Member log in

Should scientists be criminally liable for poor quake communication?


The convictions of the Italian scientists of manslaughter has more relevance here in New Zealand than you might think.

In the wake of the Christchurch earthquakes, the scientific community and GNS Science in particular came in for scrutiny over how effectively it communicated the risk of earthquake activity in the lead up to September 2010, prior to the devastating February 2011 quake – and post quake as well.

While there was much apparent frustration on the part of the public that the seismologists hadn’t been able to “predict” the quakes – and Sciblogs has already conclusively established that no one, least not Ken Ring can perform that trick – much more subjective and open to interpretation is how warnings around risk were communicated to the public. Overall, I think most people would agree that GNS Science did a damn good job of telling the public what it knew – and didn’t know.

And as the statement below, released by GNS Science this afternoon suggests, the L’Aquila situation involved risk communication gone wrong. Nevertheless, should scientists be criminally liable if they misrepresent risk through poor communication?

The worst thing that could happen is scientists watering down their comments and hedging their bets to the extent that information is of little use to the public either way – neither reassuring, alarming or enlightening.

Peter Griffin is the head of the Royal Society's Science Media Centre, home of Sciblogs, among other resources.

In 2009, a quake the size of Canterbury's February 2011 shake hit the city of L'Aquila, Italy, killing 309 people. Six days before the quake, six scientists had reassured people there would not be a major quake, despite tremors. 

The six scientists were sentenced to six years' prison for multiple manslaughter.

Scientists around the world condemned the sentence, although Northland-based New Zealand seismologist Dr Chris Buckley backed the sentence. 

Dr Buckley told RNZ's Checkpoint that the scientists were wrong to think they could draw definitive: "To my mind they made a big mistake claiming they could use the information available to say there would be no quake," he said. He did not believe the verdict would influence the behaviour of NZ scientists.

RAW DATA: GNS statement

Oct 23, 2012

L’Aquila convictions about science communication, not about quake prediction

The manslaughter conviction of six scientists and a government official in the wake of the magnitude 6.3 L’Aquila earthquake in Italy in 2009 is a complex matter involving legal, scientific, emotional and political aspects. It is also concerned with a very specific set of circumstances.

We understand that the court case was not about failing to predict an earthquake. Most people understand this is not possible with current scientific knowledge. There are no proven precursory signs such as gas measurement, micro-earthquakes, animal behaviour, electrical phenomena, or lunar phenomena that can predict earthquakes.

Despite decades of research into earthquake processes, the ability to predict earthquakes remains elusive.

The Italian case is really about the ineffective communication of science. In this instance, the scientists and government official were found to be deficient in the way they communicated the state of scientific knowledge and the possible threat of a large damaging earthquake.

The communication of risk and uncertainty is a challenging area for scientists. But to suggest that repeated small earthquakes in the area of L’Aquila were favourable because they unloaded seismic stress and reduced the chance of a big quake was unwise in our view. This, and other comments from officials, apparently inhibited many people from taking actions that might have saved their lives.

Equally, the L’Aquila area had a known history of earthquake activity and government officials could arguably have done more to prepare city infrastructure and the population for a large earthquake through measures such as setting appropriate building standards.

It is difficult to make any direct comparisons between L’Aquila and what happens in New Zealand. The roles and responsibilities of scientists and government officials are different in the two countries. However, the case does provide lessons about the communication of science and earthquake risks to officials and the public.

The most scientists can do is to estimate the probability of an earthquake occurring in a given region over a certain time frame such as months, a year, or longer. However, because natural events are inherently unpredictable, the limitations on the meaning of these probabilities need to be communicated clearly to the public.

GNS Science endorses the need for scientists to communicate meaningful information about natural hazards and probabilistic information to government agencies and the public. In this regard, for example, we update our aftershock probabilities for the Canterbury region on a monthly basis.

In relation to the Canterbury earthquake sequence, over the past two years GNS Science has undertaken hundreds of communications with a wide range of stakeholders via public seminars, briefings to government agencies, written reports, video and Youtube clips, plus many communications with the print and electronic media.

It is worth noting that in the past 60 years in Italy, only six of 26 major earthquakes have been preceded by foreshocks and many earthquake swarms have occurred without subsequent large earthquakes.

As foreshocks are usually not any different to ‘background’ earthquake activity, it is impossible to make a diagnosis that they are precursors to a major earthquake. Worldwide, most major earthquakes do not have precursory foreshocks.

