Question:
Does anyone know of any emprical research on incidence and rate of spinal/neck trauma among helmeted and non-helmeted cyclists, whether fatal or not?

Here's why I ask. The one thing most people can be counted on to do in a fall is to try to protect their head. What makes a difference is how successful they are and whether or not they even have a chance to do so in the first place. THe idea of a helmet is to provide some measure of additional protection for your head. However, in a generally head-first fall with significant forward motion and an oblique angle of impact (i.e., hitting the ground while moving, but less so in the case of a car or tree), it seems likely that spinal damage to the neck is almost as likely to kill or damage you as the impact to the brain. If you're wearing a helmet, the forward surface impact area is increased, and the helmet vents offer excellent traction to a rock, root, curb etc. It seems possible that in this (relatively-common?) type of bicycling fall, the helmet may reduce the risk of damage to the brain from cranial impact, but increase the risk of injury to the neck by snagging on the impact surface and applying additional leverage to the neck as the body continues moving past the suddenly-halted head. On the other hand, if there's a pointy surface involved, sufficient to snag a helmet vent, then without the hat you've probably got a very rough lobotomy. It would only make a difference in the relatively few cases where its the extra area/volume of the helmet that impacts when a bare head would clear, leaving the body to absorb the main impact. But I wonder is this risk significant, given that the genetic instinct is to avoid head impact, but that instinct doesn't account for the larger helmet size. After all, how many of us practice throwing ourselves from our bikes, just so we can execute a flawless aikido roll with helmet in place?

anybody seen any stats on neck trauma with and without helmets? And relative risk of severe injury from head vs neck trauma while biking?

Answer: I don't recall any specific research on this. As the cynics (certainly not myself) would tell you, since most head-injury research is conducted on behalf of the helmet industry, they would ignore neck injuries since at best a helmet does nothing to protect your neck. OTOH, some people put on a helmet and think it protects them from a head-on with a semi, so there is an additional danger there. There was a time when some helmets consisted of a foam core covered with a fabric, rather than shiny plastic. Rumors were that such covers grabbed on the road more than no helmet at all, so increased the chances of neck injury, but I don't believe this was supported by anything more than the claims of those selling helmets with plastic covers. The helmet does a bit more than that. If you look carefully at it, a helmet is normally a thick, rigid surface (foam mainly) with some padding built in for comfort. Structurally the helmet serves as a force redistribution system: it takes a concentrated force and distributes it over a large area (the whole head) so the resulting stress is significantly lower than if the force was applied directly. The helmet also stores some energy as deformation during impact, and releases it in a delayed manner so you don't transmit the full force of impact through your head at once. Put on your helmet and hit it (lightly!) with a hammer... you'll see what I'm talking about.

When you fall you try to protect your head, true, but the problem is that many times the protection you can provide to the upper and rear cranium using your arms and hands is minimal.

A helmet is designed to protect the head, not the neck, so comparing neck injuries with and without helmets is of little significance. You're asking something from the helmet that it's simply not made to do.