Conroy continued, “Wheeler was a great advocate of the idea of a participatory universe: a universe shaped by the inhabitants who observe and explain it. He had a favorite metaphor for this concept … do you know the old game of twenty questions? One person thinks of an object, and the other keeps asking yes-or-no questions, to try to find out what it is.

“There’s another way to play the game, though. You don’t choose any object at all, to start with. You just answer the questions ‘yes’ or ‘no’, more or less at random – but constrained by the need to be consistent with what you’ve already said. If you’ve said that ‘it’ is blue all over, you can’t change your mind later and say that it’s red … even though you still have no precise idea what ‘it’ really is. But as more and more questions are asked, what ‘it’ might be becomes narrower and narrower.

“Wheeler suggested that the universe itself behaved like that undefined object – only coming into being as something specific through a similar process of interrogation. We make observations, we carry out experiments – we ask questions about ‘it.’ We get back answers – some of them more or less random – but they’re never absolute contradictions. And the more questions we ask … the more precisely the universe takes shape.”

I said, “You mean like … making measurements on microscopic objects? Some properties of subatomic particles don’t exist until they’re measured – and the measurement you get has a random component – but if you measure the same thing a second time, you get the same result.” This was old, old ground, well-established and uncontroversial. “Surely that’s the kind of thing Wheeler would have meant?”

Conroy agreed. “That’s the definitive example. Which dates back to Niels Bohr, of course – and Wheeler was one of his students, in Copenhagen in the nineteen-thirties. Quantum measurement was certainly the inspiration for the whole model. Wheeler and his successors took it further, though.

“Quantum measurement is about individual, microscopic events which do or don’t happen – at random, but according to probabilities determined by a set of pre-existing laws. About … individual heads and tails, not the shape of the coin, or the overall odds when it’s thrown repeatedly. It’s easy enough to see that a coin is neither ‘heads’ nor ‘tails’ while it’s still up in the air, spinning – but what if it’s not even any particular coin? What if there really are no pre-existing laws governing the system you’re about to measure … any more than there are pre-existing answers to any of those measurements?”