I can already hear the howls of complaint: "You're writing about a new paper just because it has Stephen Hawking's name on it?" Well, yes, I guess I am, but I think it's worth saying something about the paper because not too many people actually hear much about the science that Hawking does. The paper is technical and I can only give the merest flavor of what it talks about. Much of this will be heavy going but I hope that you can get a taste of the kind of work that Hawking is now up to.
His new paper, with other physics heavyweights Jim Hartle and Thomas Hertog, was published the other day in Physical Review Letters. In it, he continues a research direction he started with Hartle back in 1983 in a paper provocatively (for physicists) titled "Wave function of the Universe". In that paper, H&H suggested that the universe might be finite but have no boundary--the no-boundary proposal.
This might seem like a mind-bending topic at first because surely the only way for something to have no edges is for it to go on forever. However, once you allow the fun of general relativity, in which spacetime can bend in all kinds of interesting ways, you can start to imagine a universe that doesn't go on forever but has no edges because the universe wraps around from one side to the other. Or, the dimensions of space and time can mix together and turn into each other in odd ways to avoid there ever being an edge. One side effect of all this mixing up of space and time is that H&H also needed to introduce the concept of imaginary time. I'm not going to go into the details here but if you want to know more, you can try on for size the text of one of Hawking's public lectures, "The Beginning of Time".
Back to the present paper: Hartle, Hawking, and Hertog, or H3, as I shall geekily call them for now, apply the no-boundary proposal to a class of possible universes in the context of the string theory landscape--the vast number of possible universes that could exist within string theory.
H3 point out that the string theory landscape by itself gives no way to explain why the universe is the way it is out of all the possible options. For that, say H3, "one has to turn to cosmology and to a theory of the quantum state of the Universe."
The authors go on to apply their no-boundary proposal to the evolution of the universe, building in the result that the universe must evolve to what we actually see today--essentially classical on most scales. In other words, the universe, despite being quantum at its foundations has evolved to a point where the obvious effects of quantum physics hide in the small and the unusual. That alone limits how the universe could have evolved quite significantly within this framework.
With this constraint, H3 predict what type of universe is most likely (and you have to work in probabilities when you are dealing with the string landscape). They find that the most likely universe is one with an inflationary past and one that has a beginning that looks semiclassical (not the singularity that many theories of the beginning of time predict).
The authors go on to predict many properties of what such a universe would look like. One is that the beginning of the universe is characterized by a bounce rather than starting from nothing. However, the arrow of time points outward from the bounce on both sides so there is very little chance that anything from one side of the bounce could influence the other, in contrast to some other versions of bounces.
Also, such a universe should have had lots of inflation bringing it to a nearly flat state such as we observe now. That just means that our universe now doesn't have much weird curviness and that each of the three space dimensions we're used to in everyday life really are something like the straight directions we naturally imagine.
There are some other even more complicated ideas the authors discuss but I'll leave those for professional physicists to ponder. One of the main conclusions of the paper is that the no-boundary proposal is able to provide some kinds of limits on the immense number of possibilities allowed by the string landscape without needing any extra outside rules to be imposed. It's heavy going but if you've stayed with me this far, perhaps you have a small idea of just what sorts of things keep the heavy hitters of cosmology occupied these days.