It's a safety system — Page 322, Lesson 530

I want to walk you through a real-world case study that illustrates how human performance limitations — the very subject we're studying — can interact with technical failures to produce a catastrophic outcome. This is the story of British European Airways Flight 548, a Trident aircraft that crashed on 18 June 1972. Let's start with the technical background. The aircraft involved was a Hawker Siddeley Trident, and it had a system called a stick pusher. A stick pusher is a device that automatically pushes the control column forward to prevent the aircraft from entering a stall. It's a safety system. During a routine maintenance visit before this flight, the stick pusher system was overhauled. But here's the critical detail: on reassembly, the mechanic left out a small lock washer. That might sound trivial, but that missing washer had a specific effect — it would very occasionally let the pneumatic pressure in the system drop for a fraction of a second before recovering its normal pressure. So the stick pusher system had an intermittent fault, one that wouldn't show up on a normal ground check. Now let's look at the day of the flight — Sunday 18 June 1972. It was a typical British summer flying day: wet and windy with poor-ish visibility. There was a great deal of pressure on the check-in staff, baggage handlers, and everyone at Heathrow because of the number of passengers flying that day. Why? They were trying to avoid the consequences of a strike planned for the following day. So the whole airport was under extra workload and time pressure. In the hour or so before the flight, the captain — Captain Key — was speaking at a pilots' meeting, trying to persuade his colleagues not to strike the next day. He failed. And unfortunately, another pilot made some sort of comment to him just after the meeting about his lack of success. At that point, Captain Key had one of his famous explosions — he turned all the colours of the rainbow and entered into a full and frank, if one-sided, discussion with this other pilot. Two other crew members, Keighly and Ticehurst, were close by and witnessed this tirade. Immediately after this outburst, Captain Key seemed to be in some discomfort. The problem seemed to be centred around his chest. It now seems likely that at this point he had a minor heart attack. So we have a captain who has just undergone significant emotional stress and may have suffered a cardiac event before even getting into the aircraft. Now the flight itself. While loading the aircraft there were a few problems. At the last moment, space had to be found for a crew of a Vanguard freighter so that it could be brought back from Brussels before the strike. This put a little extra pressure on the crew at the last moment — they had to re-do the passenger manifest and all the weight and balance calculations. Despite this, the Trident pushed back from the gate approximately on time and began the taxi out. From this point onwards, a great deal of what happened is supposition. The aircraft was only equipped with a rudimentary Flight Data Recorder — that's the device that records aircraft parameters like altitude, airspeed, and heading. There was no Cockpit Voice Recorder — no recording of the crew's conversations or sounds in the cockpit. So many of the actions and thoughts of the crew are unknown. Let me pause and make sure you have the key terms clear. A Flight Data Recorder is an instrument that records aircraft performance data. A Cockpit Voice Recorder records audio from the cockpit microphones and area microphones. The Trident in this accident had only the basic flight data recorder, not the voice recorder, which means we have a very incomplete picture of what the crew were doing or saying in the final moments. So what we have so far is a picture of multiple pressures building up: a technical defect in the stick pusher system from a missing lock washer, a captain under extreme emotional stress who may have had a heart attack, a busy airport environment with strike pressure, and last-minute changes to the loading that added workload. And we have very limited data to reconstruct what happened next. This is the foundation for understanding how human performance limitations — stress, fatigue, emotional state, physiological events — can combine with technical failures in ways that are hard to predict or recover from.