Dying in Earthquakes: A Tale of Umpteen Tremors

Earthquakes don't kill people, but collapsing infrastructure does.

Posted Aug 10, 2020

How many people have earthquakes as their cause of death? The phrasing sounds awkward, but it is deliberate in avoiding the simpler question "How many people die in earthquakes?"

 Ilan Kelman
Engineering for earthquakes in northern Japan, although many seismic resistance measures are not visible.
Source: Ilan Kelman

The reason is to focus on the "cause of death." Mortality in earthquakes has been devastatingly high: over 200,000 each in China in 1976 and Haiti in 2010. Farther back in history, perhaps over 800,000 dead in China in 1556; over 200,000 in Syria in 1138; two ninth century earthquakes in Iran each with six-figure death tolls; and similar tallies twice in Turkey in earlier centuries.

How did these people die? It was not really the shaking which killed them, but mainly collapsing infrastructure. We have had earthquakes since the dawn of humanity and many traditional design and construction techniques deal with seismicity. At what point do we realise that we could have and should have stopped collapses during shaking?

This is especially true today when we add in modern earthquake engineering alongside the understanding of how some populations are left without opportunities to help themselves coupled with the science of changing policy and practice to live with earthquakes without earthquake disasters. It might be understandable that experiences in Turkey and Iran from over a millennium ago did not reach China in time to stop the 1556 tragedy. Even knowledge from that earthquake or Austria's major tremor in 1590 would not be easy to transfer down the centuries to stop over 100,000 dying in Japan in 1923 and in Turkmenistan in 1948.

Now, we do have the knowledge, ability to access it, extensive interest in acting on it, and resources to do so. Past decades have shown that, when this does not happen, it is typically an active political decision not to move forward with reducing earthquake vulnerability which leads to infrastructure collapses and deaths in earthquakes, often most affecting those with the fewest choices. Reasons range from corruption to discrimination to apathy to other priorities to inequity to lack of caring.

The cause of earthquake deaths is then corruption, discrimination, apathy, other priorities, inequity, and lack of caring. That is, the cause of earthquake deaths is not the earthquake. To answer the question at the beginning, in modern times, effectively zero people have earthquakes as their cause of death.

An earthquake unfolds quickly as an event bounded in space and time. Before the earthquake struck, it took a long time through interconnected, amorphous societal processes for the urban planning, marginalisation, building codes, inequities, injustices, and construction to manifest so that infrastructure collapsed and killed people. Frequently, people had little choice except to live in and use these buildings. These baseline causes of earthquake disasters (rather than just earthquakes) result from humanity’s long-term decisions, attitudes, behaviour, activities, and values.

The year 2003 brought a trio of seismic events showing the contrast. 26 September, early morning, Hokkaido island in northern Japan: a massive, shallow earthquake with moment magnitude 8.3 at 27 kilometres below ground. Slightly over an hour later, the same area was rocked by a huge aftershock.

Hundreds were injured as the ground moved, but no one died right away. Reports afterwards described how someone cleaning up glass was struck by a car and died.

Three months later, 22 December, late morning, central California: very shallow at 8 kilometres deep, but moderate at moment magnitude 6.5. Just as in Japan, plenty of injuries and a lot to clean up. Two people were killed as they left their building and a tower collapsed on them. The property owner was found liable for failing to keep the structure up to seismic resistance standards.

Four days after California's earthquake, dawn over southern Iran brought a similar one: 10 kilometres deep and moment magnitude 6.7. This time, the death toll exceeded 25,000. In the World Heritage city of Bam, mud-brick houses crumbled, with the dust suffocating occupants or heavy roofs crushing them.

Japan, the USA, and Iran have long had some of the best seismologists and earthquake engineers in the world. In two of these three countries in 2003, policies and decisions had applied the expertise over the long-term to save lives in earthquakes. It can be done, yet too often it is not.

Many other comparisons provide similar conclusions. 2010 witnessed earthquakes in Haiti, Chile, and New Zealand followed by shakings in New Zealand and Japan the year after. The widely varying death tolls were caused by long-term choices to stop earthquake damage, or not, rather than from the actual tremors.

We hear so much about earthquakes being dangerous, but it is really us being dangerous through choices which fail to stop infrastructure destruction. We know exactly what to do, technically and socially, to stop buildings from falling down in earthquakes. Long-term, societal choices by those with power and resources mean that some places are made to be safer than others, irrespective of earthquakes.

The maxim has long been this: Earthquakes don't kill people; collapsing infrastructure does.


Alexander, D.E. 2016. "The game changes: 'Disaster Prevention and Management' after a quarter of a century." Disaster Prevention and Management, vol. 25, no. 1, pp. 2-10.

Bundell, D.J. 1977. "Living with earthquakes." Disasters, vol. 1, no. 1, pp. 41-46.

Lewis, J. 2003. "Housing Construction in Earthquake-Prone Places: Perspectives, priorities and projections for development." Australian Journal of Disaster Management, vol. 18, no. 35-44.

Mika, K. 2019. Disasters, vulnerability, and narratives: Writing Haiti’s futures. Routledge, Abingdon, U.K.