If I were to tell you that I had constructed a time machine that could travel forward ten thousand years, you might very reasonably ask me for proof. But that’s easy! You and I step inside, I flick the switch, and when we step out again we see—the towering filigree architecture of the far future; taller slimmer humans with much larger heads; a silver sheen to the ordinary daylight and so on.
But what if I tell you that my machine works in such a way as to take not only you and I ten thousand years forward, but the whole world? We step out into 2007 architecture, ordinary-looking people, and petrol driven cars. But the stars! We need only look up into the night sky and see how far the proper motion of the stars has, for instance, sprawled open Ursa Major.
But what if I tell you that my machine works in such a way as to take not only the whole world forward in time, but the whole cosmos? Then there’s no way to determine for sure whether my machine works or not.
About two-thirds of the questions philosophers argue over are examples of this latter category. (I mean this is the sense that it may be we have free will, and it may be that we don’t, but whichever way philosophers call it, it will feel the same to us on the ground.) Indeed, one of the ways of approaching a definition would be to say that the first of these time machines is technology, the second science and the third philosophy.