Entangled photon pairs can be created by passing a laser beam (left) through a crystal. The crystal can convert a photon into a pair of photons, each with half the original energy, one polarized vertically (red cone) and one horizontally (blue cone). Along the cone intersections (green), neither photon has a definite polarization but their relative polarizations are complementary. They are entangled. (Colors here do not represent wavelengths.)

Entanglement

By Gautam Naik
The Wall Street Journal, May 5, 2009

Edited by Andy Ross

When two photons get entangled, they behave like a joint entity. Even when they're far apart, if the spin of one particle is changed, the spin of the other instantly changes, too. This entangled behavior is called nonlocality.

Nicolas Gisin, a scientist at Geneva University, led a recent experiment demonstrating nonlocality. Gisin and his colleagues entangled a pair of photons in their lab and fired them along several kilometers of fiber-optic cables of exactly equal length. During the journey, when one photon switched to a higher energy level, its twin instantly switched to a lower one. But the sum of the energies stayed constant, showing that the photons remained entangled, and the team couldn't detect any time difference in the changes. The conclusion: there was no communication and so there is nonlocality.

Other scientists have recently resolved Hardy's paradox. In 1990, the English physicist Lucien Hardy devised a thought experiment. The common view was that when a particle met its antiparticle, the pair annihilated each other. But Hardy noted that in some cases when the particles' interaction wasn't observed, they wouldn't annihilate. The paradox: Because the interaction had to remain unseen, it couldn't be confirmed.

Scientists from Osaka University used extremely weak measurements that didn't disturb the photons' state. By doing the experiment many times and pooling the results, they got enough good data to show that the particles didn't annihilate. So when particles aren't observed, they behave differently. In their March 2009 paper, the Japanese team said the result was "preposterous." A team from the University of Toronto published similar results in January.
 

AR  Nature behaves as quantum mechanics predicts — no surprise here. The surprise is that after eighty years or so we still can't seem to get our heads around the paradox of nonlocality.
My big idea here is to say spacetime expands from moment to moment, and entangled particles first get distinct locations at the moment of disentanglement (the pop), but this raises the question of whether we can assign consistent times to all locally simultaneous pops.