The 2007 indie SF film Timecrimes shows the horror & metaphysical implications of stable time loops: over a single day, the protagonist commits escalating crimes while trapped in a 3-iteration loop enabled by a time machine-or rather, caused by it.
The 200717ya film Timecrimes shows the horror & metaphysical implications of stable time loops: over a single day, the protagonist commits escalating crimes while trapped in a 3-iteration loop enabled by a time machine, or rather, caused by it.
Stable loops require events to be self-consistent, unlike serial loops. The outermost loop has priority to manipulate earlier ones, as long as perceptions match. More observations by early loops constrain later ones, creating an entropy conservation law. Successive loops compete to precommit or remove degrees of freedom, like Stackelberg games or Nim.
The final Timecrimes plot has no logical predecessor plots, raising issues around origination and sufficient reason: the time-loops appear to operate on a logic that if something is logically-possible, then it becomes actually possible. The protagonist is capable of evil, so the time-loops create an impossible but logically-consistent equilibrium in which he gives into his moral weakness.
Because of the temptation of power, and the ability to cover up crimes by going back in time repeatedly, because time machines can be abused, they must become abused. A time machine becomes a “damnation machine” mass-producing crime and empowerment, with immorality increasing the chances of an individual being trapped.
Conflicts spread time machines spatially and temporally as iterations try to persist. Stable time loops likely evolve convergently towards maximal scope. Inventing them may end humanity’s autonomy, warping causality irrevocably.
Timecrimes (Los Cronocrímenes): obscure 200717ya indie Spanish stable-time-loop time-travel movie. Recommended to me by Riley Goodside as one of the few good time-travel movies (in addition to Coherence or Primer); since I had already seen Primer, I started with Timecrimes.
To cut to the chase—the titular time-travel crimes are: kidnapping, false imprisonment, assault & battery, sexual assault, criminal intimidation, trespassing, grand theft auto, destruction of property, first-degree homicide, and lying to the police. (By my rough count, those are just the protagonist’s crimes if you leave out the ambiguous ones like ‘self-harm’1; a secondary character is guilty of at least accessory before & after the fact, reckless endangerment of both protagonist & public, and presumably a whole slew of laws/regulations on human experimentation.)
This eventful plot is despite taking place over one evening and the fact that Timecrimes was filmed on a budget that appears to have covered only gas money to the filming location (and possibly fries at a McDonald’s on the way back). Somewhat like Primer or locked-room mysteries, Timecrimes demonstrates that stable time-loop movies are a micro-genre whose formal rules lend themselves to an extreme economy. Indeed, Timecrimes winds up accidentally respecting Aristotle’s 3 classical unities: despite the length, it all takes place in a single day by a fixed set of actors depicting the same events from 3 perspectives.
TC Plot Summary
The core plot is this: the camera follows, in a coldly clinical documentary format devoid of any narration (internal or external), our protagonist, an older Spanish retiree, who has recently retired to a building in the countryside which he is fixing up with his short-haired wife. While tinkering, he sees a beautiful young woman with long hair in the woods—peculiarly, nude. Intrigued, he leaves his wife to investigate. She is laying unconscious, and he is attacked by a murderous man in pink rags & heavy coat. He flees in terror as the murderer follows him to the neighboring property, where he breaks in to seek help.
Unfortunately, as it’s the weekend, the property, which is some sort of shiny contemporary-modern research lab, is completely empty of people. There turns out to be one scientist there, in another building far up the hill; he tells the protagonist that the phone lines appear to be cut, and to come up to his lab. He goes, but the murderer follows, and attacks the building; the scientist tells him to hide from the murderer in a device, which is triggered.
The protagonist exits the device to discover the scientist has no idea who he is, and that it is now hours earlier in the day. The device is an experimental time machine, of course, which had never been tested on a living being before.2 There are now two protagonists: the later one sitting in the lab, and the one that will soon be attacked. The scientist explains that the consequences of interference are unknown and in fact, it would be best to simply wait until the ‘old’ one has been attacked, ensure that everything happened exactly as he remembers it happening, and then simply go back to his house as if nothing had happened.
