Have you ever made a mistake that you wish you could undo? Correcting past mistakes is one of the reasons we find the concept of time travel so fascinating. As often portrayed in science fiction, with a time machine, nothing is permanent anymore – you can always go back and change it. But is time travel really possible in our universe, or is it just science fiction?

Our modern understanding of time and causality
comes from general
relativity. Theoretical physicist Albert Einstein's theory combines space
and time into a single entity – "spacetime" – and provides a
remarkably intricate explanation of how they both work, at a level unmatched by
any other established theory.

This theory has existed for more than 100
years, and has been experimentally verified to extremely high precision, so
physicists are fairly certain it provides an accurate description of the causal
structure of our Universe.

For decades, physicists have been trying to use general relativity to figure out if time travel is possible. It turns out that you can write down equations that describe time travel and are fully compatible and consistent with relativity. But physics is not mathematics, and equations are meaningless if they do not correspond to anything in reality.

**Arguments against time travel**

There are two main issues which make us think
these equations may be unrealistic. The first issue is a practical one:
building a time machine seems to require exotic matter, which is
matter with negative energy.

All the matter we see in our daily lives has positive
energy – matter with negative energy is not something you can just find lying
around. From quantum mechanics, we know that such matter can theoretically be
created, but in too
small quantities and for too short times.

However, there is no proof that it is
impossible to create exotic matter in sufficient quantities. Furthermore, other
equations may be discovered that allow time travel without requiring exotic
matter. Therefore, this issue may just be a limitation of our current
technology or understanding of quantum mechanics.

The other main issue is less practical, but
more significant: it is the observation that time travel seems to contradict
logic, in the form of time
travel paradoxes.

There are several types of such paradoxes, but
the most problematic are consistency paradoxes.

A popular trope in science fiction, consistency
paradoxes happen whenever there is a certain event that leads to changing the
past, but the change itself prevents this event from happening in the first
place.

For example, consider a scenario where I enter
my time machine, use it to go back in time five minutes, and destroy the
machine as soon as I get to the past. Now that I destroyed the time machine, it
would be impossible for me to use it five minutes later.

But if I cannot use the time machine, then I cannot go back in time and destroy it. Therefore, it is not destroyed, so I can go back in time and destroy it. In other words, the time machine is destroyed if and only if it is not destroyed. Since it cannot be both destroyed and not destroyed simultaneously, this scenario is inconsistent and paradoxical.

**Eliminating the paradoxes**

There's a common misconception in science
fiction that paradoxes can be "created." Time travellers are usually
warned not to make significant changes to the past and to avoid meeting their
past selves for this exact reason. Examples of this may be found in many time
travel movies, such as the *Back to the Future* trilogy.

But in physics, a paradox is not an event that
can actually happen – it is a purely theoretical concept that points towards an
inconsistency in the theory itself. In other words, consistency paradoxes don't
merely imply time travel is a dangerous endeavor, they imply it simply cannot
be possible. This was one of the motivations for theoretical physicist Stephen Hawking to formulate his chronology protection
conjecture, which states that time travel should be impossible. However,
this conjecture so far remains unproven.

Furthermore, the universe would be a much more
interesting place if instead of eliminating time travel due to paradoxes, we
could just eliminate the paradoxes themselves. One attempt at resolving time
travel paradoxes is theoretical physicist Igor Dmitriyevich Novikov's self-consistency conjecture,
which essentially states that you can travel to the past, but you cannot change
it.

According to Novikov, if I tried to destroy my time machine five minutes in the past, I would find that it is impossible to do so. The laws of physics would somehow conspire to preserve consistency.

**Introducing multiple histories**

But what's the point of going back in time if
you cannot change the past? My recent work, together with my students Jacob
Hauser and Jared Wogan, shows that there are time travel paradoxes that
Novikov's conjecture cannot resolve. This takes us back to square one, since if
even just one paradox cannot be eliminated, time travel remains logically
impossible.

So, is this the final nail in the coffin of
time travel? Not quite. We showed that allowing for multiple histories (or in more
familiar terms, parallel timelines) can resolve the paradoxes that Novikov's
conjecture cannot. In fact, it can resolve any paradox you throw at it.

The idea is very simple. When I exit the time
machine, I exit into a different timeline. In that timeline, I can do whatever
I want, including destroying the time machine, without changing anything in the
original timeline I came from.

Since I cannot destroy the time machine in the
original timeline, which is the one I actually used to travel back in time,
there is no paradox.

After working on time
travel paradoxes for the last three years, I have become increasingly
convinced that time travel could be possible, but only if our Universe can
allow multiple histories to coexist. So, can it?

Quantum mechanics certainly seems to imply so,
at least if you subscribe to Everett's "many-worlds"
interpretation, where one history can "split" into multiple
histories, one for each possible measurement outcome – for example,
whether Schrödinger's cat is
alive or dead, or whether or not I arrived in the past.

But these are just speculations. My students
and I are currently working on finding a concrete theory of time travel with
multiple histories that is fully compatible with general relativity. Of course, even if we manage
to find such a theory, this would not be sufficient to prove that time travel
is possible, but it would at least mean that time travel is not ruled out by
consistency paradoxes.

Time travel and parallel timelines almost
always go hand-in-hand in science fiction, but now we have proof that they must
go hand-in-hand in real science as well. General relativity and quantum
mechanics tell us that time travel might be possible, but if it is, then
multiple histories must also be possible.