BOAC,
A. My reference to this considered that there was an ‘invented’ stall recovery procedure in normal law involving a pull up because the protections were there (#816).
B.
AFAIK, correct.
C.
AFAIK, correct.
Thus combining A (on the premise of B) with C, and due to poor situation assessment, surprise, inappropriate training, application of automation, etc, the PF pulled up believing that the aircraft had stalled at the time of speed failure, the resulting climb did actually stall the aircraft. Then by supposition, the ‘normal law’ mindset maintained the control input resulting in complete confusion, particularly when back stick stopped the stall warning and forward stick started it (AoA cut-out logic).
HN39 “IMHO you are reversing cause and effect.” 
Hence my consideration of the ADC, or even turbulence / poor handling with the lower stall threshold?
“
Loss of speed …”, I agree, but if the mindset was that of ‘a higher priority’ stall warning, then what?