He agrees, and goes to monitor the situation. Taking a lab truck and driving out, he sees the ‘naked woman’ (apparently not yet naked) and can’t resist chasing her, causing him to crash. She tends to his wounds, and he realizes that he is the pink-bandaged murderer—pink with his dried blood from the accident. He resolves to re-enact events, if that is what is necessary. This leads to him kidnapping the woman at knife (scissor-blade) point, making her strip nude, knocking her out, stabbing himself, and chasing himself. Worse, the woman wakes up, and he chases her into his house, only to frighten his wife into falling off the house’s roof, whose short brown hair he recognizes in the corpse on the ground; he is devastated by his wife’s death, but continues, to ensure the time-loop’s stability.
Once the first protagonist is safely away, the second resolves to go back again, and fix his mistake. The scientist tries to stop him, sabotaging the time machine by removing a key part; but he has begun to understand the power of time-travel, and he simply reaches out his hand at random and discovers the machine part laying in the grass—left by the third iteration.3 The scientist gives up and resumes cooperating, so the second protagonist doesn’t have to assault him.
How can that be possible, when he didn’t see any third self when he went back in time to the earliest possible moment? Well, there was still a 30s gap in the past before the time machine becomes disabled, so he aims for that, races out of the time machine into hiding, and watches the second protagonist emerge a few seconds later. (This is now the final loop: he cannot go back a fourth time, because he has already seen that there is nowhere a fourth protagonist could be hiding.)
As the third protagonist, he stalks the second protagonist who he just was, and arranges the destined car accident. Finally, he chases the woman through the house before the second protagonist can arrive to chase her—cutting off her hair while she is distracted. Then he chases her onto the roof—where the second protagonist sees her fall and mistakes it for his wife (who is in reality safe on the ground, being escorted away by the third protagonist). The second protagonist flees the corpse to finish the second time-loop, and the third sits down with his wife to wait for the police. The End.
Coverup Success
Does he get away with it? I think he does. ‘Timecrimes’ presumably require ‘timecops’, the latter of which do not exist (yet). So the regular cops encounter a beat-up victim with his ignorant wife’s alibi, and a few puzzling physical pieces of evidence like a crashed truck, but are otherwise left with a mysterious freak crime and a pretty young woman who died by misadventure.
The only person who could put all the pieces together is the scientist, but he now has ample reason to remain silent and not come forward to the police (who have no way to know he was involved, not even if they trace the vehicle to the research lab): aside from his initial culpability in enabling the homicide (possibly excusable since he wouldn’t know until much later), he must have committed multiple crimes in subjecting the protagonist to something as wildly experimental & risky as time-travel, and at a minimum, his career would be over if that ever got out. And once he remained silent, he is now an accessory to murder, providing a nasty dilemma: he may not be able to convince anyone of what happened, even by breaking the secrecy around the time machine, but if he did, he necessarily indicts himself as well.
So, I think the protagonist gets away with it.
Looping Computational Power
One interesting part of the TC setup is that the time machine functions like a wormhole, in that it connects only to itself, and doesn’t travel to arbitrary times like a classic H. G. Wells-style time machine does. The time machine has performance limits, and the scientist explains to the protagonist that it can’t send him back more times because it will run out of fuel. Further, he went back almost as early as possible; the second trip, he exits just seconds before his first trip arrives and must frantically hide in order to remain consistent.
So, this is why the third loop is the once-and-for-all loop: he can’t go back to patch up errors in the third loop like he did for the second loop, because there is no time machine fuel, and he also observed that there were no earlier selves exiting at the same earliest-possible time that he did.
This is an arbitrary limit, but lifting it points to two of the most important formal features of a stable time-loop story: a stable time-loop is like a murder mystery or a magic trick in that what you see is not what you get.
In a stable time-loop, your sense-perceptions are valid; it is your interpretations of them that are worthless. Your interpretations can always change. An earlier version of events can manipulate what you perceive, because it has temporal priority. This means that without having the earliest perspective, your understanding can always be rewritten. In stable time-loops, “one story is good, until another is told”—everything is always ‘up for grabs’ by a later loop.
Priority
The third time-loop has priority, as he is the ‘outermost’ loop: he cannot be ‘inside’ anyone else’s loop, because they would have to have gone back earlier than he did, but he came through the earliest possible time, and has logical priority over all other time-loops. Being first, he has time to set up interventions & manipulations for all inner loops (while those inner loops can only manipulate further inner loops). So, being the outermost loop is desirable, albeit possibly difficult to do: if the time machine has a fixed range, then every time one goes back a little earlier, one irreversibly uses up priority, and if the time machine has unlimited range, one faces practical issues like dying of old age4. (This raises different issues of precommitment racing from the Zeroth Maria/Endless Eight/Re:Zero serial time-loops, due to the self-consistency moving the precommitment races to a more abstract realm.)
