There is another factor in play here, and it is likely present in the AF447 accident.
Part of my work as a contractor to the FAA has been to build an inflight and ground icing accident database. This involved the review of a very large number of icing incidents and accidents; for the 309 events dating back to 1983 that were deemed applicable, I reviewed all of the available data, i.e., NTSB dockets, pilot statements, ATC transcripts, DFDR plots, etc. In this set of events, I identified 19 cases yielding enough information to determine that the pilot’s first response to an ice contaminated stall was to pitch the aircraft nose up. This ranges from twin Cessnas to the ATR at Roselawn to an A300 at West Palm Beach.
Now, in many of these cases, the pilot’s next response was to push the nose down. But what has interested me is this initial response; it is almost always immediately following the initial pitch down and/or altitude loss resulting from the stall.
This has led me to suspect that we have a strong “compliance” reflex. The first response is aimed at returning to compliance.
Several years ago, shortly after returning to work for the world’s largest airline, I was in the right seat of a 767 returning from Europe. The relief pilot was in the left seat as we approached Labrador. We had been given a re-routing at the other end of the North Atlantic Route, and when we loaded that into the FMC, it dumped the NAR out of the route. Consequently, we were about four miles from a route discontinuity. The poor relief pilot was in a near panic, because he did not want to be seen as straying off course. As it happened, the correct track for the now-missing NAR was exactly the track we were on, so all that was imminent was a change from LNAV to HDG HLD, followed shortly by a return to LNAV once we reloaded the NAR. (In the end, I was able to reload the NAR before we reached the discontinuity.)
In another example, while recently executing a fairly prompt return to landing when my MD80 started emitting a burning odor shortly after departure, my ex-military, newly hired first officer was likewise very concerned about the prospect of landing a thousand pounds overweight. I pointed out afterward that such was the value of declaring an emergency.
It occurs to me that these are forms of the same compliance reflex. The relief pilot in this 767 case was visibly stressed about being seen as non-compliant by Canadian ATC. My MD80 FO was worried about deviating from the approved limitations. It seems possible that such a stress response may be at the root of the 19 events in our icing database. The same thing may have been involved in the first few seconds of the AF447 sequence; I don’t know.
So in parallel with the question of automation dependence may be a fear of violation or non-compliance. This goes straight back to the sense of autonomy that should be inherent when ensuring that “the successful outcome of the maneuver is never in doubt” (ATP Practical Test Guide…or at least it used to say that).
Many companies and some authorities tend to behave in ways that reinforce this. Years ago I remember an internal company training video in which an MD80 captain was interviewed regarding a high altitude stall that he had managed to get himself into, and obviously out of, with a 4000 foot altitude loss. They had actually blurred the fellow’s face as if he was in some kind of witness protection program, and the whole theme was about how you really, really did not want to have to “one of those meetings” with the FAA.
After 30+ years of accident investigation, I can tell you that what you really, really don’t want to do is have “one of those meetings” with the crewmembers’ family members. The FAA can take a hike.
Just food for thought…