to send that information back — so you never received it, so you never acted — and there’s the loop again. You’ve only moved it into the future instead of the past. Or you go to the future with no information at all and do something once you arrive. It has zero effect: from their point of view the change at point 1 is simply what happens, and by point 2 nothing has changed, because it was the natural flow of time all along. Which means time was already determined. Which means there was never anything to alter.
This is why I can’t sit through them. Every time-travel story is quietly picking one of these three doors and betting you won’t look. Once you’ve seen the doors, you can’t stop seeing them — and the story dies in front of you, every time, the moment someone says they’ve found a way back.
⁂