Hi,
I was wondering why phonons should have spin zero, too. After all, the quantized system looks just like QED, only with more (!) polarizations, so even in the simplest acoustic case we have one longitudinal and two transversal modes. This looks like spin 1.
On the other hand, spin is defined by the Lorentz-group, i.e. the behavior under rotations, which is not exactly usable in a lattice. So my guess is that this reduction of the full rotational group to the discrete rotation group causes the concept of spin to break down.
Is this about right? Books are amazingly useless when it comes to phonon spin...
regards,
torus