If we take a Freeman Dyson view of the future of the universe, of mankind, or maybe robotic descendants, or some engineered descendant of human beings, spreading out through the solar system and eventually through the galaxy, harnessing natural energy on galactic dimensions, we'd be talking hundreds of millions of years of development here. At that stage our descendants might be capable of manipulating entire stars or black holes, and creating something like a wormhole, but it's not the sort of thing that's going to be done in a hundred years or even a thousand years — unless there's another way of doing it. This is of course always the excitement in a scientific topic: have we overlooked something? And given that we know time is elastic, that time can be manipulated, some way of traveling into the past seems to be possible. So is there a much easier method that we've overlooked? The great hope for building a time machine in the foreseeable future is that that is the case, that something involving maybe weird aspects of quantum physics is going to do it for us, some other type of physical process that we haven't yet discovered — but it's going to have to have gravitation in there somewhere.

EDGE: Maybe it's just that little red pill.

DAVIES: Sorry, but no. Here is where H. G. Wells got it wrong. His time traveler sat in this machine and then pressed a few buttons or something and effectively threw the great cosmic movie into reverse. Everything ran backwards. Then when he got to where he wanted to go he hit the stop button, just like the fast rewind on a video player. But the time travel that I'm talking about is not like that. It's not a method of somehow reversing the arrow of time. It is going off on a journey through space, in a closed loop, and arriving back at your starting point before you leave. There is no reversal of the arrow of time, no putting the great cosmic movie into reverse. Everything around you continues in a forward direction, so in your local neighborhood the arrow of time is unchanged. Eggs still break and don't reassemble themselves. It's not that you're going backwards in time, it's that you visit the past. There's a distinction between going backwards in time, in the sense of reversing through time, and going to the past, which is what I'm talking about.

EDGE: How does all this fit in with the views expressed by Julian Barbour in his book The End of Time?

DAVIES: Barbour argues that time doesn't really exist, to express his work somewhat simplistically. Clearly time exists at the practical level — at the level of gravitation and engineering and everyday Newtonian mechanics. To say there's no time is rather like saying there's no matter, on the basis that ultimately matter is made up of vibrating superstrings or something, You might be tempted to say about matter, well, it's not really there at all. The truth is, matter manifests itself in our everyday quasi-classical quasi-microscopic world, and space and time manifest themselves in that world too. I concede that space and time may not be the ultimate reality. It could well be that space and time — and we really have to link them together — are ultimately derived concepts or derived properties of the world. It could be that ultimate reality is something more abstract, some sort of pre-space-time, component out of which space-time is built. Just like matter, time may be a secondary or derived concept. But nevertheless, at a sufficiently large level of size, there is the familiar space-time we know. You can't wish it away, or define it away through mathematics — it's something that you can try to explain. Wood, for instance, is not a primary substance, it's made up of something else, which in turn is made up of something else, and so on. But that doesn't mean that wood is unreal. It's still there. The same goes for time. We know that time is real at one level because it can be manipulated ­ stretched and shrunk by the processes I have been discussing.

Your question is very pertinent though, because before the theory of relativity, it was fashionable in some quarters, and maybe it still is, to try to make out that time is somehow merely a human construct, deriving from our sense of the flux or flow of events, that it's something to do with the way we perceive the world as a temporal sequence. I'm not denying that we perceive time as flux, but time is not solely a human invention or a human category. For the physicist, time and space, along with matter, form part of the equipment that the universe comes with. Or rather, it's what the universe is made of. To say that it doesn't exist at all is nonsensical.