I've been following the discussion. When reading the interim reports, I had a great deal of difficulty in understanding what was going on in the pilots' minds. After reading the final report, I would classify the accident as initiated by spatial disorientation and exacerbated by mode confusion. The ultimate factor was a stall with a botched recovery.
This wasn't just an Airbus problem or a sidestick problem. Look at the Northwest B-727 in December 1974. When I read the report on that one, all I could think was "how dumb could he be?" A few months later, I found out just how dumb I could be. The only difference was I flew out of the cloud before losing the airplane.
In all three cases, there were multiple indications caused by a single common cause. The problem is that human pilots are not very good at working back from multiple indications to a common cause. If you had asked all three pilots what the effect would be of iced pitot tubes, I'm sure that they could have told you. I think that given multiple indications, first on A/S, then a second, then autopilot disconnect, then ... and so on, it would be much more difficult.
There have been a number of studies of multiple failures in transports and in light aircraft. Look at Gideon Singer's doctoral dissertation for example. We should all remember that we've had thirty seven months to study the cause. The crew had a couple of minutes.