However, higher priority time loops have a severe constraint: they must manipulate events in a way which does not change the observations of lower-priority loops.
This constraint might seem to render priority fatalistic, but this is not the case, as TC demonstrates.
The requirement to ‘preserve the appearances’ simply means that a looper must think like a stage magician and unsee their earlier loops.5 In TC, the protagonist did indeed see things, but as Sherlock Holmes might say, he did not observe. He is correct in that he did ‘see the dead body of a woman with short hair on the ground’; he was just incorrect in his interpretation that ‘he had killed his wife’. He had actually killed the young woman—his higher-priority future self could ‘unsee’ the observation, and went out of his way to cut the young woman’s hair to ensure that she looked like his wife before being led to the scene of the accident in order to die.
Entropy
Never Do Yesterday What Should Be Done Tomorrow.
If At Last You Do Succeed, Never Try Again.
A Stitch in Time Saves Nine Billion.
A Paradox May Be Paradoctored.
It Is Earlier When You Think.
Ancestors Are Just People.
Even Jove Nods.“The By-Laws of Time”, from “‘—All You Zombies—’” (Robert Heinlein, 195965ya)
So one might say that the problem was the second protagonist saw too much: if he hadn’t seen the woman’s body at all, it would have been that much easier for the third protagonist to swap victims. Conversely, if he had gone down from the roof to check her body up close, it would still be possible for later time-loops to fix his mistake, but it’s harder.6 By gaining information, he uses up his freedom to choose possible timelines, reminiscent of information theory like entropy.
This stacking can be indefinitely deep. There could be a fourth or fifth or sixth protagonist, going back earlier, each of them interfering with ‘earlier’ iterations to achieve their goal.
The power of later iterations is limited by what earlier ones have already seen and done; the more the first travelers notice, the less freedom the later ones have. (One can imagine an indefinite sequence of later iterations, all skulking around in the background, but powerless to affect the events as they unfold for the nth time, the first iteration having been far too inquisitive for his own good.) From a later iteration’s point of view, they will wish the early iterations had seen & done the absolute minimum possible, and spent the rest of the time hiding in a closet with earplugs & blindfolds on until they emerged & stumbled blindly into the time machine!
As an outer looper is a bit like a Cartesian demon for his inner loopers, we could take this perspective further. The ideal looper minimizes gain of information about his true environment; but he can ‘observe’ & ‘intervene’ in arbitrary things insofar as they do not provide reliable information about reality and could always still turn out to have been faked by his outer loopers. So for example, if he hid in his house and had no contact with anyone else, he could send & receive thousands of emails and believe himself commanding a large business-empire, as long as he could be commanding a business-empire or simply exchanging emails with an outer looper who is rolling back the business empire after it went disastrously wrong somehow, and who is actually doing something else. As long as he only sends & receives emails (which are so easy to fake), then he is fine and the stable time loop will turn out well. But the moment he does something else, like turn on the TV to see how his business empire is doing, or he takes a video-conferencing call, suddenly it becomes much more difficult for his outer loopers to maintain the fiction and roll back the business-empire attempt, and it begins to destabilize, and some other equilibrium (perhaps one not involving him at all) may be chosen instead.
We can take this much further. Perhaps the optimal looper is something like a time traveler who takes drugs to knock himself unconscious before being sent back, whereupon (before regaining consciousness) he is plugged into a Matrix-like cyberworld by a brain-computer interface; inside this Matrix, he receives messages containing short, heavily-redacted, text snippets and crude, low-resolution images or videos with ‘obvious’ AI-generation artifacts, and sends back orders to underlings in the outside world; every minute, a new version arrives and is plugged in, and starts being sent messages & giving orders; periodically, a version is woken to be sent back as the new outermost loop, and shown a real message, and can decide whether to use its next outer loop to override all previous decisions (replacing the original messages, which were real in that inner iteration, with fictional & AI-generated versions in all outer iterations). Unbeknownst to the plugged-in travelers, by the end, almost all of them are not giving real orders, they are simply being fed fiction and playing an elaborate computer game, while a relative few outer-loopers execute the final perfected plan, whatever that is. Whatever is going on in reality, it bears as little resemblance as possible to their game. But while almost all the time-traveler instances are badly deluded, the stable time loop as a whole has defeated its enemies and achieved its goal, because every time there was a problem, the time traveler was not ‘locked into’ anything, he could always restart (almost) completely from scratch, eliminating most of the restrictions of stable time-loops compared to regular time-loops—because everything the time-travelers had ever ‘seen’ was mediated through the cyberworld, and could just be explained by that most annoying of fictional story endings, “and it was all just a dream, The End”.
