This has been discussed exhaustively over on Bogleheads. I think (others obviously may disagree) it boils down to:

 A) Bogle is saying it is unlikely to be worth a 1% management fee to have a frequently updated rebalancing - it simply does not pay:

"...suggests that formulaic rebalancing with precision is not necessary...."

 B) That said, the purpose of an AA is to have a portfolio that meets your personal RISK requirements. As long as it does that, you do not need to rebalance. IF you find over time it has 'drifted' (some people use 'bands), then you rebalance to put it 'right', OR if your risk tolerance changes, you should rebalance.

 C) I personally wish that the whole concept that there is some measurable and useful GAIN to be had by re balancing had never been floated, because, IMHO, it has completely distracted from the whole purpose  -- it is not to maximize your return, ti is to maintain your risk stance.I believe that is consistent with Bogle's past teachings.

d) I am not looking for a fight, but JWR's reasoning for not rebalancing has been that diversification in general is only useful to "avoid fraud." (?)

I have yet to see him explain/support that interpretation. The fact two people agree (i.e 'buy pork bellies!') does not necessarily mean they both working from equally sound reasoning - one could have an advanced algorithm for futures trading, the other could just want some bacon. (smile)