Thought through this a little more and came up with something that I think is worth presenting here.
When you apply forward cyclic, you reduce pitch on the right and increase it on the left. This means that while tilting the disk, you generate a dissymetry of lift. However, since the forward blade experiences increased alpha and thus generates more lift, the disk tilts to the right.
So you decrease lift on the right but at the same time tilt the disk to the right. Wouldn't that just be crazy if those two effects cancelled each other out with the result that no lateral cyclic results in no roll during abrupt manoeuvring?
What I was trying to say before is not quite that the quote was a load of crap, just that it wasn't complete. It didn't consider other transient effects on the disk, and it didn't consider whether the (weewa?) effect was desirable or not.
Matthew.