One of the great appeals to me of cosmology isn’t so much what we learn about the universe, as what we learn about ourselves. Take the shape of the universe, for example. Cosmologists have long wondered about it. What is the overall structure of the space-time continuum with its myriad galaxies, including ours?
The actual answer – for the moment – is ‘nobody knows’, but a fair number of ideas have been proposed, all of them owing as much to imagination as to empirical observation. Eighteen of them, to be precise. The top three are the simplest: the universe is a sphere, it’s a saddle, or it’s flat. Others are more exotic, such as a kind of Moebius strip, or a one-sided Klein bottle. To this can be added a further complicating factor: any of these shapes, although fixed of themselves, might be geometrically bent – just like a flat piece of paper can be bent into a cylinder or a parachute shape.
Current thinking leans away from those more exotic shapes. If the universe was shaped like a Moebius strip, for instance, we’d expect to see some slightly odd things in the distance. Specifically, if you looked at a distant galaxy, then turned your telescope around and looked for it in the other direction, you’d also see it, but it would be flipped upside down because a Moebius strip has a 180 degree twist in it. So far, nobody’s seen anything like that.
That suggests that the simpler shapes are probably the ones to explore further – and at the moment, thinking leans towards flat. This has been deduced by analysing light from the cosmic background radiation – the oldest known light in the universe, which has travelled across the width of the visible universe to reach us. If the beams from that light are exactly parallel, then the universe is flat. If they diverge, it’s likely saddle-shaped; and if they converge, it’s probably a sphere.
As you can imagine, the sensitivity needed to make that kind of measurement is very delicate, and it’s open to revision. But that is where thinking is going at the moment.
The thing is that a flat universe, by definition, could well be infinite. But it might not be. It could be a flat surface bent into a geometric shape, like a cylinder. That would make the universe the same shape as the world in the original Civilisation game, which was basically cylindrical – you couldn’t go off the edges to north and south, but if you went east or west far enough you’d end up back at your starting point. Again, though, we haven’t seen any evidence of this in the real universe.
If the flat universe also happened to be bent into a cylinder, then an astronomer with a powerful enough telescope could theoretically see themselves from behind. In practise, we might pick up our own galaxy or the characteristic pattern of the Local Group in which our galaxy sits. Nothing like that has happened, although the picture is complicated by the fact that the universe is expanding, and the rate of expansion limits the distance to which we can see. Beyond that distance, the rate of expansion exceeds light-speed and so it’s invisible to us.
Incidentally, even bent into a cylinder, a flat universe would still be technically flat because parallel lines drawn on its surface remain parallel. Whereas on a sphere, which doesn’t have a flat surface, they are not. It’s the difference between Euclidean and non-Euclidean geometry.
If all of this sounds like some kind of really weirdo mind-game with no practical application, you’re probably right. But it’s where cosmology is at the moment, and the thinking around it has as much to do with our ability to imagine and conceive the abstract as it does with the empirical evidence. One thing is certain: out on the edge of what we know, the universe gets very weird indeed.
If you want to know more about our weird universe, check out my short book on that very subject – available on Kindle, right now.