Scientists must weigh up the evidence carefully and be cautious about the possibility of saying too little and delivering a false sense of security that could cause complacency, or delivering a false alarm that could cause panic.

There is a need for balanced information so government agencies and the public have the ability to make informed decisions about their actions.

Part of GNS Science’s core purpose, established by the Government, is to increase New Zealand’s resilience to natural hazards and reduce risks from these hazards. As its role requires, GNS Science will continue to communicate measured and meaningful information about natural hazards to government agencies and to the public.

More by Peter Griffin

Comments and questions

there are no known person on earth how can tell when an earth quake will hit any town or county, or till the time they will come to whom the bell tolls.

And so they should be held responsible for their predictions - just as the Metoffice should be for its forecasts! Even moreso because I rely on them every day rather than the 1 in 200 year earthquake. There are laws against this, on the books for 100s of years - it's called soothsaying. Burn them at the stake I say!

What nonsense - but then it was done in Italy where they are superstitious beyond belief.

They are not being jailed for failing to predict, the case was about did they accurately present the uncertainity. According to the minutes of the meeting they told officials it was extremely unlikely an earthquake like previous one in 1706 would happen. This was beyond the limits of what science can predict and they should not have been do confident.

Should they go to jail? I don't know because there will be more to the story than the headlines. But let's stop saying they were jailed for failing to predict they were jailed for over confidence.

How stupid has the world become. This is the mentality of the dark ages.

No scientist will ever offer any opinion based upon their (or any) knowledge in Italy any longer.

A scientific opinion is worth no more than the evidence on which it is based. Just report the evidence and forget the opinions.

How is this a good idea? Presenting evidence without any kind of conclusion is ridiculous. The general public don't know how to interpret raw data - there would be many who jump to extreme conclusions and cause unnecessary panic.

This case was never about the power of prediction - it was about what was interpreted to be an inadequate characterisation of the risks; of being misleadingly reassuring about the dangers that faced their city.

They should have told the truth, that is, that they and every other seismologist in the world don't have the faintest idea of when or where the next earthquake will be, but they couldn't resist the temptation to grandstand and give an opinion, and that was their mistake.

I agree. They made a statement that was unscientific nonsense and paid the price. Had they stuck to the science they would not be going to jail.

The most common cause of death in earthquakes is that things hit and fall on top of people. it was clear that the people lived in an earthquake zone (most of Italy is) and were obviously ill prepared for a risk that they or the civic authorities were aware of. Maybe the answers they gave were not couched in caveats customary for financiers and politicians, but were they aware of just how poorly the community had been prepared for such an event, the real cause of the loss of life?

Wait till we can lock them up over the carbon & climate scam.
Happy days !

While we are on the topic of scientists being liable for biased scientific communication of risks, should NIWA reconsider their communications surrounding "catastrophic" outcomes from climate models? Since knowingly sabotaging the NZ economy is now classified as economic terrorism.

One word - Berlusconi. A nice distraction from the trial, don't you think?

Are GNS libel for all the information they provide the govt & public with to create a new "seismic building code" ?
This earthquake strengthening industry is growing without supporting evidence for released figures and claims like "34% earthquake strengthened.".
Like the ETS and Carbon trading its another scam fed by misinformation and fear.
Stephen it was not about providing the govt with knowledge it was about providing them with misinformation.
They should not be charged with manslaughter, but I do not think GNS should continue defending the provision of misinformation.
Sometime if you do not know it is good to say "I do not know" the scientist would agree.

Should bureaucrats be liable for their ideological convictions imposed on a non-consenting population. Think rural urban boundary dense cities causing house prices to be extraordinarily high.

Peter, you wrote: "But to suggest that repeated small earthquakes in the area of L’Aquila were favourable because they unloaded seismic stress and reduced the chance of a big quake was unwise in our view."

"Unwise" doesn't capture it for me. The "unloaded seismic stress" statement was untrue, or at best, unproven. So they lied (or failed to rebut the lie of their fellows) in the service of reassuring frightened people. Scary and tragic as the earthquake was, don't lose sight of the main lesson learned by the people of l'Aquila: you can't trust those patronizing pat-you-on-the-head scientists who treat you like children having a nightmare, instead of adults able to cope with a frightening situation.

Probabilistic forecasts are the way to go - any non-zero probability of disaster covers your liability. In meteorology, ensembles of computer models now enable us to assess (a vague term) the likelihood of extreme events, including outcomes from climate models. Those issuing such forecasts will also need to communicate the uncertainty of their probability forecasts!