I should rephrase, below the resonant frequency of the system you no longer have mechanical damping - you're relying entirely upon the electrical damping inside the amp to control the cone excursion. And it gets worse when you consider the non-linear behavior of the air on the cone as the piston becomes smaller and smaller relative to the wavelengths being reproduced.
One of the analogies I've heard compares displacing X volume of water with a rod (~1 cm^2) and a frisbee (~20 cm^2). Though you have to move the rod 20x further for the same displacement, it's going to move much easier than the frisbee. (The analogy of course assumes that there is an appropriate baffle present). Though moving to more drivers increases the overall displacement, it also increases the area of the piston - making the displacement more effective.
Another factor that comes into play is the non-constant pressure in front of and behind the driver. When a low frequency sound is sent out into the room, some of that is going to bounce back and arrive in front of the driver (same thing happening in the rear too). The changing pressure seen by the driver is going to change how it behaves - if a large transient hits while there is a null in front of the driver, you're going to realize far larger excursions, yet measure the same SPL. That cool fan sub takes advantage of this effect by creating air pressure with the fan blades - and then only small excursions are needed for the same SPL. The opposite effect is that a lower pressure is going to cause over-excursion. I've been told that this same physical system is why passive radiators often need 4x the excursion of the active driver.
I'm by no means a driver expert, but I thought some of these ideas were rather interesting. I don't think it's a huge sacrifice to trade cracking walls for a system that doesn't blow itself apart
