I’m a bit uncomfortable with the use of rudder to recover from upsets. The usual technique for recovery from unusual attitudes is unload, speed, roll, pitch. In other words reduce the G-load, control the speed trend with power bearing in mind the pitch/power couple, idle thrust for nose low, mid to full thrust for nose up, roll to the horizon, then pitch. Use of speed-brake should be cautiously considered against its effect of reducing available lift and compromising the roll authority of the spoilers.
I’m afraid I can’t think of any primary role for the rudder in this. Unlike the wings or the horizontal stabiliser, it is not built to withstand high manoeuvring loads. The rudder is primarily for achieving longitudinal balance in flight and counteracting moderate out of trim loads in the event of engine failure. Yes, the secondary effect of rudder is roll, but it has only a very limited application in the recovery from upset scenario. If yaw is required then assymetric thrust would be far quicker.
In all the years I’ve been flying big aeroplanes, I cannot remember any procedure that required me to rapidly apply full rudder followed by an immediate reversal. I have been conditioned to avoid the use of large rudder deflections at high speed. In many years on the 767, from memory the failure of the rudder ratio system after a hydraulic system loss (can’t remember which one) specifically mentions the necessity to avoid large rudder inputs to avoid structural failure.
Hindsight is a wonderful thing, but we do all need to analyse the actions of the unfortunate crew involved in case there’s anything we can learn. To do anything less would be to dishonour their untimely deaths. I’ll leave blame to the lawyers and suits, that’s about all they’re good at.