Maybe some readers recall Immanuel Velikovsky's 1950 mega bestseller Worlds in Collision. The book, which caused a real sensation at the time, was an attempt to "explain" many of the big cataclysms and "miracles" recorded in mythic and folkloric narratives of ancient cultures as real astrophysical events. Velikovsky's thesis was that narratives of floods and mass destructions were not just allegorical or metaphorical but records of events that did take place. Mythic and biblical catastrophism had a historical and a scientific value to them.
In Velikovsky's theory, Venus was ejected from Jupiter as a kind of comet sometime around the 15th century BCE. Its periodic passage by Earth caused all sorts of havoc. His mythic inspiration came from Greek mythology, in particular the fable where Athena (Venus) was ejected from the head of Zeus (Jupiter). In spite of its popular appeal, the astronomical community summarily dismissed Velikovsky's ideas, a key player having been none other than Carl Sagan, who knew a lot about Venus and its atmosphere. In any case, Velikovsky's celestial catastrophism follows on the footsteps of many claims of apocalyptic endings due to upheavals in the skies. And even if his thesis was unfounded, bad things can and do happen from time to time due to collisions between "worlds." Think, for example, of the demise of the dinosaurs 65 millions years ago due a collision with a seven-mile wide asteroid.
Still, Velikovsky's doomsday imaginings are a child's play compared to some catastrophic ideas that modern cosmologists have been putting forward. I don't mean the devastation caused by a collision with an asteroid or comet, but of whole universes colliding with one another, including with our own.
Welcome to cosmic catastrophism.
The universe began its existence 13.7 billion years ago and has been expanding ever since. However, current observations indicate that this expansion wasn't always at the same rate. Right at the beginning of time, the cosmos underwent a short period of hyper-accelerated expansion called inflation. According to this theory, proposed by MIT cosmologist Alan Guth in 1981, our whole universe could have emerged from a tiny patch of space that was stretched like a rubber band by the enormous factor of one hundred trillion trillion times (1026) in a fraction of a second. The universe we observe today fits within this stretched region, like an island in an ocean.
Now imagine that other portions of space, neighbors to that tiny patch that gave rise to our universe, also got stretched at different rates and at different times. We would have a universe filled with island-universes, each with its own history and possibly even types of matter, etc. This ocean of island-universes is called the multiverse.
Since physics is an empirical science, any hypothesis needs to be tested before being accepted by the community. This is as true for a ball rolling down a hill as for Guth's inflating universe or the multiverse. For the ball, we know how to apply Newton's laws of motion to describe its rolling and the results come out in excellent agreement with observations. Cosmic inflation predicts that our universe is geometrically flat (or almost) like the surface of a table but in three dimensions; it also predicts that space should be filled with radiation with a uniform temperature, as bathwater fills a bathtub. These two predictions have been confirmed, although a skeptic could argue that inflation was designed to accommodate these two observational facts about the universe. To its merit, inflation also offers an explanation as to how galaxies were first born and then grouped together in clusters, something that no other theory can do satisfactorily. Cosmologists like inflation a lot for its simplicity and range of explanation.
Since we can't receive information from outside our universe (or better, from outside our "horizon", the sphere that delimits how far light travelled in 13.7 billion years), how can we possibly test the existence of other universes "out there"? This has been a sticky point with the multiverse and indeed, the notion that the multiverse extends perhaps to spatial infinity is untestable. Infinity makes sense mathematically and may even be realized in Nature; but we will never know for sure.
However, we can do the next best thing, and see if at least neighboring universes exist. Just as with soap bubbles that vibrate when they collide with one another without popping, if another universe collided with ours in the distant past, the radiation inside our universe would have vibrated in response to the perturbations caused by the collision. These perturbations would be registered in the cosmic radiation and could, in principle, be observed. Matthew Kleban from New York University and his collaborators, and Anthony Aguirre from the University of California at Santa Cruz and his have been studying what kinds of signals would be left over from these dramatic events. Kleban found a unique signature, concentric rings where the radiation temperature would show a characteristic fluctuation. On top of the rings the radiation itself would be polarized, that is, it would oscillate in tandem in a specific direction of the sky. At least for now, no telltale rings have been found in the cosmic radiation, although the European satellite Planck promises to deliver more accurate polarization data that may shed light on the issue.
The bad news is that the probability of a collision with another universe increases with time: we could disappear at any instant: live life to the fullest!
The good news is that, although the multiverse as a whole may not be a testable scientific hypothesis, with some luck we may at least know if one or a few other universes exist. An observational test distinguishes science from idle speculation.