I think I understand what TTex is pointing at.
There was only a small trim change in THS position leading up to the stall, but if that trim change had not occurred, the aircraft would have sought to fly at its original trimmed speed and PF would have had to work mighty hard to slow it down to a stall. Dozy does not really appreciate the significance of this because his piloting experience is minimal.
Yes the Airbus THS trim worked exactly as it was designed to. So What? In retrospect, that might have not been the best design from a human factors standpoint.
Yes, any self-respecting Airbus pilot who could fly the thing in ALT2 law at altitude would appreciate the flight path stability offered by the current implementation of Alt2 which involves similar behavior to Normal Law in pitch and Direct Law in roll.
However a pilot who badly overcontrols an Airbus in roll and doesn't know how to stop it will have serious trouble controlling other axes at the same time. With such a loss of precise pitch control ability, the flight path stability of the Airbus becomes a liability, because it tends to mathematically integrate inadvertent pitch inputs.
Such a pilot would benefit from a direct drop to manual THS trim following loss of speeds. This essentially means Direct Law in pitch also, since Alternate law in pitch without autotrim does not make sense.
Perhaps the law of lowest common denominators (i.e. weakest pilot skills) should apply as the difficulty of flying increases. From all accounts Alt2 at altitude is not particularly difficult if one is smooth, but can be a real bear if one is using the stick to wipe the cockpit.
I'm going to offer the opinion that if AF447 dropped to Direct Law on loss of speeds, even with the crew she had on that night, the odds of arrival at the other end would have dramatically improved. Then the raw airframe's natural speed stability would have helped them avoid the low speed trap that they fell into. At least, they would not have crashed in the same manner.
