I have thought about it and currently it is mostly considered to be an absolute fact that time is uni dimensional and only goes one way, or naturally only go one way, or forward, but if time weren't naturally going only one way, but forward and backward naturally, what would be the implication on causality?

Causality implies a unidimensional time that only goes one way, but what about a unidimensional time that go either way? How would causality work in such a world where time is bidirectional?

  • 1
    Causality does not imply unidirectional time as time travel stories attest. Moreover, one can set up mathematical equations where events at later times affect those at earlier times and study the consequences. This has been done by some physicists, including Wheeler and Feynman, see Retrocausality and Retrocausality in Quantum Mechanics. Of course, the idea is speculative and controversial, and lacks any compelling applications at the moment.
    – Conifold
    Commented Apr 25 at 1:01
  • 1 Earth. Going 1 direction. (While also rotating 1 way. We can state where Earth was. We can predict accurately where the 1 Earth is going, and where it will be when. Any model of "time"... that doesnt go 1 direction... would have to explain why there is not more than 1 Earth. Commented Apr 25 at 1:40
  • If time is bi-directional, one could not distinguish causes from effects. The two concepts would be equivalent.
    – Dcleve
    Commented Apr 25 at 2:59
  • 2
    @sayaman I've edited the question and removed references to multiple time dimensions, since they were not necessary in the question. I suggest posting a different question about the dimensionality of time.
    – RodolfoAP
    Commented Apr 25 at 3:37
  • 2
    "it is mostly considered to be an absolute fact that time is uni dimensional and only goes one way" that's completely wrong, in a sense. In the current basic theory of physics, Einstein's general relativity, it "goes in both directions".
    – Fattie
    Commented Apr 25 at 10:46

5 Answers 5


I think you are mixing notions of time and causality, as in metaphysics or psychology, with the physical notions of time and causality.

The laws of physics are indeed insensitive to the direction of (physical) time: play the billiards movie back, i.e. reverse the flow of time, and it still is a physical system, in particular, it still satisfies the (physical) laws of causality.

(Thermodynamics is in some sense irreversible, e.g. an egg would not unscramble, except that thermodynamics is a coarse-grained picture, and that an egg would unscramble is in fact not impossible, just extremely unlikely.)

  • "The laws of physics" in general are clearly not insensitive to the direction of time, as playing back, for example, the breaking of an egg clearly demonstrates: Some macroscopic events -- indeed, if we look closely, all of them -- are irreversible. This is an emergent property of collectives: We have an element of (true) random at the microscopic level which results in a non-deterministic future, which tends to evolve into more likely states. Reversing time would require evolving into unlikely states, which is increasingly unlikely. Commented Apr 26 at 9:01
  • On the microscopic level, time is reversible, but only together with parity and charge (CPT symmetry). Commented Apr 26 at 9:03
  • I was going in the opposite direction, i.e. one of generality: and Unitarity in QM is reversibility, in state space, which is more fundamental and not even at the same ontological level as a standard model. But even more generally, it's a principle of conservation of information that is becoming ubiquitous in physics, which again is a name for reversibility. Thermodynamics, OTOH, is a coarse-grained picture and it is simply a different issue. Commented Apr 26 at 20:20
  • Indeed, to the point, thermodynamics is a coarse-grained picture with loss of information, whence the irreversibility. Commented Apr 27 at 7:37

To deal with you question about bidirectional time one first needs a scientific context.

  • A suitable context are world models on the base of General Relativity. Here your question becomes the question about causality in a 4-dimensional spacetime with closed timelike curves. Moving on a closed timelike curve: Locally it is not unusual, but globally one registers the same events which already happened before.

    A 2-dimensional analog would be turning around on a circle on the earth and finally arriving at one’s starting point. A main point of discussing such models is the question of causality.

  • A model with closed timelike curves on the basis of General Relativity has been constructed by Kurt Goedel. For an introduction see Closed timelike curves with a link to further references, notably to Goedel’s paper from 1949.

