Chris,
I am a physicist, or, rather, used to be. If you have a link to the video I would like to see it. I believe your claim. I do not, however, think its anything beyond classical mechanics and electromagnetism though.
First, I would have to replicate the process (or at least have some corroboration beyond the single video), as I won't simply believe an internet video. Have you done this yourself?
If the facts of this are true it sounds like a classic problem that would appear on a physics graduate exam. In fact, if provided the characteristics of the system any physics graduate student worth his salt should be able to calculate the rotational speed of the object and the required stopping force.
One thing that possibly jumps out at me is the combination of the magnetic force (due to the Earth's field) and the Coriolis force (due to the Earth's rotation). You can observe the effects of the Coriolis force with a ball on string, swinging as a pendulum. If you add to that a magnetic force you could possibly get a twisting action. It would all depend on geometry though and I'd really have to see the set up.
The reason it would only spin in one direction is the Coriolis force only acts in one direction, opposite the earth' s rotation.
The other possibility is the combination of two magnetics on the end of the string, set in opposite directions.
These are both just wild guesses mind you. You find me the details on it and I'll think about and get back to you. And I promise you that it doesn't involve a violation of the current understanding of physics. Its just an interesting combination of known principles. But without the details I can't tell you what those principles are.
I don't know. If you look at what was in the books 50 years ago, there are many things different from our understandings of today. Same applies to medicine, and other areas. Fifty years ago many thought they had it figured out, just as did those fifty years previous.
Take a gallon can of paint filled 7/8 full and immerse a sphere so its equator is at the level of the paint, the sphere being half immersed. The sphere is attached to a vertically-extending arm attached to a motorised cam, that causes the sphere to bob up and down at a low amplitude, like a fishing bobber. At a sufficient speed, ripples will be sent out and reflect from the walls of the can, causing standing waves to exist on the surface, these waves having crests and troughs. Now take a tiny second sphere having less density than the paint and place it in one of the troughs, say the third orbit out and give it a little push. It will orbit the center sphere even though there is no attraction between the bobbing sphere in the center and the little one in orbit. You can "promote" the little sphere by hitting it with enough energy to overcome the crest and it will settle in, and become trapped in the fourth orbit. If you hit it hard enough and the can is of large diameter, it should be able to escape to an area where the crests / troughs are of such low amplitude so as to not cause an effective barrier to motion.
Then consider the sun, its rotational axis does not align with its magnetic axis, and I can prove that these two can never be in perfect alignment. The result is a vibrating magnetic current and from the orbits of the planets one can calculate the boundaries of the solar system from which these waves are reflected. But, the factor of aether drift needs be taken into account, since it is what gives rise to the orbits being elliptical, not circular. I propose on this model therefore, that there is no gravitational attraction between the earth and sun, that the earth is merely in a trough. It is the same reason that electrons are not attracted to nuclei, otherwise all matter would collapse.