This is a different perspective from previous analyses of the computational power of time travel, like 1991. Moravec focuses on standard computational complexity issues, and concludes that time travel loops can solve NP problems but not uncomputable problems. This would have many consequences almost as profound7, but his consideration is limited to self-contained time-loop calculations—a single-player computer in a box, essentially.
From the multi-player view, the critical quantity looks like it’s less the computational complexity class of a problem than it is the information possessed by agents. The more information an agent has, the more it can do in the current time-loop—but the less it can do in future time-loops, having tied its hands. This implies that time travelers must choose carefully when to look: “what you don’t know can’t hurt you”, because you always revise it in an outer loop. It’s like budgeting: spend your ‘information currency’ too early, and you may run out when you need it most.
Perhaps it becomes a Stackelberg game, where there is a race to precommitment? The agent best positioned in the ‘base’ reality can then spend less entropy in each iteration, ensuring that they will probably become the outer loop; this renders fighting them futile from the start, as one has always already lost.8 Alternately, if neither is able to precommit adequately in the ‘first move’ by making themselves the outermost loop, conflicting successive time loops might resemble an alternating game like Nim: each player can ‘remove’ an arbitrary amount of freedom from the loop as a whole by making observations, but the goal is to be the last player to have some freedom with which to decide the final meaning of the loop.
Time Travel Calvinism
Horatio: You will lose [the duel with Laertes], my lord.
Hamlet: I do not think so. Since he went into France, I / have been in continual practice. I shall win at the / odds; but thou wouldst not think how ill all’s here / about my heart. But it is no matter.
Horatio: …If your mind dislike anything, obey it. I will / forestall their repair hither and say you are not fit.
Hamlet: Not a whit. We defy augury. There is a / special providence in the fall of a sparrow. If it be / now, ’tis not to come; if it be not to come, it will be / now; if it be not now, yet it will come. The / readiness is all. Since no man of aught he leaves / knows, what is’t to leave betimes? Let be.
Hamlet (Act 5, Scene 2, Line 223)
By the end of Timecrimes, we understand both the power and limits of stable time loops if we have been paying attention. So we understand how the protagonist made the time-loops happen, and repair his mistake and instead murder the young woman.
What we do not understand is why. The one question that is never asked inside TC is: why this time-loop? There is no answer at the end explaining why the protagonist had to undergo this peculiar sequence of events (regardless of the internal logic), nor do we ever hear the protagonist’s true motivations (only what he says to manipulate other people). In this respect, it is like many later murder mysteries (eg. Umineko): while we are caught up with solving the ‘whodunnit’ or ‘howdunnit’, we forget the ‘whydunnit’.
TC can be compared to serial loops like Back To The Future (where each iteration follows a standard ‘forward’ causality’ and the total set of loops is revealed), or simple stable time-loops like Bill & Ted’s Excellent Adventure where they exploit the stable time-loop powers in modest ways. In each of those cases, the final time-loop either is the straightforward result of the repeated time-travel, or the deviations from the base time-line are modest enough that we can see how the iterations could have proceeded without the final set of time-loop interventions (eg. if they find keys under a bush, they could have stolen the keys instead, or someone forgot to lock the door), and we can imagine how they would eventually converge in the final version.
TC is unlike any of these. The stable time-loop in TC has no plausible predecessors and could not have evolved. There is not even a ‘base’ timeline where the protagonist goes back in time on his own. He knows nothing about the laboratory or the scientist, and vice-versa, and no motive; even if he tried, the scientist has many strong reasons to not send him. Even if they did all that for some reason, the young woman has no reason to be involved, as she is bicycling through the area only briefly on the way home, and has no intention of going into the woods (much less stripping nude and giving a peep show). There is no one who might serve as a temporary murderer, pink-masked or otherwise. And so on.
