Lyman
A ballistic STALL is gentle, almost graceful, and there is no "Drastic increase in drag", and no compelling urgency for the Nose to drop. Buffet is likewise attenuated, and as in this case, less likely to cause alarm, or alert. The AoA was extreme, and this causes the functional Cg to locate aft, further reducing the normal signals of STALL as well as make recovery more problematic.
I really cannot agree with anything you have written here. Drag certainly increases dramatically when the wing stalls and the loss of energy came from an excess of drag over (thrust plus weight component along the flight path). For an explanation on the mechanics of wing stall on a modern airliner see Thread 5 #176 and 297 and Thread 6 #457. What evidence do you have to justify your claim that buffet is attenuated, when it is clear from the data published that there was a clear change in cockpit 'g' characteristics at the first stall and BEA stated in their final report that AI pilots had confirmed the (subjective) stall limit as severe buffet at 10 deg AoA? [I accept that on 447 it might have been masked by response to turbulence but that is not at all the same thing]. What on earth is 'functional Cg'? and how does AoA affect it?Are you postulating fuel slosh? and if you are would you like to put some numbers to it rather than vague speculations?
And wing stalling AoA is independent on whether the pitching moment necessary to get to that AoA is provided by THS, elevator or THS and elevator acting in concert. On AF447 BTW, once the aircraft had departed from initial trim the THS was giving a ND moment, offset by a lot of up elevator.