Science of parallel universes
Everyone loves the idea of parallel universes - maybe it's the appeal of an ideal world where you have second chances and things turn out differently - an alternate reality where you do get into Hogwarts and the Star Wars prequels aren't made and you finally plug in your asymmetric computer cord correctly on the first try... but is there really a place in science for such wistful speculation? I mean, if "the universe" is everything that there is, you can't have two versions of it, right? Otherwise the pair would really be everything and what you started off calling the universe, wasn't. The problem here is terminology: physicists speaking informally often say "universe" when they really mean "observable universe" - that is, the part of the whole universe that we've so far been able to see. And it's perfectly fine to talk about multiple different observable universes - for example, an alien near the edge of OUR observable universe will see parts of the Whole Universe that we cannot yet see, but that's a well-understood question and not what physicists normally talk about when they discuss multiple observable universes, or "multi-verses." So let's cut to the chase: in physics, the word "Multiverse" normally refers to one of three distinct and largely unrelated proposed physical models for the universe - none of which has been tested or confirmed by experiment, by the way. The three "multiverse" models are: Type 1) Bubble universes or baby black hole universes. This is the most straightforward kind of multiverse: the basic idea is that perhaps there are other parts of the universe which are so far away that we will never see them (or are inside black holes so similarly we will never see them). This kind of model was created as an attempt to explain why our universe is so good at making stars and galaxies and black holes and life - as the argument goes, if each of these separate mutually un-seeable "bubbles" in the universe had slightly different laws of physics, then by definition we could only exist in one that had the right physical laws to allow us to exist. If you're not convinced by this logic, don't worry too much: there's not yet any experimental evidence for this kind of multiverse. Multiverse type 2) Membranes and extra dimensions. Inspired in part by the inability of the mathematics of string theory to predict the right number of dimensions for the universe in which we live, string theorists proposed the idea that perhaps what we think of as our universe is actually just a three-dimensional surface embedded within a larger super-universe with 9 spatial dimensions. Kind of like how each page of a newspaper is its own two-dimensional surface embedded within our three-dimensional world. And of course, if space had 9 dimensions rather than three, there'd be plenty of space for other three-dimensional surfaces that appeared, like ours, to be universes in their own right, but, like the pages of a newspaper, were actually part of a bigger whole. These kinds of surfaces are called "membranes" or "branes" for short. And as a reminder, there is not yet any experimental evidence for this kind of multiverse. Multiverse type 3) The many-worlds picture of quantum mechanics. Surprisingly, physicists still don't fully understand how the collapse of the wavefunction in quantum mechanics happens, and the many-worlds hypothesis makes an attempt at explanation by proposing that every possible alternate timeline for the universe is real and they all happen in an ever-larger, ever-branching way. Like, a universal choose-your-own-adventure where every possible story happens! If this were the case, we might not realize it because we'd be stuck living out just one of the infinitely many possible lives available to us. In some ways, many-worlds is similar to the bubble multiverse model by proposing "maybe anything that can happen, does. And we just happen to exist in the series of happenings that were necessary for us to exist." If you're still not convinced by this logic, don't worry: there is not yet any experimental evidence for this kind of multiverse. Of course if you want to get imaginative, you could also combine several of these models together into a multi-multiverse... a new super-speculative model based, itself, on speculative and experimentally unconfirmed models. But that's not to say we couldn't test these multiverse hypotheses. For example, if our observable universe were really just one of many disconnected bubbles or membranes and if it happened to collide with another bubble or membrane some time in the past, then that collision would certainly have had some sort of effect on what we see when we look up at the night sky. On the other hand, the many-worlds interpretation might be tested fairly soon since experimentalists are becoming increasingly able to manipulate and control ever-larger quantum mechanical systems in their labs - systems that approach the line between the quantum realm and our everyday experience. So as always, we must remember that physics is science, not philosophy; and in our attempts to explain the universe that we observe, we have to make claims that can in principle be tested - and then test them!