For the last few weeks our Safety Moments have been on themes of Incident Investigation and Analysis. Topics have included:
- Safety Moment #21: Happy Trails
- Safety Moment #20: Young Man, It’s Turtles All The Way Down
- Safety Moment #19: Root Cause Analysis
- Safety Moment #18: Incident Investigation: Words, Words, Words
Let’s run one more post on this topic before we switch to another theme.
I would like to start by considering the difficult word prejudice. In modern English this word has, unfortunately, developed a negative tone as in, “She is prejudiced against < a group that she does not like >”. (It can occasionally connote a positive attitude — for example, when someone is prejudiced in favor of their hometown sports team.)
But let’s look at the makeup of the word “prejudice”. It consists simply of two parts: “Pre” + “Judge”. And, in that context we are all prejudiced because we all tend to jump to conclusions and because we all tend to select the facts and observations that support those pre-determined conclusions. Even the most thoughtful and experienced investigator will fall into this trap.
I recall working on one facility where one of the senior technicians blamed all our problems (and at that time on that facility there was no shortage of problems) on the instruments we were using. He would say, “These instruments from Company ABC are no good — so I don’t believe what they are telling me.” What had happened is that he had worked on another facility where the instruments were indeed of poor quality and they had been manufactured by Company ABC. So he carried this prejudice with him, even though the instruments on the newer facility worked well.
In another situation a high pressure cylinder broke loose and could have seriously hurt someone. I investigated this event — which was technically quite complex. After much analysis I and the other team members came up with a technical elegant explanation for what had happened. But when we looked at the damaged cylinder in the shop it became obvious that our explanation was incorrect. Yet we were reluctant to give up our wonderful theory. The facts were in our way.
The well-known literary critical Harold Bloom (1930- ) in his book How to Read and Why said, “Clear your Mind of Cant”. Cant means, 'Monotonous talk filled with platitudes' or 'Hypocritically pious language'. Bloom's advice can mean simply “Clear your mind”. It is critical that those conducting incident investigations follow his advice. This is not easy.
In Safety Moment #19: Root Cause Analysis, we quoted Oscar Wilde who said,
A truth ceases to be a truth as soon as two people perceive it.
What he meant is that, while facts are facts, each person will select and interpret facts according to their own background and experiences — their own version of reality. We all do it.
Failing to clear one’s mind of cant can lead to the problem of fixation. The examples that I have quoted to this point are relatively minor, and no one was injured. But a much more serious event — the meltdown of a reactor core at the Three Mile Island (TMI) nuclear power plant in the year 1979 — was partially attributable to fixation.
In a nuclear reactor it is vital that the water level be kept above a minimum level, otherwise the core will overheat. In the case of TMI the operators had many instruments telling them that the water level in the reactor was low and that they needed to add water. But one instrument was wrongly telling them that the level was too high. They chose to believe that instrument and they successfully worked to reduce the water level. Their fixation made a bad situation catastrophic.
Clearing one’s mind of cant also means that investigators should open their mind to lines of thought that do not fit with their own, normal way of thinking. As we discuss in Safety Moment #21: Happy Trails, most investigators tend to have a technical background. Yet it could be that insights from the IT or Human Resources groups could lead to higher quality analyses and investigations.
Copyright © Ian Sutton. 2018. All Rights Reserved.