I saw this old closed question on Physics.SE and thought it would be perfect for here. The original author is long gone, so I thought I'll just have to post it myself.

We can consider factors of plot and literature, not just hard science. But I do intend for it to be hard-SF.

In Greg Egan's The Arrows of Time he had a crew visit a region where the thermodynac arrow of time was in the opposite direction. The ambient sunlight was not visible to them and they would unmake footprints.

But I think that's not right. Consider what would happen if you prepared such a region by somehow reversing the momentum of every particle. It is very precarious with every gas molecule having to be exactly right! As soon as you observe it, you peturb that delecate state and time runs normally from that point forward.

Consider an experiment where a gas cloud in a large isolated chamber will rush into a bottle, because the state of every molecule is just so. If you go into the chamber, it won't work anymore because you mess up those perfect trajectories. (Unless the trajectories were chosen with knowledge of your exact effect in mind...!)

The Setting

Suppose you had a wormhole that led to another part of the universe where time ran backwards or the orientation of the exit mouth were such that it dropped you into the time dimension facing the wrong way.

You don't experience the environment to be anti-matter: as in Egan's story it would be anti-matter if your time arrows were aligned, so you have matching matter polarity but opposite entropy gradiants and (whatever that means if it's more than just entropy) opposite time directions.

What happens? How?

Maybe you are like the gas bottle cloud and the large environment at your destination peturbs you. Does time do funny things? Do time and entropy break up into separate concepts? Could you go down in an elevator that's ascending, escaping in the same ride that the police use to arrive at the scene of the crime?

  • 1
    $\begingroup$ Diffusion until the entropy matches up? $\endgroup$
    – King-Ink
    Commented Jan 24, 2016 at 13:23
  • $\begingroup$ Of course, science-based is for sure. $\endgroup$
    – JDługosz
    Commented Jan 25, 2016 at 15:44
  • $\begingroup$ I'm pretty sure if you did this by /wormhole/ the wormhole time arrows must align to avoid an explosion. $\endgroup$
    – Joshua
    Commented Jan 15, 2019 at 20:25

5 Answers 5


The title of your question and the question you're asking seem to be two different things: In the title you're asking what would happen at the interface, but in the body of the question you ask what it would be like to experience such a world.

I can answer both: You wouldn't be able to experience such a world because the interface can't happen. Or more to the point, if it did happen, nothing would be able to cross it.

Consider a person trying to cross the boundary. In our universe their muscles are pushing their flesh. In the other universe? Their muscles are following. On the boundary? It's one of two things: Either there's a continuous (but tiny) change from one set of thermodynamic laws to another, or there's a straight discontinuity where the two universes meet.

In case 1 it's fairly obvious that the boundary is an area where time doesn't travel in either direction. If time isn't flowing: you can't cross the boundary, so you can't get to the other universe past the wall of particles that are undergoing an asymptotic temporal dilation as they approach the boundary.

In case 2 there's going to be an interesting moment when you cross the discontinuity. The cause of your skin crossing the boundary (the effect) will have to happen after the effect in the new universe. Sadly that can't happen, because the thing causing that effect (in this case a push from your muscles) is still in the wrong universe where time is going the other way. In this case the boundary causes causal violations on both sides, as particles try to cross but can't as they can't move over the boundary until they've already crossed on one side, and can't cross the boundary until they've moved over it on the other side. Confused? Not half as much as you would be at the boundary. By which I mean every quark in your body will be in a confused jumble floating back and forth at the boundary. They can't move back in this universe and can't move forward in the other one.

Essentially: Anything that has a temporal component can't work as it crosses the boundary. Sadly that includes everything (including spacetime. I can't even picture the Minkowski diagrams for this one.)

  • 1
    $\begingroup$ This takes the view that there's "something" about the direction of the arrow of time beyond the low-level physics that imposes a direction, as opposed to the laws being symmetrical and the direction caused simply by entropy. For point 1, you can just "tunnel across" the barrier :) $\endgroup$
    – JDługosz
    Commented Jan 25, 2016 at 15:50
  • $\begingroup$ @JDługosz: If it's identical laws but with entropy running in opposite directions then crossing the boundary violates at least one set of thermodynamic laws. As for point 1: I'm not sure you can tunnel through it. The temporal weirdness going on here means that not only is our boundary infinitely high (from a quantum mechanics point of view) it's also infinitely wide (as it would take light an infinite amount of time to get to the boundary). Either way, I'll let you try cross it first! ;-D $\endgroup$
    – Joe Bloggs
    Commented Jan 25, 2016 at 16:32
  • $\begingroup$ Plot trumps physics. It needs to sound plausible to be good SF and follow the modified rules carefully. Better-than-just-handwaving to get accross the barrier is a good plot element. I imagine it like pushing through a soap film, that instantly moves around the other side of the object pushing through when it's stressed enough. $\endgroup$
    – JDługosz
    Commented Jan 25, 2016 at 16:53
  • $\begingroup$ @JDługosz: True enough! Quantum tunnelling is a fantastic plot-hole-filler for sci-fi stories! $\endgroup$
    – Joe Bloggs
    Commented Jan 25, 2016 at 18:46

Sir Arthur Eddington introduced the concept of time's arrow in thermodynamics. This is what he had to say about it, quoted from Wikipedia Arrow of time:

  • It is vividly recognized by consciousness.
  • It is equally insisted on by our reasoning faculty, which tells us that a reversal of the arrow would render the external world nonsensical.
  • It makes no appearance in physical science except in the study of organization of a number of individuals.


