@ PJ2 ....."There are only a few reasonable theories which can explain why a rational, trained airline pilot would pull back half the available stick deflection and, with back-and-forth variations, keep it mostly in the NU position, even in spite of the fact that a minute later, pulling back does not arrest the rate of descent and the stall warning sounding almost constantly."
I think I agree, but would suggest the insertion of an alternate to 'reasonable'. Can we try 'obvious'? Many possibilities have been mentioned, and rejected, in some cases, without objectivity. Bias is a subjective species of opinion, and should play no part in a detached investigation. "Reasonable" is a matter of opinion, and led in some cases to the repetitive presentation of fatal accidents, simply because the cause was not 'reasonable'.
Mention airframe damage, and the gallery frowns and stomps its feet, likewise computer failure of massive nature. In this regard, I have great respect for BEA, they do not make conclusions, only supposition. The final report will be similar. I expect an objective and thorough document. Throughout each report, the authority is careful to not make dramatic conclusion. The downside is that by exclusion, readers of the reports "conclude" fact. Damage, for instance, is possible, but not seriously addressed; therefore the popular conclusion is "Intact at Impact". A serious sceptic would respond: "Define Intact"........