I like this quote from D.P. Davies regarding "superstalls".
There is no point in discussing the irrecoverable case any further, except perhaps to say that those aeroplanes which have been lost in such manoeuvres finally reached the ground substantially level laterally, having defied all efforts to roll or spin them out of the stabilized condition; only slightly nose down in pitch, with little or no forward speed; at an extremely high incidence; rotating only very slowly in yaw; with (in one case) all the engines flamed out because of being exposed to such massive angles of incidence; and finally with an enormous vertical velocity.
The quote seems very relevant to AF447.
HN39 quote:
As far as I am aware, no one on this thread really knows how the airplane was certificated in this respect. Some contributors, for example PBL, on the basis of a literal interpretation of the relevant regulations, have suggested that an investigation of its stalling characteristics beyond "alpha max" was not required, by virtue of the AoA protections incorporated in its flight control system in "normal law".
In the vicinity of alpha max I would expect predictable stall behavior.
Well beyond alpha max, I would expect stall characteristics approaching those of a swept wing flat plate airfoil-in other words, terrible.
That work-of-art Airbus airfoil, optimized for cruise, was probably never even tested well past alpha max. I expect that the Airbus AOA protection system was considered to be better than a stick pusher and thus sufficient to prevent flight outside the design flight envelope.
Somehow, AF447 exceeded its design envelope, and that is the question which the information from the black boxes needs to answer.
It may well be more complex than pitot icing. Let us hope that the information is still available.