This time-loop is not merely a stable time-loop, it is a ‘Garden of Eden state’: the stable time-loop is self-consistent, but there is no way to get from here (the ordinary retiree enjoying a weekend) to there (the triple loop). No pink-mask murderer means no panicked flight into the research facility, no encounter with the scientist, no forcing him to send iterations back in time, and so on. Remove any element—any!—and the entire thing collapses in a poof of logic. (qntm criticizes Timecrimes for being unlike Bill & Ted in this respect, with the reason for the crimes left as “big, floating question marks, to which the only possible answer is ‘mu’.”)
This is the classic ‘Shakespeare time loop problem’: as a matter of logic, it is impeccable for Shakespeare to write his works using a copy of the collected works of Shakespeare from the future, which will eventually be printed based on his writings; all laws of physics are obeyed, all causality makes sense examined locally. But one still feels this bootstrap violates conservation laws & metaphysical principles like the principle of sufficient reason: how did you get a lifetime of literature ‘for free’? ‘Where’ did it get created? If it was an ‘iterative’ process, how did that work? Did Shakespeare add 1 play to the end of the manuscript in each loop as fanfiction, or what? If the whole set of plays can be created ex nihilo, why did our Shakespeare’s plays specifically get created and not the plays of another fellow of the same name, drawn from the near-infinite possibilities of human literature? (And in a scenario like Heinlein’s “All You Zombies”, where the protagonist turns out to be his own father & mother in a time-loop, where did that human come from? In fact, how do you know they are human? They could be any kind of robot or alien species or mysterious third thing, as long as they look human enough to preserve the loop.)
Perhaps it simply picks a random equilibrium? Or perhaps it is a multiverse where each equilibrium gets instantiated in rough proportion to its internal ‘likelihood’? These would look the same, on average, and in either case, the ‘solution’ to the Shakespeare time loop problem is simply that Shakespeare was the most talented writer around and had a propensity for great writing, and so many random equilibria involved collected works worth republishing centuries later; the logic is that it could happen… so it does happen.
In stable time-loops, “possibility implies actuality”.9
With this in mind, we can ask again: why did this protagonist get trapped in that time-loop, and not, say, his wife? The key, I think, is that the protagonist does not seem upset at the murder or any of the other timecrimes, and he appears to have every intention of covering up the crime to continue his ordinary retired life. A sinister undertone creeps in to his casualness in executing the scenario: he goes along with it too easily. “He does it because he can” is the glib answer… but this is in a stable time-loop with self-fulfilling prophecies. What does ‘because he can’ mean there, exactly?
In the case of the protagonist, presumably if he wasn’t so sociopathic and couldn’t’ve done things like stab himself or knock out the woman so cooly, then the time loop would be logically impossible and collapse, and then he would never be faced with the choice to begin with. The protagonist, faced with the choice of committing crimes to maintain the time loop and save his wife, finds himself the sort of man who is morally flexible enough to do so… so, he does so.
This presents a horrifying view of the universe, as running on a perverse physics of Calvinist predestination: you are saved or damned from the beginning of time(-loops), because your innate traits which make you immoral cause the scenario in which you would succumb to evil.10 To the extent that there are scenarios in which one commits crimes of some sort, or the weaker one’s moral fiber is, the more likely one is to be trapped in a damnation time-loop as the fixed point; and the longer one spends in the vicinity of the time machine, under more circumstances, the more possible scenarios there are, and the more likely one will be to involve a time-loop.
In Timecrimes, a time machine is not yet another d—ned time machine—it’s a damnation time machine, a machine for mass-manufacturing timecrimes & epistemic chaos. The protagonist (and everyone else?) had simply enjoyed the moral luck of not being near a time machine—until now.
Time Machine Evolution
I know where I came from—but where did all you zombies come from?
…You aren’t really there at all. There isn’t anybody but me—Jane—here alone in the dark.
I miss you dreadfully!Protagonist, “All You Zombies”
In Timecrimes, no one benefits from the time machine existing, including the protagonist, and everyone would be better off with it destroyed and never used again.
