Is gravity special in your view?

Yes. Physicists define all the other forces in terms of fields evolving in space-time. Gravity alone tells us about the geometry and curvature of space-time itself. None of the other forces describe the universal background geometry that we live in like gravity does.

At the moment, our best theory of quantum mechanics uses this background structure of space-time — which gravity defines. And if you really believe that gravity is quantized, then we lose that background structure.

What sorts of problems do you run into if gravity is classical and not quantized?

For a long time, the community believed it was logically impossible for gravity to be classical because coupling a quantum system with a classical system would lead to inconsistencies. In the 1950s, Richard Feynman imagined a situation that illuminated the problem: He began with a massive particle that is in a superposition of two different locations. These locations could be two holes in a metal sheet, as in the famous double-slit experiment. Here, the particle also behaves like a wave. It creates an interference pattern of light and dark stripes on the other side of the slits, which makes it impossible to know which slit it went through. In popular accounts, the particle is sometimes described as going through both slits at once.

But since the particle has mass, it creates a gravitational field that we can measure. And that gravitational field tells us its location. If the gravitational field is classical, we can measure it to infinite precision, infer the particle’s location, and determine which slit it went through. So we then have a paradoxical situation — the interference pattern tells us that we can’t determine which slit the particle went through, but the classical gravitational field lets us do just that.

But if the gravitational field is quantum, there is no paradox — uncertainty creeps in when measuring the gravitational field, and so we still have uncertainty in determining the particle’s location.

So if gravity behaves classically, you end up knowing too much. And that means that cherished ideas from quantum mechanics, like superposition, break down?

Yes, the gravitational field knows too much. But there’s a loophole in Feynman’s argument that could allow classical gravity to work.

What is that loophole?

As it stands, we only know which path the particle took because it produces a definite gravitational field that bends space-time and allows us to determine the particle’s location. 

But if that interaction between the particle and space-time is random — or unpredictable — then the particle itself doesn’t completely dictate the gravitational field. Which means that measuring the gravitational field will not always determine which slit the particle went through because the gravitational field could be in one of many states. Randomness creeps in, and you no longer have a paradox.

So why don’t more physicists think gravity is classical?

Well, it is logically possible to have a theory in which we don’t quantize all the fields. But for a classical theory of gravity to be consistent with everything else being quantized, then gravity has to be fundamentally random. To a lot of physicists that’s unacceptable.