What is a fact is that Boeing FCOMs in last 25 years, ever since EFIS FMC a/c were introduced, now contain less & less technical information than previous generation a/c.
the notion that the aircraft commander IS the legal commander responsible for the safety of the flight, and in the end is the sole decision-maker on board the aircraft is gradually being made subservient to the audit process where such authority is "modified"
To expand a bit on my earlier post, these two observations actually form the important pattern.
Soon after WWII, Herbert Simon, who later was an early developer of artificial intelligence, said the following, one of my favorite quotes:
“Two persons, given the same skills, the same objectives and values, the same knowledge and information, can rationally decide only upon the same course of action. Hence, administrative theory must be interested in the factors that will determine with what skills, values, and knowledge the organization member undertakes his work.”
This idea, the suggestion that the human mind is simply a binary computer, is at the root of observations like
The manufacturers believe that knowledge will lead to thinking instead of performing like automatons in the event of abnormal events.
This is not necessarily a safety strategy so much as it is a management style. The only way a human being will fit into a flow chart is if you convince yourself that Simon’s idea works. (To be fair to Simon, he was a remarkable polymath who evolved considerably and made substantial contributions to many areas.) Thus, as a manager, you are primarily concerned with “what skills, values, and knowledge the organization member undertakes his work.” Naturally, it is easier to specify the course of action that you would like your human to rationally decide upon if there are fewer skills, simpler values, and less knowledge.
This exists in areas much broader than aircraft systems knowledge. We have new hire first officers straight out of the military who have no idea what an Operations Specification is. How would they know? The company doesn’t issue them to the pilots. They didn’t have OpSpecs in the military, and no ATP course here in the States would ever go so far as to explain how OpSpecs work and why they matter. The company certainly doesn’t want to spend any time on this; the instructors themselves don’t have the OpSpecs. The same ex-military pilot may not know that there is an FAR that says you can’t take off if you know you will arrive at the destination over the max landing weight…nobody ever provided any training in what the FARs say. (Not an indictment of ex-military pilots; many, particularly the former KC-135 guys, hand fly quite well for some reason…)
Reading Langewiesche’s piece on AF447, the most poignant passages are the CVR transcriptions detailing the frustration and near panic felt by the subordinate pilots when the captain did not immediately respond to their calls. Assuming that the translation is accurate (I always worry that meaning is lost in these endeavors), it is clear that neither pilot believed he had the technical skill necessary to resolve the situation. Moreover, they somehow believed that the captain, by virtue of his greater experience, would have that technical knowledge.
Thus, in the stress of the moment, they revealed an inner perception of themselves as inadequate by virtue of inexperience. The “skills, values and knowledge” that they possessed were not up to the requisite course of action. They knew it, and they were clawing their fingernails raw trying to get at that knowledge. How on earth did they graduate from a technical program that awarded them a type rating without the confidence they needed at that moment?
The problem, as I have said in an earlier post, is that all of the management theory generated by Taylor, Simon and others operates within a linear mathematical paradigm. It uses the same understanding of cause-and-effect that babies use when predicting the motions of billiard balls. They probably could not have done otherwise; chaos theory and the understanding of nonlinear behaviors did not exist until the 1970’s for all practical purposes. However, in the cockpit, we actually operate in an environment that frequently exhibits nonlinear behavior. We always have. Turbulent flow off a wing is nonlinear. Weather is nonlinear. Even the function of neurons in the brain is nonlinear. Complex systems, which we already had in aviation and then expanded exponentially with automation, exhibit emergent behavior at the least, in which the system output is something more or less than the sum or product of its components.
Standard operating procedures, when well designed, function to protect margins of safety. The margins exist to provide resilience and can absorb nonlinear effects. There are two ways to comply with SOP. In one, you simply do what you are told. In the other, you understand the margins that you are protecting, understand how that SOP accomplishes that, and you comply intelligently, as an act of executing your authority as well as an act of mastery over the aircraft. In the former approach, you become fearful of noncompliance, and pull the nose up as soon as the nose drops regardless of why. In the latter approach to SOP, you preserve the protections and error traps built into SOP while being much less likely to follow the book into the ground.
Linear management theories don’t see the difference.
Nonlinear behaviors are why I am not terribly worried about autonomous airliners in my lifetime. Eventually, sure…but not for quite a while. But much, much more important is how we tailor our profession to meet our obligation as the final authority as to the operation of the aircraft, in an increasingly complex system managed by people who actually think that “Two persons, given the same skills, the same objectives and values, the same knowledge and information, can rationally decide only upon the same course of action…”