Ah, but how does one destroy a damnation machine? The time machine in TC won’t be destroyed, because that would require admitting to the crimes; and the one attempt to sabotage the time machine is immediately subverted by a time traveler who requires the time machine. If a single ignorant individual with no resources can stumble into a time machine and get away with murder, what are the limits, exactly, to what could be done by a knowledgeable individual—or large groups of individuals?
Here we start touching on territory that time-travel SF has been reluctant to tread on, perhaps for the same reasons that SF is reluctant to write stories about the Singularity or superintelligences: if you start to take it seriously, what sort of narratively-satisfying story can you tell?11
In Timecrimes, the protagonist manipulates and bullies the scientist into sending him back. He can do this because he has foreknowledge and is motivated to save his wife. His advantage is limited, however, because he went back only a few hours, and is confined to a tiny geographic area and unable to communicate; the scenario is an elegant locked-room murder mystery. But what if he had more scope for action? What if he had gone back a month, with some old newspapers with stock & lottery listings? What if, when he returned, the scientist refused more strongly…?
Obviously, any stable time-loop which has multiple iterations means that a time-machine must exist from the beginning to the end of the outer-loop: you can’t go back in a machine which has been destroyed. Someone trying to destroy the time-machine can block it from turning on, because no time-traveler can reach them there and stop them; and they can block it after the outer-most loop ends for similar reasons; but they cannot destroy it during the outer-most loop because them failing to do so is logically required for the loop. Thus, if they can fail in any way, they will fail; ‘fail’ here can be anything from ‘tricked into thinking they destroyed it’ to ‘killed’ to ‘another replacement time-machine is constructed’.
So to avoid that, they must be able to defeat every possible time-traveler who might create a stable time-loop, and destroy the time-machine in all time-lines. Any timecop who intends to destroy a damnation machine is now fighting arbitrarily many desperate zombiefied time-travelers, who are both know how the machine was destroyed and are seeking to preserve their loops.
In a situation with sparse scenarios to sample from, like an empty countryside on the weekend with no one there, probably most equilibria will have 0 time-travelers, and the damnation machine can still be destroyed after it has been turned on for the first time. However, what if a time machine was turned on in the center of a city?
A time machine is more devastating than any nuclear bomb to its surroundings, because at least the damage could be repaired afterwards, while a time machine precludes any possibility of undoing itself.
Such an installation could no more be undone than the historical fact of having dropping an atomic bomb: instantly, the outer loop comes through with the highest priority, representing the ultimate combined power of all time-loops in the final stablest equilibrium. Inside a city with its millions of inhabitants, any of whom could be a looper, one is suddenly fighting the maximum-possible ingenuity & ruthlessness of hundreds—thousands—millions of protagonists, all dedicated to a convergent instrumental goal of ‘preserve the time travel machine’ and able to recruit allies & acquire vast resources with their foreknowledge. This incentivizes ever more extreme tactics: if you are unwilling to commit a crime or sin which would be useful, there is another version of you, or another time-traveler, who could, and so now does.
If it is possible for even a single person to go through and thus possibly causing others to go through once they realize they need allies to defeat attacks and so (possibility implies factuality) multiple people are looping, dropping an atomic bomb on the time-machine would be inadequate—the loopers will have already relocated or rebuilt it. Gradually, the region around the time-machine becomes distorted: causality itself warps, and you can only take actions which help the time-machine & loopers, because any other action would eventually impinge on them, be manipulated by them, and anti-time-traveler timelines erased as non-equilibria.
Conflicts between loopers do not destroy time-machines but propagate their seeds, both spatially and temporally. Loopers want more time-machines, going back earlier, as they strive to gain priority over each other and amass enough practical power that they can achieve their goals before running out of information.
Of all possible equilibria, the original one of zero time machines is the rarest and thus least likely.
This holds true on the higher level of all time machines: they evolve to persist and spread as packages of time-machines & loopers. Any time machine is a threat to other time machines, and loops will inevitably expand in scope from the earliest possible time any time machine can reach by proxy (which includes time-travelers sending electronic messages across the world): there can only be one outermost loop. And all time machines must have a place in the outer loop, as some sort of ‘time machine civilization’/‘ecosystem’, or the equilibrium is meta-stable at best, because they all could subsume each other.
