Newton begs to differ
Instead of looking at the rotor disk as a whole, break the problem down into the individual rotor blades.
If we move our frame of reference from the rotor hub to the CG of the rotor blade, the blade is observed to be rotating about its CG just like a top or gyroscope. The aerodynamic forces as well as the forces imparted to the blade by the blade grip can all be transfered to the blade CG as simply one force and one moment acting at the CG. This is exactly the classical case of a force/moment applied to a gyroscope, and hence the blade responds to the force/moment EXACTLY 90 degrees out of phase with regards to where the force/moment was applied. That's Newton at work... there's no getting around it or arguing it away.
What confuses the matter is when the system is viewed from the rotor hub. It then becomes an articulated gyroscope with all the quirky behavior that such an object entails (such as responding at a phase angle other than 90 degrees)... but in the end it's still a gyroscope.
In my area of expertise (simulation), we live or die by being able to accurately predict the behavior of complex mechanical systems... one such system being a rotor head and its associated rotor blades. When you break down the equations of motion for such a system, they become exactly the same equations used to predict the motion of a spinning top.