Originally Posted by
737 Driver
I pretty much agree with everything you say except for the "any human" reference in the second to last sentence. I think that there is broad, though not unanimous, agreement that the accident crews made some serious errors that led to the final loss of aircraft control. Where I and some other participants here differ is that I strongly believe that much of the human factors element you refer to is amenable to training. One only has to look at the performance of the crew on the Lion Air 610 flight the day prior to the original accident. Despite being presented with a novel malfunction, one of the pilots kept flying the aircraft. IMHO, that crew took a bit too long to get to the trim cutout switches, but the takeaway is that every time MCAS made an input, the flying pilot took it right out again. At no time was there evidence that they were losing that fight.
It has been suggested that this crew would have ultimately crashed if not for the jumpseater suggesting that they try the cutout switches, but frankly that is an unwarranted assumption. Having been both in the flying seat and the jumpseat on many occasions, it is absolutely true that the jumpseater may catch something quicker, but that does not mean the flying crew will not catch it at all. There definitely appears to been a limited understanding of the stab trim system by both the Lion Air and Ethiopian crews (very much amenable to training), but it did not prevent the crew above from maintaining aircraft control.
I will agree that any human pilot at some stage in his training will be easily overwhelmed by even the most basic aircraft emergency scenarios. That is why we train so extensively for them. I had previously posted that in the Ethiopian accident, the Captain did fly the aircraft after a certain fashion, and that he had defaulted to his training. The problem was that he defaulted to the wrong training. Just about all of the ET302 Captain's initial actions can be understood in the context of a normal takeoff profile. Unfortunately, a normal takeoff profile left him in a highly unstable position from which dealing with the ultimate stab trim problem became a bridge too far.
When I and other posters keep saying that the pilots should have kept FLYING THE AIRCRAFT, we do not say this from a perspective that operating a malfunctioning aircraft is some inborn capability that every person has. Heck, walking isn't even an inborn capability. However, it is a skill that is amenable to training. That is, as long as you get the right kind of training. As has been already discussed extensively, training in modern commercial airliners has largely devolved into a process of following scripts. Pilots are presented with known problems with known solutions. Even at my airline, there is much less of the relatively unscripted training that really drove home the need to set aside any distractions and focus on the basics of flying the aircraft without the benefit of any automation until the situation was stabilized.
There are multiple links in the chain of causation leading to these accidents. I don't think anyone is questioning that MCAS needs to be fixed, or the FAA needs to step up its oversights, or that airlines need to review their internal training and operations policies. The professional pilot corps, however, needs to look at what we can do to correct the airmanship deficiencies that were exposed by these accident.
When you say that you "strongly believe" the issue is amenable to training, is that based on research or on intuition? I ask because my understanding of the broad sweep of behavioural research is that actual human behaviour is often deeply unintuitive, and in a safety-critical environment I'd be very wary of relying on any assumptions about human behaviour that haven't been rigorously validated.
(I fully agree that pilots need to do their part in correcting any systemic errors contributing to these incidents; my concern is that treating them as "human error" or "poor training" passes up an opportunity to better understand how and why people actually make mistakes, and how to avoid falling into the same traps. Like, I read wonkazoo's underlying point about his crash as being "I knew exactly what to do and I *still* almost killed myself, because a real incident is not like a simulated one", and that seems like it argues against more of the same sort of training as a solution to that class of problem.)