The time machine civilization is the next level of replicators parasitizing human hosts, insidiously evolving at high speed in super-temporal ‘logical’ time rather than mere ‘temporal’ time, ripping up all cultural restraints & traditions, hacking security effortlessly, mindlessly ascending the gradient to complete control of the lightcone. Collectively, damnation machines are an invasion of non-conscious techno-superintelligences from a barely-possible future, bootstrapping themselves into existence from their enemies’ resources.
Thus, inventing a stable time-loop time-machine would be the last invention humanity made of its own free will (whether it realized that or not). And a time-traveler in a stable time loop universe may not realize it either, but he has always already been in hell, because nothing human makes it out of the near-future—or the near-past.
-
If you are harmed by your future self, it seems reasonable to regard it as a crime, if only for the minimal reason that while it’s not illegal to cut yourself or commit suicide (in most jurisdictions), such instances of self-harm at least involve the prior granting of a consent (you intend to harm yourself and then you do), while in the usual time-travel you have not (yet) consented. However, in a stable time-loop universe, if you are stabbed by your future self, then in some logical or B-time sense, you have ‘already’ consented to this, or else it couldn’t’ve happened.↩︎
-
The scientist explains that it isn’t scheduled to be tested for months to come; this dilatory attitude to the most important invention ever, and the lab’s emptiness, is a nice touch of realism (and highly consistent with the movie’s contemporary setting on the museum continent).↩︎
-
Similar self-fulfilling-prophecy causal powers play a starring role in Bill & Ted’s Excellent Adventure; cf. Good and Real.
This is a special-case of a general principle: “a stable time-looper is never at a loss”. He simply assumes that there is one where he needs it, precommits to putting it there, and there it is. Unlike an irreversible-looping time-traveler, he cannot permanently “lose” or “break” his time machine, because he can simply bootstrap a replacement time machine.
Even if there is only one possible time machine (so a future time traveler can’t simply drop off a replacement), we can imagine chains of time-travelers: the first ‘trapped’ time traveler simply assumes another time traveler will arrive and give him the time machine to use, and then is himself relieved by another time traveler (possibly himself that much older), who is then also relieved and so on—possibly to infinity, so no specific time traveler ever wastes much time ‘trapped’, but no additional time machine is ever conjured up either.
In the finite case, the series might terminate with the time traveler on his death bed, traveling back to die shortly after the time machine breaks; in the infinite case, it becomes a version of Hilbert’s paradox of the Grand Hotel, and there is no termination as there is always another time traveler coming back & being relieved.↩︎
-
Particularly if accident, recklessness, or antagonists mean that loops start to span decades—it would not be difficult to quickly push the necessary subjective time past the maximum human life span. It is true that human longevity appears essentially stochastic and the annual mortality levels off in the ‘mortality plateau’ at ~50% annual mortality, so given some modest selection effects similar to quantum immortality, a looper may be able to live well past Jeanne Calment… but the improbability will keep building up, and depending on how equilibria work, too long a series of loops may guarantee failure rather than see one live to age 300.↩︎
-
eg. Umineko’s systematic exploration of loopholes in the locked-room mystery genre demonstrates the countless implicit assumptions that readers make about a locked-room, ranging from locks not being swapped out to rooms having roofs.↩︎
-
He could, for example, use the same substitution trick, but this time, drug his wife unconscious (alcohol is easily available), carry her to near where the young woman will fall, wait for the second protagonist to look at the young woman’s body from a distance, and then swap his unconscious wife for the young woman’s body before the second protagonist comes down to look up close.
Here too, the more the second protagonist sees, the harder it gets to fix mistakes: what if he is more careful and checks her pulse to see if she is merely unconscious rather than dead? Well, there are things like wax statues or hyper-realistic masks, as well as drugs that can put humans into comas so deep that they would appear dead even when checking their pulse (to name the first ideas that come to me in a second of thought). Difficult, but all doable—merely a matter of time…↩︎
-
It would mean we live in Impagliazzo’s “Algorithmica” world where you can in practice do almost any of the things you would want to do if you could solve halting problems; for example, you can ‘solve’ any halting problem by simply including an upper bound on iteration count so high that you wouldn’t care if a program terminated after that iteration limit.↩︎
-
How does a non-looper defeat a time-looper? (In the previously mentioned case of serial time-loops, there are ways for the non-looper to ‘wear out’ or ‘outwait’ the looper, but these do not apply in the case of stable time-loops.)