  • so far as physics is concerned, time's arrow is a property of entropy alone.

From this, we gather that the arrow of time is just a vivid metaphor for entropy. It has no real physical existence apart from its alter ego, entropy.

Therefore, in the alternate universe, entropy is reversed. Nothing happens there: everything unhappens.

So far, we've been discussing this unhappening at a macro level, except for a passing reference to quarks by Joe Bloggs. But is it conceivable at the level of fundamental particles? For example, all fusion must unhappen over time, leaving a universe of hydrogen. What would a time-reversed Schroedinger equation lead to? The implications are so bizarre that they make string theory look like elementary arithmetic.

LR's answer opines that two arrows of time pointing in opposite directions can't coexist. Since all our equations of physics are time-reversible, if one exists the other must exist. Therefore the only resolution to this paradox is that there is no such thing as an arrow of time. There is entropy, of course, but it would seem there is no way to reverse it: the genie just won't go back into the bottle.

Joe Bloggs' answer, considering what would happen if two areas with contrary arrows of time were to abut, seems to approach the same conclusion: there can't be an arrow of time. I think this is an interesting line of thought to pursue - and that's an understatement.

  • 1
    $\begingroup$ Re time-revered equasions: but the point is that the low level physics, e.g. Shroedinger's equasion, is time symmetric. A time-reversed version is the same thing! $\endgroup$
    – JDługosz
    Commented May 17, 2016 at 17:49
  • $\begingroup$ That's the thing that worries me about that equation, but delving into that would be way off topic. $\endgroup$
    – frank
    Commented May 17, 2016 at 17:55
  • $\begingroup$ "What would a time-reversed Schroedinger equation lead to? The implications are so bizarre that they make string theory look like elementary arithmetic." it would be no different. Is this paragraph meant to indicate the lack of an arrow of time? That's not clear, as it seems to contradict your following statements instead. $\endgroup$
    – JDługosz
    Commented May 17, 2016 at 18:02

Crossing the interface

@JoeBloggs is right there are two scenarios for crossing the interface between two universes with opposite thermodynamic arrows of time. His case 2, where the interface is sharp discontinuity, requires anyone passing to do so quickly. Otherwise the person would be a mixture of matter and antimatter. The model of quantum tunneling through a soap film like wormhole would make for a safe passage.

What would this look like? This can be demonstrated with a thought experiment. Assume the interface lies at the mid-point of a tunnel. A small vehicle A is used to cross from one universe to another. It travels at one metre per second and the tunnel is twenty metres long. When the vehicle enters the tunnel its passengers we see an identical vehicle B enter the tunnel at the other end and moving towards them backwards. The people in the backwards will be waving hello to those in the vehicle from our universe.

The two vehicles approach each other at the mid-point and instead of colliding from the point of view of vehicle A vehicle B will vanish from in front of them while immediately reappearing behind them and heading backwards towards the entrance where vehicle A had originally entered.

Vehicle A continues forward and goes through the tunnel mouth leading the reversed universe as they do they wave goodbye to the people in vehicle B.

Obviously vehicles A and B are the same vehicle viewed at different parts of its worldline and in close proximity, spatially, to each other.

Case 1 is much more interesting. Here crossing the interface will happen gradually in small degrees. This reminded me of Kurt Godel's rotating universe solution to general relativity where a spacecraft travelling through its spacetime will gradually have its light cone tipped over until its relationship to past and future are reversed. This reversal of time is relative to the starting point of the Godelian spaceship. To anyone on the spaceship time will still go forward as if nothing had changed.

Details of the Godelian universe and its metric can be found here: https://en.wikipedia.org/wiki/G%C3%B6del_metric

This seems plausible. The interface between the two universes in case 2 would most likely be vast, perhaps several light years across. Perhaps, it might be as big as a universe in its own right. In that case, crossing between two universes with opposite directions in time will include an intermediate universe. Anyone travelling between the universes has to cross this intermediate universe before going from one universe to another. It would need something like this, and possibly as big as another universe to allow the progressive tipping of a traverser's light cone in being able to cross the interface. In a SF story the writer can scale down the size of a Godelian spacetime as interface for the purposes of the plot.

