Here's something (not what I remember, though):
At issue, Michael, is the formation of sediment. That part of sediment that is not tartrate crystals has been shown to be polymerized phenolics. The problem here is that the polymerization of phenolics is an oxidative process. I can do it in the lab, and I guarantee you that it doesn't work without some sort of oxidant. Since we know that sediment forms even before that cork is removed, we need to find the source of oxidation and oxygen is the obvious (perhaps even correct!) culprit. I do realize that Peynaud has argued strongly against this view, but I haven't seen the data that led him to this conclusion.
Mark Lipton