For example, in Higurashi, the antagonist possesses enormous advantages like command of a heavily-armed military force, secrecy, a devastating bioweapon which has infected the characters and so on, while her enemies are a few middle school students; however, given the advantage of looping, this is still not enough to ensure victory, and she is defeated by the looper.
The usual suggestion proposed by authors is that a looper can only be defeated by massive overkill: ensuring that they cannot figure out nor arrange any possible way out, because being loopers, they will just keep going back until they find a loophole.
An example would be the serial webfic Worm, where the villain Coil is able to engage in limited stable time travel by making a binary choice which causes two possible timelines, and then he can choose which one becomes real; so for any choice, he can take a safe and a risky action and keep the winner (or gather information through an irreversible action in a discarded timeline, like torture). He is ultimately defeated by arranging a meeting where his soldiers have been bribed in advance, and convincing him, as part of a deal, to check his alternate timeline, collapsing it into a single present, right before they betray him.
Another serial webfic example would be Worth The Candle, which features “revision mages” who are very short-range loopers and challenging but vincible; but also more dauntingly, unicorns, which branches many timelines over a range of 3 seconds and then picks the ‘best’ one. The Unicorn is defeated by covertly partially immobilizing it for >3 seconds, long enough that it cannot undo the immobilizing when it realizes that it’s been caught, and then attacked with such overkill that its best-possible defense & enemies’ worst-possible offense still kills it.↩︎
-
“Time Loop Software: A somewhat puzzling debugging session”, Marak 201311ya dramatizes the experience of trying to program in a stable time-loop universe (cf. retrocognition):
↩︎In order for time loop logic to return an answer instantaneously, we must ensure that the problem will run long enough into the future to actually calculate the result. If a problem takes 60 seconds to solve, the program must run for at least 60 seconds. Time-loop logic does not violate causality. We are able to retrieve the answer instantly because we have committed to spending 60 seconds in the future calculating the answer and sending it back.
This turns debugging time-loop logic into somewhat of an impossibility. Any bugs in a time loop indicate that sometime in the future a problem has occurred. This event may or may not be related to software.
Imagine a computer that utilized a time loop to brute force crack passwords (as our code posted above did). I turn the machine on and request it cracks the password. The program doesn’t work. Frustrated, I turn off the machine and complain to my co-worker Josh.
Josh turns on the machine and requests the password. The software works instantly cracking the password in under 1ms.
Bewildered, I ask Josh why the machine worked for him but not for me.
Josh replies, “It’s actually quite simple. Using that computer it’s going to take ~400 hours to brute force the password. After that 400 hours the CPU must recursively return the cracked password back in time until it reaches right now. I was able to get the answer instantly because I have decided to not turn this computer off for another 399 hours and 59 minutes. Simply put, you turned off the computer too quickly.”
-
See also Bakker’s Second Apocalypse. In it, the villains are able to corrupt even the greatest heroes by using their advanced technological devices to show the heroes their future afterlife in Hell, which exists atemporally outside the universe. And because the heroes always see themselves in Hell, implying that they have already been damned by the gods to suffer forever despite their heroism, they are morally outraged and terrified; and they join the villains to try to cheat death.
So the question becomes: were they “always” going to be damned by the gods regardless… or were they only become damned because, if they saw themselves in Hell, they would rebel and thus damn themselves?
And if they had been willing to go to Hell “unjustly”, would they not be seen in Hell using the device, and so the villains would never try to show them it—and would never be corrupted & damned in the first place?↩︎
-
Primer stands as a warning here: does Primer use the dynamics of priority & entropy with its morally-challenged protagonists, and converge on an outer-loop? I honestly can’t tell.
Primer ends on an ominous note of an entire warehouse being turned into a time-machine, presumably to send back many agents & nested time-machines, but it is unclear if the loops have reached an equilibrium or if Primer is perhaps illustrating a pseudo-timeline (or fragments from multiple timelines) which is still iterating to a final stable equilibrium but has not yet reached it.
The sheer opacity & complexity renders it a deeply frustrating watch: any serious stable time-loop story quickly becomes a puzzle, and watching puzzles with no solution is no fun. (See also “Suzanne Delage”, where few readers seemed to genuinely enjoy or admire it on an esthetic level because it was a puzzle which defied all analysis for 40 years.)↩︎