  • From a philosophical point of view there is the concept of eternal recurrence. Friedrich Nietzsche propagated this teaching and elaborated some of its consequences in his books “The Gay Science” and “Thus spoke Zarathustra”, see Eternal Return.


According to physics,

  • there is NO present (see the Einstein's Train Thought Experiment, which shows that what is simultaneous to one observer is sequential to another);
  • there are multiple possible futures (which is a natural fact from our perspective, due to the nature of time we experience),

and, believe it or not,

  • there are multiple pasts, you happen only to know and remember one (such is an evident consequence of the Many Worlds Interpretation of Quantum Mechanics) (see Sean Carroll on multiple pasts).

In strict terms, physically...

  • Time is an arbitrary partial knowledge of a multiplicity of QM wave function collapsing facts that happen outside of our minds.

So, in essence, there is no time as a physical fact. Time is metaphysical (purely rational). And in consequence, time "flows" only in our minds. Time is essentially subjective memory.

Causality is, therefore, also a subjective synthetic a priori truth (see Kant's Transcendental Idealism). Causality is not a physical fact. If time is subjective, so is causality.

So, regarding causality, if time would flow in two directions, two options are possible:

  1. In backwards time, naming would be the same, causes would also follow consequences, but our knowledge of causal facts would be the opposite (fire causes a match to be scratched).

  2. Naming would be opposite, and facts would be the same: consequence, then, cause. So, after consequences, causes arise. In backwards time, a match on fire provokes a cause: the scratching of a match.

Notice that time being the knowledge of an ordered sequence of facts, you can reverse it by thinking facts backwards (which is just like seeing a movie forwards and backwards). For example, I saw a match on fire, then, I lighted the match and then, I picked up the matchbox. Learning just requires practice.

Making time flow backwards can be a voluntary act. Furthermore, we do it frequently. For example, when we lose the keys: I am here, I came from the entry, after playing with the dog... Eureka! My keys are in the dog's house. OR, from this consequent fact, I can find the causal fact. Such reverse backwards time flow mechanics sustain a branch of knowledge that comes as an opposite to causality: teleology. Teleology is a term that aligns with the opposite perspective of "causality¨.

In fact, if time wouldn't be rationally reversible, it wouldn't be possible. From the point of view of memory, irreversible time would be like not having any memory at all, we would just experience a sequence of facts without learning them (not even like a fish, which can remember only 5 seconds). From the point of view of causality, there would be no causes, just facts, so, we wouldn't be able to relate recent facts with previous facts, etc.

So, as a fact, time does flow in both directions.

  • 1
    What an interesting answer! It seems totally wrong from my perspective, but very original. Commented Apr 25 at 14:54
  • Causality is not only in our minds. In fact, because you mention relativity theory: Exactly the unintuitive consequences of it (e.g. that different observers disagree about the order of events) are necessary to preserve causality (light cone and all). Causality is a fundamental feature of our universe. Commented Apr 26 at 9:10

Causality is a human concept directly "linked" with time. When a system progresses from state A to state B it is supposed that some kind of force(?) acts upon this system to cause this progression: causality. This cannot be undone, a new force must act to reverse the process; so, although the effect can be undone, time still flows forward, because time is - this context - inside which things happen : by definition, time is this progression.


One of Einstein's great insights was simultaneity (if there is such a word) is not universally well-defined - it depends in surprising ways on how you move relative to the events you are observing.

Time itself is (assumed to be) universal in the sense that there is time everywhere in space, of course, since it is a dimension of space. Whether causality 'causes' time or is at all connected with the direction of time is not clear, but there is no doubt in my mind that the sequence of events is the same everywhere: if a supernove goes off 13bn lightyears away and becomes a black hole, then that sequence won't be switched around if we were observing it from somewhere else.

So, what about time being unidirectional? Well, in Einstein's theory, since the perception of time is local, you might still be moving forward in your personal time, but somehow coming back to what is that past to some other observers. When Einstein in his later life worked with Theodor Kaluza and Oskar Klein (Kaluza-Klein Theory), they were considering two-dimensional time, where this kind of thing might be more easily imagined.

You must log in to answer this question.

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