Backwards O Time, Backwards

There are no surprises here. Once in the time reversed universe everything seen exactly the same in their home universe. Time would appear to go forward in exactly the same way. Entropy increases as usual. If the traversers had a viewing devices to observe what was happening in their home universe it would appear to be running backwards. This omits any possible effects of relativity on the viewer. Things become more than difficult considering how this might work.

If the traversers could cross the interface back their home universe, there might be time travel. This is more likely with an abrupt discontinuity. Whereas with the case 1 huge interface the time taken to it might cancel out any possible time travel.

  • $\begingroup$ @JDlugosz Thanks for the edit. Glad to have my post improved.. I'll check out the toolbar. $\endgroup$
    – a4android
    Commented Jun 30, 2016 at 8:05
  • $\begingroup$ And welcome to Worldbuilding! $\endgroup$
    – JDługosz
    Commented Jun 30, 2016 at 8:09
  • $\begingroup$ @JDlugosz Thanks for the welcome. And for posting such a delicious question. Everyone's answers are fascinating and challenging in the best possible way. $\endgroup$
    – a4android
    Commented Jul 1, 2016 at 6:37

There are actually two scenarios differing in “polarity” with which two space-times would be connected.

If, when traversing the interface, a world line faces the reversal of thermodynamical time direction (but possibly travels the same coordinate time direction), namely:

calendar:   Monday   Tuesday   Wednesday
(our world) →→→→→→→┑
(antiworld)             ┖←←←←←←←←

, then Joe Bloggs is right: no one will be able to traverse the interface. The conscience (as well as physiology) is not able to run towards decreasing entropy, so a human travelling the way pictured above will essentially die. Of course, Ī̲ dismiss the possibility that a human body can maintain its own direction of time that defies the one of the surrounding universe — that’s anti-materialistic, at least.

Another possibility is the interface that we cross along the Arrow of Time, but reversing coordinate time direction, such as:

calendar:   Monday   Tuesday   Monday     Sunday    Monday
(our world) →→→→→→→┑                   ┎→→→→→
(antiworld)             ┖→→→→→→→→→→→┙

There is no death upon transition, but implications will not be very different from the usual “travel to the past” stuff. If two-ways crossing is allowed, then mathematically, the first thing to be noted is that the united space-time of two interconnected universes isn’t causal (there are closed time-like curves and no past–future partial ordering).


In the universe that you are describing effect comes before cause, and enthropy always decreases. This is the universe where you would be stealing because the police is chasing you. Gas returns to a bottle because it is its natural behaviour. For the inhabitants of this universe, rain after the wet ground would be completely expected. They would not perceive it as abnormal.

I cannot imagine what a person coming from our universe would feel like there. Supposing he survives the journey, I think he would start functioning according to the local laws of physics. Imagine, if he comes to a place where gravity is repulsive, he might be surprised but he would still fall up. And in a place where time flies backwards the very neurosignals would be sent backwards. He would maybe be able to sonsciously understand cause before effect, but for him them it would be unnatural. He would just go along with the flow.

On large scale, I think that if two universes with different time arrows would meet, it would cause some very grave consequences. I don't know whether complete anihilation or the two would somehow mix in some very weird way, but it seems impossible that the two would be able to coexist.

  • 2
    $\begingroup$ I think brains would run in reverse too, so it would seem exactly the same. $\endgroup$
    – JDługosz
    Commented Jan 24, 2016 at 17:07
  • 1
    $\begingroup$ @JDługosz I was thinking the exact same thing, they both experience time normally from their perspectives, and both would see the other universe as the reversed one. Also while it may or may not be obvious, this scenario allows fairly easy time travel, pop into one universe and then back to get to the past just thought I should point that out. $\endgroup$ Commented Jan 24, 2016 at 18:47
  • $\begingroup$ @JDługosz I actually don't think they would. If the brains run in reverse, it just means information flows the opposite way. The brain feels pain, sends a message to the hand, the hand touches the stove. And from their perspective it would be quite natural to put the hand on the stove after it gets burned. But in any case, the perception of pain came first. Then again you might consider time like a book that can be read either beginning to end or end to beginning. In that case you can say our perception of time is just arbitrary. In this case we ourselves might be going from effect to cause. $\endgroup$
    – L.R.
    Commented Jan 24, 2016 at 18:56
  • $\begingroup$ You seem to be imposing a perception of time or before/after labeling in contrast to the events occuring in the physics. You are saying that some external imposed arrow exists and can be at odds with the thermodynamic arrow, whatever that means. $\endgroup$
    – JDługosz
    Commented Jan 25, 2016 at 8:42
  • 1
    $\begingroup$ sends a message to the hand why is the nerve signaling not reversed? Why is burning skin not reversed? The chaotic mess of protoplasm releases oxygen and pulls itself into organized cells, with the excess kenetic energy flowing to the hotter coil, then the brain removes the memory of being in pain by pulling ions up the emitters and restocking the potential. $\endgroup$
    – JDługosz
    Commented Jan 25, 2016 at 8:48

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .