In fact there are almost too many other potential universes. Physicists have proposed several candidate forms of "multiverse", each made possible by a different aspect of the laws of physics.
The trouble is, virtually by definition we probably cannot ever visit these other universes to confirm that they exist. So the question is, can we devise other ways to test for the existence of entire universes that we cannot see or touch?
Worlds within worlds
In at least some of these alternative universes, it has been suggested, we have doppelgängers living lives much like – perhaps almost identical to – our own.
Giordano Bruno speculated that the Universe might be infinite
That idea tickles our ego and awakens our fantasies, which is doubtless why the multiverse theories, however far-out they seem, enjoy so much popularity. We have embraced alternative universes in works of fiction ranging from Philip K. Dick's The Man in the High Castle to movies like Sliding Doors.
Indeed, there is nothing new about the idea of a multiverse, as philosopher of religion Mary-Jane Rubenstein explains in her 2014 book Worlds Without End.
In the mid-16th century, Copernicus argued that the Earth is not the centre of the Universe. Several decades later, Galileo's telescope showed him stars beyond measure: a glimpse of the vastness of the cosmos.
So at the end of the 16th century, the Italian philosopher Giordano Bruno speculated that the Universe might be infinite, populated by an infinite number of inhabited worlds.
The idea of a Universe containing many solar systems became commonplace in the 18th Century.
By the early 20th Century, the Irish physicist Edmund Fournier d'Albe was even suggesting that there might be an infinite regression of "nested" universes at different scales, ever larger and ever smaller. In this view, an individual atom might be like a real, inhabited solar system.
Scientists today reject that notion of a "Russian doll" multiverse, but they have postulated several other ways in which multiverses might exist. Here are five of them, along with a rough guide to how likely they are.
The patchwork universe
The simplest multiverse is a consequence of the infinite size of our own Universe.
There must be worlds identical to Earth somewhere out there
We do not actually know if the Universe is infinite, but we cannot rule it out. If it is, then it must be divided into a patchwork of regions that cannot see one another.
This is simply because the regions are too far apart for light to have crossed the distance. Our Universe is only 13.8 billion years old, so any regions further than 13.8 billion light years apart are utterly cut off.
To all intents and purposes, these regions are separate universes. But they will not stay that way: eventually light will cross the divide and the universes will merge.
If our Universe really does contain an infinite number of "island universes" like ours, with matter and stars and planets, there must be worlds identical to Earth somewhere out there.
It may sound incredibly unlikely that atoms should come together by chance into an exact replica of Earth, or a replica that is exact except for the colour of your socks. But in a genuine infinity of worlds, even that strange place must exist. In fact, it must exist countless times.
The Universe began as an infinitesimally tiny point and then expanded incredibly fast in a super-heated fireball
If so, then somewhere almost unimaginably far off, a being identical to me is typing out these words, and wondering if his editor is going to insist on radical revisions. [nice try Phil - ed]
By the same logic, rather farther away there is an entire observable universe identical to ours. This distance can be estimated at about 10 to the power 10 to the power 118 metres.
It is possible that this is not the case at all. Maybe the Universe is not infinite. Or even if it is, maybe all the matter is concentrated in our corner of it, in which case most of the other universes could be empty. But there is no obvious reason why that should be, and no sign so far that matter gets sparser the farther away we look.
The inflationary multiverse
The second multiverse theory arises from our best ideas about how our own Universe began.
According to the predominant view of the Big Bang, the Universe began as an infinitesimally tiny point and then expanded incredibly fast in a super-heated fireball. A fraction of a second after this expansion began, it may have fleetingly accelerated at a truly enormous rate, far faster than the speed of light. This burst is called "inflation".
There are many, perhaps infinitely many, universes appearing and growing all the time
Inflationary theory explains why the Universe is relatively uniform everywhere we look. Inflation blew up the fireball to a cosmic scale before it had a chance to get too clumpy.
However, that primordial state would have been ruffled by tiny chance variations, which also got blown up by inflation. These fluctuations are now preserved in the cosmic microwave background radiation, the faint afterglow of the Big Bang. This radiation pervades the Universe, but it is not perfectly uniform.
Several satellite-based telescopes have mapped out these variations in fine detail, and compared them to those predicted by inflationary theory. The match is almost unbelievably good, suggesting that inflation really did happen.
This suggests that we can understand how the Big Bang happened – in which case we can reasonably ask if it happened more than once.
The current view is that the Big Bang happened when a patch of ordinary space, containing no matter but filled with energy, appeared within a different kind of space called the "false vacuum". It then grew like an expanding bubble.
Perhaps our Universe is simply one of a crowd
But according to this theory, the false vacuum should also experience a kind of inflation, causing it to expand at fantastic speed. Meanwhile, other bubble universes of "true vacuum" can appear within it – and not just, like our Universe, 13.8 billion years ago, but constantly.
This scenario is called "eternal inflation". It suggests there are many, perhaps infinitely many, universes appearing and growing all the time. But we can never reach them, even if we travel at the speed of light forever, because they are receding too fast for us ever to catch up.
The UK Astronomer Royal Martin Rees suggests that the inflationary multiverse theory represents a "fourth Copernican revolution": the fourth time that we have been forced to downgrade our status in the heavens. After Copernicus suggested Earth was just one planet among others, we realized that our Sun is just one star in our galaxy, and that other stars might have planets. Then we discovered that our galaxy is just one among countless more in an expanding Universe. And now perhaps our Universe is simply one of a crowd.
We do not yet know for sure if inflationary theory is true.
However, if eternal inflation does create a multiverse from an endless series of Big Bangs, it could help to resolve one of the biggest problems in modern physics.
The fundamental constants of the laws of physics seem bizarrely fine-tuned to the values needed for life to exist
Some physicists have long been searching for a "theory of everything": a set of basic laws, or perhaps just a single equation, from which all the other principles of physics can be derived. But they have found there are more alternatives to choose from than there are fundamental particles in the known universe.
Many physicists who delve into these waters believe that an idea called string theory is the best candidate for a "final theory". But the latest version offers a huge number of distinct solutions: 1 followed by 500 zeros. Each solution yields its own set of physical laws, and we have no obvious reason to prefer one over any other.
The inflationary multiverse relieves us of the need to choose at all. If parallel universes have been popping up in an inflating false vacuum for billions of years, each could have different physical laws, determined by one of these many solutions to string theory.
If that is true, it could help us explain a strange property of our own Universe.
The fundamental constants of the laws of physics seem bizarrely fine-tuned to the values needed for life to exist.
Things have to be the way we find them: if they were not, we would not be here and the question would never arise
For example, if the strength of the electromagnetic force were just a little different, atoms would not be stable. Just a 4% change would prevent all nuclear fusion in stars, the process that makes the carbon atoms our bodies are largely made of.
Similarly, there is a delicate balance between gravity, which pulls matter towards itself, and so-called dark energy, which does the opposite and makes the Universe expand ever faster. This is just what is needed to make stars possible while not collapsing the Universe on itself.
In this and several other ways, the Universe seems fine-tuned to host us. This has made some people suspect the hand of God.
Yet an inflationary multiverse, in which all conceivable physical laws operate somewhere, offers an alternative explanation.
In every universe set up in this life-friendly way, the argument goes, intelligent beings will be scratching their heads trying to understand their luck. In the far more numerous universes that are set up differently, there is no one to ask the question.
If you don't want God, you'd better have a multiverse
This is an example of the "anthropic principle", which says that things have to be the way we find them: if they were not, we would not be here and the question would never arise.
For many physicists and philosophers, this argument is a cheat: a way to evade rather than explain the fine-tuning problem.
How can we test these assertions, they ask? Surely it is defeatist to accept that there is no reason why the laws of nature are what they are, and simply say that in other universes they are different?
The trouble is, unless you have some other explanation for fine-tuning, someone will assert that God must have set things up this way. The astrophysicist Bernard Carr has put it bluntly: "If you don't want God, you'd better have a multiverse".
Cosmic natural selection
Another kind of multiverse avoids what some see as the slipperiness of this reasoning, offering a solution to the fine-tuning problem without invoking the anthropic principle.
A "mother" universe can give birth to "baby" universes
It was formulated by Lee Smolin of the Perimeter Institute for Theoretical Physics in Waterloo, Canada. In 1992 he proposed that universes might reproduce and evolve rather like living things do.
On Earth, natural selection favours the emergence of "useful" traits such as fast running or opposable thumbs. In the multiverse, Smolin argues, there might be some pressure that favours universes like ours. He calls this "cosmological natural selection".
Smolin's idea is that a "mother" universe can give birth to "baby" universes, which form inside it. The mother universe can do this if it contains black holes.
A black hole forms when a huge star collapses under the pull of its own gravity, crushing all the atoms together until they reach infinite density.
It is a neat idea, because our Universe then does not have to be the product of pure chance
In the 1960s, Stephen Hawking and Roger Penrose pointed out that this collapse is like a mini-Big Bang in reverse. This suggested to Smolin that a black hole could become a Big Bang, spawning an entire new universe within itself.
If that is so, then the new universe might have slightly different physical properties from the one that made the black hole. This is like the random genetic mutations that mean baby organisms are different from their parents.
If a baby universe has physical laws that permit the formation of atoms, stars and life, it will also inevitably contain black holes. That will mean it can have more baby universes of its own. Over time, universes like this will become more common than those without black holes, which cannot reproduce.
It is a neat idea, because our Universe then does not have to be the product of pure chance. If a fine-tuned universe arose at random, surrounded by many other universes that were not fine-tuned, cosmic natural selection would mean that fine-tuned universes subsequently became the norm.
So far, there is no evidence that this is the case
The details of the idea are a little woolly, but Smolin points out that it has one big advantage: we can test it.
For example, if Smolin is right we should expect our Universe to be especially suited to making black holes. This is a rather more demanding criterion than simply saying it should support the existence of atoms.
But so far, there is no evidence that this is the case – let alone proof that a black hole really can spawn an entirely new universe.
The brane multiverse
When Albert Einstein's theory of general relativity began to come to public attention in the 1920s, many people speculated about the "fourth dimension" that Einstein had allegedly invoked. What might be in there? A hidden universe, maybe?
Perhaps the fifth dimension was curled up into an unimaginably small distance
This was nonsense. Einstein was not proposing a new dimension. What he was saying was that time is a dimension, similar to the three dimensions of space. All four are woven into a single fabric called space-time, which matter distorts to produce gravity.
Even so, other physicists were already starting to speculate about genuinely new dimensions in space.
The first intimation of hidden dimensions began with the work of the theoretical physicist Theodor Kaluza. In a 1921 paper Kaluza showed that, by adding an extra dimension to the equations of Einstein's theory of general relativity, he could obtain an extra equation that seemed to predict the existence of light.
That looked promising. But where, then, was this extra dimension?
The Swedish physicist Oskar Klein offered an answer in 1926. Perhaps the fifth dimension was curled up into an unimaginably small distance: about a billion-trillion-trillionth of a centimetre.
In the modern version of string theory, known as M-theory, there are up to seven hidden dimensions
The idea of a dimension being curled may seem strange, but it is actually a familiar phenomenon. A garden hose is a three-dimensional object, but from far enough away it looks like a one-dimensional line, because the other two dimensions are so small. Similarly, it takes so little time to cross Klein's extra dimension that we do not notice it.
Physicists have since taken Kaluza and Klein's ideas much further in string theory. This seeks to explain fundamental particles as the vibrations of even smaller entities called strings.
When string theory was developed in the 1980s, it turned out that it could only work if there were extra dimensions. In the modern version of string theory, known as M-theory, there are up to seven hidden dimensions.
What's more, these dimensions need not be compact after all. They can be extended regions called branes (short for "membranes"), which may be multi-dimensional.
If branes collide, the results could be monumental
A brane might be a perfectly adequate hiding place for an entire universe. M-theory postulates a multiverse of branes of various dimensions, coexisting rather like a stack of papers.
If this is true, there should be a new class of particles called Kaluza-Klein particles. In theory we could make them, perhaps in a particle accelerator like the Large Hadron Collider. They would have distinctive signatures, because some of their momentum is carried in the hidden dimensions.
These brane worlds should remain quite distinct and separate from each other, because forces like gravity do not pass between them. But if branes collide, the results could be monumental. Conceivably, such a collision could have triggered our own Big Bang.
It has also been proposed that gravity, uniquely among the fundamental forces, might "leak" between branes. This leakage could explain why gravity is so weak compared to the other fundamental forces.
If their idea is true, there is an awful lot of space out there for other universes
As Lisa Randall of Harvard University puts it: "if gravity is spread out over large extra dimensions, its force would be diluted."
In 1999, Randall and her colleague Raman Sundrum suggested that the branes do not just carry gravity, they produce it by curving space. In effect this means that a brane "concentrates" gravity, so that it looks weak in a second brane nearby.
This could also explain why we could live on a brane with infinite extra dimensions without noticing them. If their idea is true, there is an awful lot of space out there for other universes.
The quantum multiverse
The theory of quantum mechanics is one of the most successful in all of science. It explains the behaviour of very small objects, such as atoms and their constituent fundamental particles. It can predict all kinds of phenomena, from the shapes of molecules to the way light and matter interact, with phenomenal accuracy.
Quantum mechanics treats particles as if they are waves, and describes them with a mathematical expression called a wave function.
When we make a measurement we only see one of those realities, but the others also exist
Perhaps the strangest feature of a wave function is that it allows a quantum particle to exist in several states at once. This is called a superposition.
But superpositions are generally destroyed as soon as we measure the object in any way. An observation "forces" the object to "choose" one particular state.
This switch from a superposition to a single state, caused by measurement, is called "collapse of the wave function". The trouble is, it is not really described by quantum mechanics, so no one knows how or why it happens.
In his 1957 doctoral thesis, the American physicist Hugh Everett suggested that we might stop fretting about the awkward nature of wave function collapse, and just do away with it.
Everett suggested that objects do not switch from multiple states to a single state when they are measured or observed. Instead, all the possibilities encoded in the wave function are equally real. When we make a measurement we only see one of those realities, but the others also exist.
