Yes, I've seen Brandt's reasoning in the past. But in the absence of any clear evidence in support of cleaning I really don't see the point of doing anything other than a quick wiping of the exterior of the chain. And that's just from the standpoint of cleanliness, not because of any expectation of increased chain durability.

My counter to Brandt is based on my experience with telescope mirror grinding where grit of a specific size is deliberately introduced to grind away and thereby change the surface shape of the glass. After only a few moments of moving the blank against the tool, the grit is pushed out to the edges and only clear water remains in the interface. So one must constantly keep moving the grit back into the middle. I'd expect the same process to occur inside a chain which is adequately oiled. I.e. any grit quickly migrates to the outside leaving clear lubricant at the actual interface where wear occurs.

I haven't seen any indication of a replication of the split-chain experiment either. Would have been nice if some of the people who criticized the particular cleaning technique had conducted their own experiment to confirm that a better technique would actually lead to greater durability.