The vast array of research pertaining to the "core" has made me consider the importance of variability when looking at spinal stability.
It seems as though the greater the variability in the "system", the greater the stability. Peter Reeves of Michigan State University uses a "plant and controller" analogy to describe this in great detail. He likens plant to the spine and its interbertebral joints and musculature. The controller, conversely, includes the reflexive and voluntary pathways that travel through the spinal cord.
When considering the system as a whole, that is the plant and the controller, the greater the stiffness, the less likelihood of dysplacement of that system...therefore, stability!
Unfortunately, however, a stiffer system does not always equate to optimal performance.
Consider the example of walking on a log.
Our system (the plant and the controller) may be better off as one that is open and available to react and respond to the dynamic and changing environment. Reeves doesn't believe that a system that displays greater stiffness is ideal in this situation. He suggests the possession of greater variablity is ideal...motor control, if you will.
Unfortunately, this system may fail in the presence of LBP. That is, when feedback control is altered. However, we must fear not. A "bottom up" approach that incorporates some form of manual therapy may be the intervention of choice for our patients.
Back to stiffness.
Don't get me wrong. I do think stiffness is important. However, it may just be an intermediate step. Consider a maximum effort deadlift. Consider the player receiving a body check in ice hockey. Here, system stiffness would be valuable for stability sake. Please remember that one goal in rehabilitation is to have the motor control system subconsciously learn more stable patterns (thanks Dr. Liebenson).
So perhaps Pr. Stuart McGill's research on rate of contraction/relaxation (pulse) may allow "stiffness" to play a larger role in the dynamic environment. The only downside is that this often seems to be voluntary and does not necessarily include the reactive component of neuromuscular control (edit: the "bottoms up" kettlebell exercises do).