Academic Airmanship Paper

This week I read an interesting research article on airmanship in standardized airline cockpits. The lead author is Torgeir Haavik, a professor at the Norwegian University of Science and Technology with an engineering background in oil drilling, who more recently earned a PhD in the sociology of risk and safety. The paper is wildly — for an academic journal — titled: ‘Johnny was here: From airmanship to airlineship’. It was published by Applied Ergonomics journal in 2016. So:

• Who’s Johnny?
• What’s airlineship?
• And can pilots learn anything useful from the journal Applied Ergonomics?

Read on, all will be revealed!

“Airmanship belongs to a class of notions or phenomena that lack a precise definition, but have a great impact in the domains where they are used. The impact is not necessarily measurable in terms of predictable successful outcomes in the operations that take place in these different domains, but rather as something that success is ascribed to post hoc, thus contributing to the nurturing of professional identity and vaguely defining a professional gold standard that is sufficiently elastic to account for outcomes where other explanations fall short.”

Agreed. Airmanship is hard to define, but we agree it is the core of personal professional processes that make the difference between a pilot and a passenger. It gets even trickier to define in military or airline environments where the rulebooks run into hundreds of pages, and years of standardized training in standardized procedures leads to nearly uniform basic performance. The research project “seeks to understand professional competence in a context of increasing standardisation of regulation, procedures and technology.” Good stuff.

They did this by office interviews with pilots and sitting in the jumpseat of an airline B737-800 during revenue fights and observing the pilots’ actions, noting things like variability, adaptations, tinkering, tailoring, and so on. Again, good stuff. But here is where I start to wonder about this research: The observers were not qualified line pilots, and they observed a total of five roundtrip flights. So ten legs! I can tell a lot about an Airbus pilot simply by watching the taxi-out, but I have 20,000 hours of flight time, with well over half of that in the ‘bus. To do research like this, I’d want a much bigger sample size. Something like, for example, the cool work the LOSA Collaborative do for airlines, with 25,000 rides and counting. LOSA has a high degree of aeronautical understanding, and would never write “Although it is possible to descent and land fully automated, the pilots need to work on the flaps during the descent in order to reduce the speed.” Scientific researchers understand sample size power and false implied causation. So just ten flights by observers who think working on the flaps reduces the speed is the first strike.

So I don’t think they gathered much good data. OK. But what are their results?

They found, unsurprisingly, that the “most striking characteristic of pilot work is the little degree of variation regarding how the pilots performed their work during the different flights.” In a general sense that is true, but it is the differences that define us, and maybe the researchers just didn’t have the knowledge to see those fine details. Plus, the likelihood of what some call ‘angel performance’ is high when people are being observed, however much they are told the data is confidential or de-identified or whatnot.

The interviews did produce an interesting airmanship insight:

“Good airmanship is to deviate from the SOP when necessary”, one pilot explained. To do this, experience is crucial. And experience is personal. Thus, however standardised, however depersonalised: adaptations of procedures as well as technical environments do take place in order for individual pilots as well as airlines to perform well and with good flow.

Pilots know this of course. But it’s good to see it laid out in public. There is also mention of alienation due to decreases in pilot working conditions, and the possible negative consequences for safety. Good stuff. I never got comfortable with their word airlineship though. They define it as:

The inter-organisational efforts to create predictability and safe practices through de-identification and interchangeability of personnel and aircrafts. 

And later state that while “airmanship in terms of the individual pilot’s discretionary manoeuvres still represents a useful, though not sufficient, reference, there are also numerous examples where the complex interplay between pilots and organisational, regulative and techno-material structures call for a more systemic trans-local perspective.” This is what they call airlineship. I don’t think that word is going to catch on. I’d put money on that! The systems interplay is real, but it exists in several other industries, and deserves a more domain general, yet more specified, name.

I do think their conclusions on airmanship are valid:

We share the opinion of our informants that pilots and airmanship are still vital to aviation. Airmanship is about discretionary performance of standard operation procedures, about tinkering the right ways with an underspecified system. 

Maybe this cartoon (not from the academic journal) sums it up:

Part — just part — of airmanship is using the yellow book. It’s good to read these academic journals, to deep dive into important ideas about how we make decisions as pilots and cockpits and companies and industries.

Oh, and who’s Johnny? He’s a pilot who brings his personal headset to work, and doesn’t always unplug the company headset. Other pilots know that Johnny was the pilot before them because of his non-SOP trace in the social-techno system. A little cog that actually has a name. Is it important? No, I think it just makes for a cute title.

.

Haavik, T. K., Kongsvik, T., Bye, R. J., Dalseth Røyrvik, J. O., & Almklov, P. G. (2017). Johnny was here: From airmanship to airlineship. Applied Ergonomics, 59, 191–202. https://doi.org/10.1016/j.apergo.2016.08.028