How Swiss Cheese can explain tragedies

Photo by Trainholic, CC BY-SA 4.0 <https://creativecommons.org/licenses/by-sa/4.0>, via Wikimedia Commons

By Moritz Krömer

It is the 6th of April 2014 and the MV Sewol, a passenger ferry carrying 476 people, departs from Incheon, South Korea. It is heading to Jeju Island. While driving through the Maenggol Channel, the crew loses control over the steering. The ship turns and tilts. The Captain is not at the bridge. He had earlier transferred the control and responsibility to his less experienced crew members. Soon after the original tilt, the MV Sewol capsizes. Water enters the ship, making it sink slowly but steadily. Still, the crew encourages passengers to stay in place. They even recommend via announcements not to leave the ferry. However, as soon as the first rescue team arrives, they are the first ones to escape, leaving the sinking ship together with all its passengers behind. More and more lifeboats arrive at the scene, but the rescue teams are not able to help most passengers, and the situation develops into a tragedy. Over 300 people die or go missing, including 250 high school students, leaving the Korean society in a state of tremendous shock and grief.  

Following the tragedy, the South-Korean media and law enforcement agencies focused on finding people to blame for the accident. They put the emphasis on the people involved, as is typical in similar situations. As a result, 154 people were declared (partly-)responsible and were put in jail, including the Captain and his crew. The actions of the captain especially seem to indicate negligence. He left the bridge despite the consequences of his actions being easily foreseeable. The Captain should have been aware of the risks, and he should have acted accordingly. But he did not and he was found guilty of murder and sentenced to life in prison. 

But do we maybe judge too fast? We should better consider some questions before drawing hasty conclusions. Why did the crew lose control over the ferry in the first place? And why did the rescue teams not manage to save more passengers? Is it just to blame a tragedy like the sinking of the MV Sewol mainly on human error of the Captain, the crew and other people involved?    

In the aftermath of such catastrophes, people are overly motivated to explore the underlying causes.  Especially catastrophic events call for an explanation, and people tend to look for someone to blame. Blaming individuals is easy and emotionally satisfying. Further, there is a widespread false impression that by removing the individual who committed the error, you can prevent similar tragedies in the future. Adding up to this misconception, knowing the outcome of an event can lead to the false impression that the outcome could have been predicted and foreseen. This phenomenon is called hindsight bias. Falling prey to hindsight bias leads to distorted beliefs of foreseeability. These beliefs, in turn, can influence and strengthen the pre-existing opinion that unskilled people cause disasters. This extreme person approach might have contributed to perceiving the captain as being the main culprit.

Yet, sometimes it helps to see the flaws of a system as a whole. As shown by Oh (2017), Professor at the Department of Public Administration at the University of Akron, multiple causes lined up before the MV Sewol tragedy happened: The ferry was already heavily overloaded. The personnel of loading and lashing companies did not check the cargo load´s weight before departure. Neither did they tie the cargo load properly. These factors severely contributed to the instability and a lack of control over the ship. Moreover, lead government agencies failed in guiding coordinated and collaborative responses to the incident. The Korea Coast Guard (KCG) could not arrange joint operations as planned, and the MV Sewol crew lacked proper training for emergencies. Additionally, Hyungju et al. (2016), researchers of the University of Trondheim, found a total of 23 different causes, suggesting a combination of failures, related to the shipping company and government-affiliated organisations.  

Visual Illustration of James Reason Swiss Cheese Model of Accident Causation

Source: Davidmack, CC BY-SA 3.0 <https://creativecommons.org/licenses/by-sa/3.0>, via Wikimedia Commons

We can visualise the interplay of different causes by using slices of Swiss cheese as an analogy. This is referred to as the Swiss cheese model of accident causation. The model states that there are many defensive layers to prevent catastrophic events. These layers, pictured as slices of Swiss cheese, rely on people (e.g., proper evacuation of passengers), technology, administrative controls and procedures (e.g., weighting the cargo before departure). However, although there are many layers between hazards and accidents, no defence layer is without error. These errors are visualized by the characteristic holes of Swiss cheese. Errors can either be active failures (human errors) or latent conditions that cause error-provoking circumstances in the local workplace (e.g., time pressure) or long-lasting weaknesses in defences (e.g., untrustworthy equipment). The mere presence of a hole does not necessarily lead to a bad outcome. Still, as soon as holes of different slices line up, there is an opportunity for accidents. The Swiss cheese model offers an alternative viewpoint on how disaster might happen. Moreover, looking at the MV Sewol case while acknowledging hindsight bias, may lead to rethinking about our first beliefs of foreseeability and the connected judgement of responsibility and guilt.  

To sum up, there often are many causes contributing to catastrophic events like the sinking of the MV Sewol. Although human error contributed to this tragedy it is necessary to acknowledge the danger of falling prey to hindsight bias when judging foreseeability and responsibility in retrospective. Additionally, to prevent similar events from happening in the future it is crucial to investigate the defence layers of systems and find error-provoking latent conditions. Decreasing human blaming can contribute to finding latent conditions by increasing transparent reporting of errors and “almost-errors”, necessary to establish a report-culture. 

References

Bogner, M. S. (2002). Stretching the search for the why of error: The systems approach. Journal of Clinical Engineering, 27, 110-115.

Hyungju, K., Stein, H., & Bouwer, U. I. (2016). Assessment of accident theories for major accidents focusing on the MV Sewol disaster: similarities, differences, and discussion for a combined approach. Safety Science, 82, 410–420. https://doi.org/10.1016/j.ssci.2015.10.009

Oh, N. (2017). Dimensions of strategic intervention for risk reduction and mitigation: a case study of the MV Sewol incident. Journal of Risk Research, 20(12), 1516–1533. https://doi.org/10.1080/13669877.2016.1179210 

Reason, J. (2000). Human error: Models and management. British Medical Journal, 320, 768-770.

Matarese, J. (2017, November 4th). The sinking of the MV Sewol [Audio podcast]