Intuitive physics

[snousle]

Here's a thought experiment that touches on a lot of different intuitions and assumptions. I'm curious to see how people answer:

You have two hockey-puck shaped permanent magnets, and you place one on each side of a thin sheet of plastic such that their mutual attraction clamps them strongly together. All surfaces are smooth and nearly frictionless (oil them if you like). If you hold the sheet stationary while rotating one of the magnets, does the other one rotate as well? Why or why not?

Comments

[come-to-think.livejournal.com]

I agree with the others on this thread that the problem is incompletely specified. However, I believe that it is usual with hockey-puck-shaped magnets for the N-S axis to be perpendicular to the flat faces; at any rate, that is true of the pair that I just plucked off my fridge (whether they stick together or repel each other depends only on which face goes with which, not on the angular position). In that case, the assembly is cylindrically symmetric, its potential energy will not be changed by a mutual rotation, and thus there can be no torque on the second magnet. (I have just tried it, using a piece of paper, and that is indeed the result.)

That's a magnetostatic argument, tho. One might ask whether, if the magnets are conducting and the first one is rotated rapidly, there might be induced currents that would spoil the result. I believe that, once again, the cylindrical symmetry prevents that from happening; but I would be prepared to be surprised, because there are subtleties in applying such arguments to pseudovector fields. (If in addition there were an electric field in the plane of the paper, I would be fully prepared for something weird to happen; but that would break the rotational symmetry too.)