This is known as the "many worlds interpretation" of quantum mechanics.
To avoid wave function collapse, you must make another universe
Everett was not very specific about where these other states actually exist. But in the 1970s, the physicist Bryce DeWitt argued that each alternative outcome must exist in a parallel reality: another world.
Suppose you conduct an experiment in which you measure the path of an electron. In this world it goes one way, but in another world it goes another way.
That requires a parallel apparatus for the electron to pass through. It also requires a parallel you to measure it. In fact you have to build an entire parallel universe around that one electron, identical in all respects except where the electron went.
In short, to avoid wave function collapse, you must make another universe.
This picture really gets extravagant when you appreciate what a measurement is. In DeWitt's view, any interaction between two quantum entities, say a photon of light bouncing off an atom, can produce alternative outcomes and therefore parallel universes.
The quantum multiverse must be in some sense real, because quantum theory demands it and quantum theory works
As DeWitt put it, "every quantum transition taking place on every star, in every galaxy, in every remote corner of the Universe is splitting our local world on earth into myriads of copies."
Not everyone sees Everett's many-worlds interpretation this way. Some say it is largely a mathematical convenience, and that we cannot say anything meaningful about the contents of those alternative universes.
But others take seriously the idea that there are countless other "yous", created every time a quantum measurement is made. The quantum multiverse must be in some sense real, they say, because quantum theory demands it and quantum theory works.
You either buy that argument or you do not. But if you accept it, you must also accept something rather unsettling.
The other kinds of parallel universes, such as those created by eternal inflation, are truly "other worlds". They exist somewhere else in space and time, or in other dimensions. They might contain exact copies of you, but those copies are separate, like a body double living on another continent.
Who are we to judge what is weird and what is not?
In contrast, the other universes of the many-worlds interpretation do not exist in other dimensions or other regions of space. Instead, they are right here, superimposed on our Universe but invisible and inaccessible. The other selves they contain really are "you".
In fact, there is no meaningful "you" at all. "You" are becoming distinct beings an absurd number of times every second: just think of all the quantum events that happen as a single electrical signal travels along a single neuron in your brain. "You" vanish into the crowd.
In other words, an idea that started out as a mathematical convenience ends up implying that there is no such thing as individuality.
Testing the multiverse
Given the strange implications of parallel universes, you might be forgiven some skepticism about whether they exist.
But who are we to judge what is weird and what is not? Scientific ideas stand or fall, not by how they "feel" to us, but by experimental testing.
And that is the problem. An alternative universe is separate from our own. By definition, it is beyond reach and out of sight. On the whole, multiverse theories cannot be tested by looking for those other worlds.
Yet even if other universes cannot be experienced directly, it might be possible to find evidence to support the reasoning behind them.
For example, we could find strong evidence for the inflationary theory of the Big Bang. That would strengthen, but not prove, the case for an inflationary multiverse.
It's proven remarkably hard to write down a theory which produces exactly the universe we see and nothing more
Some cosmologists have proposed that an inflationary multiverse might be more directly tested. A collision between our expanding bubble universe and another one should leave detectable traces in the cosmic microwave background – if we were close enough to see them.
Similarly, experiments envisaged for the Large Hadron Collider could search for evidence of the additional dimensions and particles implied by the braneworld theory.
Some argue that experimental verification is over-rated anyway. They say we can gauge the validity of a scientific idea by other means, such as whether it rests on sound logic spun from premises that do have observational support.
Finally, we might make statistical predictions.
For example, we could use the inflationary multiverse theory to predict which values of the physical constants would be expected in most universes, and then see whether these are close to the ones we see – on the basis that there is no reason to expect us to be anywhere special in the multiverse.
At any rate, it does seem odd that the multiverse keeps cropping up wherever we look. "It's proven remarkably hard to write down a theory which produces exactly the universe we see and nothing more," says physicist Max Tegmark.
Multiverse theories cannot be tested by looking for those other worlds
Even so, it is not clear that newspaper headlines will announce the discovery of another universe any time soon. Right now, these ideas lie on the border of physics and metaphysics.
In the absence of any evidence, then, here is a rough-and-ready – and frankly subjective – ranking of the probabilities of the various multiverses, the most likely first.
The patchwork multiverse is hard to avoid – if our Universe really is infinite and uniform.
The inflationary multiverse is rather likely if inflationary theory is true, and right now inflation is our best explanation for the Big Bang.
Cosmic natural selection is an ingenious idea but involves speculative physics, and there are a lot of unanswered questions.
Brane worlds are far more speculative, because they can only exist if all those extra dimensions do, and there is no direct evidence of that.
The quantum multiverse is arguably the simplest interpretation of quantum theory, but it is also vaguely defined and leads to an incoherent view of selfhood.