By David Oliver Kasdan
![]() |
There were factors in the context of this disaster unique to Itaewon that make it difficult to assign responsibility. While negligence and a lack of foresight had influence on the response, there is little evidence that any criminal oversight caused the crush.
This is not victim blaming or an excuse for what happened, but rather a reasoned plea for understanding that this event was the outcome of unpredictable behaviors in an uncertain context.
My research concerns socio-cultural behaviors and decision-making under conditions of risk (events with known probability) and uncertainty (events with unknown probability). The model of disaster management is a cycle that starts with mitigation and preparedness before the event, followed by response and recovery after the hazard strikes.
The hazard may be either natural, such as storms and earthquakes, or a human-made hazard that comes from the activities of society and industry. From that perspective, the deaths and injuries from the Itaewon crush are the result of human-made hazards, or as several officials have called it, a "social disaster," that involved both risk and uncertainty.
As we are now moving from the response to the recovery phase of the disaster management cycle, it is time to analyze the scant data that was available to learn how we may avoid future tragedies like this. President Yoon Suk-yeol admitted that Korea has not done enough research on crowd control.
The ministers and other officials have conceded that various contingencies were not considered. The police and dispatchers missed opportunities to react appropriately.
These faults are understandable as they involve a number of uncertainties, such as how people would behave in that particular context. People were in a good mood, perhaps a little drunk, and personal safety was not forefront in their minds as they were following the crowd to unknown attractions.
The normally high level of street safety in Korea allowed people to let their guard down and any worry of danger was put aside to see what was happening in that narrow alley where so many others had gathered.
There were also risks involved that would normally be contained. While the crowd size was larger than usual with estimates of about 130,000 people, that is not a statistically significant difference from the crowds of preceding Halloweens ― even before COVID-19 ― when Itaewon would host 80,000 to 100,000 people for the festival weekend.
There was increased police presence, but the focus was on specific tasks (illegal drugs and crime) rather directed crowd management. This was a calculated risk that, by all reasonable estimates, would take priority over the very small probability that a surge of dire magnitude might take place.
The social science behind the deployment of the police made sense at the time. Disaster management addresses the known risks first, then the uncertain threats. There is rarely the time, money, or foresight to consider the unknown uncertainties.
Again, this is not to excuse poor decision-making or to lay particular blame. Rather, this cold analysis is meant to convey that disasters happen because things get out of hand. There was never any intention for the crush to happen; it was the result of a confluence of circumstances that would defy logic.
We may compare the Itaewon crush to what happens when parents take their children to a playground. While mom and dad sit on a bench at the edge, the kids play on the slides and swings, all of them expecting that the environment is safe enough.
When the child falls off the side of the slide because there was a push to go first, the parents come running to their aid. The parents were not negligent. The child is not at fault.
The playground equipment manufacturer and property owner do not have liability. Sometimes these things happen and we may blame ourselves or others, but the more important thing is to not repeat the sequence of events that led to the accident.
The point of this analysis is to direct our concern toward the future of Korean safety by learning what we can from the facts. The people involved had every expectation that Saturday night would be an exciting, but relatively safe time.
With a little self-responsibility and caution, there was no reasonable cause for specific concern about the alley, the expected number of people, or how they would behave.
The response suffered from management and attention constraints; there is little doubt that we can improve on directing police and medical resources more efficiently. A broader approach to contingency planning ― thinking about the risks and uncertainties from multiple social perspectives ― can help to reduce, if not anticipate, unknown hazards.
The political attacks on pre-disaster responsibility are neither productive nor legitimate at this point. This was a horrible accident from unintentional influences; we should grieve for the losses and then watch out for the risky and uncertain hazards of society.
David Oliver Kasdan, Ph.D., is professor of public administration in the Graduate School of Governance at Sungkyunkwan University in Seoul. He researches disaster management and behavioral economics.