Earthquake scientists guilty of manslaughter in Italy – experts respond

Overnight six Italian scientists and a government official were found guilty of manslaughter for underestimating the risks of a killer earthquake in the town of L’Aquila in 2009. All seven were members of the Major Risks Committee and were sentenced to six years in prison for failing to warn the population of the risks just days before L’Aquila and surrounding towns were hit by a quake which killed 309 people.

in September last year, several New Zealand scientists commented on the case and signed a letter in support of the scientists.

L’Aquila after the earthquake

Our colleagues at the Aus SMC collected the following expert commentary.

Professor Paul Somerville, is Deputy Director of Risk Frontiers Natural hazards Research Centre at Macquarie University. Risk Frontiers is an independent research centre sponsored by the insurance industry to aid better understanding and pricing of natural hazard risks in the Asia-Pacific region.

“On April 6, 2009, a magnitude 6.3 earthquake killed 308 people in the Italian city of L’Aquila, following an earthquake swarm that had produced earthquakes daily for four months. The case against the scientists was initially understood as relating to their failure to predict the earthquake (an error of omission), but is now focused on their having provided “incomplete, imprecise and contradictory information” about earthquake risk (an error of commission).

“Specifically, the local government’s prosecution argument is that the reassuring information the scientists provided at a meeting held one week before the earthquake, to the effect that a major earthquake was not imminent, inhibited the citizens from taking precautions that would have saved lives, especially as two large foreshocks occurred the day before the early morning mainshock.

“It appears that the government’s objective in holding the meeting was to debunk unreliable but alarming earthquake predictions that were being made by L’Aquila resident Giampaolo Giuliani, who is not a seismologist, and that the scientists were distracted in this direction instead of focusing on information about earthquake risk that the citizens needed. Further, it appears that the scientists found themselves answering questions about deterministic prediction of earthquakes (which they acknowledge is not currently possible) instead of probabilistic forecasting of earthquakes (which they can do).

“However, probabilistic forecasting has very low absolute probabilities, even when the increase in probability is high. For example, Italian seismologists estimated that the probability of a large earthquake in the next three days increased from 1 in 200,000 before the earthquake swarm began to 1 in 1,000 following the two large foreshocks of L’Aquila earthquake.

“Such low probabilities make it difficult for scientists to place a large degree of importance on their forecasts. Scientific errors made by members of the Commission exacerbated the situation. Dr De Bernardinis, who is an expert in floods, not earthquakes, incorrectly stated that the numerous earthquakes of the swarm were releasing stress and thereby inhibiting the occurrence of a larger earthquake. Dr Calvi, who is a structural engineer, misunderstood the seismologists as thinking that the earthquake swarm had no impact on the likelihood of a larger earthquake; in fact the probability was estimated by the scientists to be several hundred times higher, as we have seen.  Unfortunately, the members of the Commission who participated in the press conference after its meeting did not include any of its seismologists or earthquake engineers.

“As pointed out by Professor Thomas Jordan in a recent Science article (12 October 2012), the best way to avoid such problems in the future is to clearly delineate the role of the scientists and that of authorities responsible for civil protection. Experts should provide carefully constructed probabilistic statements regarding the risk, which decision-makers would then use to choose the best course of action.

” Ironically, the conviction of the scientists is likely to imperil the very need that this incident has highlighted: for open and clear communication between the scientists and the public. In a further irony, only now has the first action been taken against the engineers who designed modern buildings that collapsed and caused fatalities and no action has yet been taken against the government officials who were responsible for enforcing building code compliance. It has occurred to some observers that the local government officials may be scapegoating the scientists to avoid prosecution themselves.” (this commentary has been corrected from a previous version)

Dr Gary Gibson is a Seismology Principal Research Fellow at the University of Melbourne
“The experts in the L’Aquila case were not accused of failing to predict an earthquake. A prediction includes a region, a time span, and a magnitude range for an earthquake that will happen. They were not accused of failing to forecast an earthquake. A forecast gives a region, time span and magnitude range with an estimate of the probability that the earthquake will occur, perhaps 1%, 10% or 50%.

“They were accused of giving an overly reassuring picture of the risk facing the town, or of not providing an appropriate earthquake alert. An alert is given when something unusual is happening and it is possible that it may lead to a significant event, so it might be appropriate to take precautions. Shallow earthquakes often occur in swarms, as were being experienced in L’Aquila, but a large majority of such swarms do not lead to a moderate or large earthquake.

“If the average recurrence interval between large earthquakes in a region is something like 200 years, the probability of the next one happening in the next week is about 0.01% (or 0.0001). If an unusual swarm of small earthquakes is occurring, perhaps the probability of a large event may be something like 1% (or 0.01), or higher, or lower. When we learn more about shallow crustal earthquakes it may be possible to quantify this, and issue a forecast. In the meantime, an alert describing the unusual activity and its effect on the likelihood of a larger event is the best that can be expected.

“Earthquakes vary from place to place depending on the type and level of stress (compressive, tensional or shear), and their effects vary significantly depending on local geology and earthquake depth, so it is difficult to quantify any particular forecast. Even if their magnitude is moderate, shallow earthquakes can severely damage a relatively small region (e.g. L’Aquila, Christchurch, both with magnitudes only 6.1 to 6.3).

“In the majority of cases reassurance given by experts in similar circumstances would appear to have been vindicated when the swarm has dissipated with no major event. This has probably happened many times. Giving reassurance is a very human response, and sometimes it is not easy to balance reassurance with reality, especially an improbable reality.”

Professor Rick Sarre is from the School of Law at the University of South Australia

“This is very unusual path for the criminal law to tread, and would be unheard of here in Australia in the absence of evidence of gross negligence or reckless indifference by the accused scientists. To prosecute these Italian seismologists is one thing; to sentence them to significant terms of imprisonment is quite another. Indeed, the outcome is, quite frankly, bizarre.

“The public policy repercussions are significant too. From now on, one might safely predict that one of two things will happen: either Italian seismologists (or epidemiologists, or climatologists or any scientists for that matter) will remain quiet about their predictions unless they are absolutely certain (which will never be the case) or they will predict a worst case scenario, which will inevitably lead to resources being spent (and largely wasted) on anticipating things that will rarely, if ever, happen.

“No scientific prediction is without uncertainty. I think we can safely say that the law, however, has more certainty when it comes to both the mental element required to establish guilt, and the aims of punishment in the event of guilt. I am confident that the appeal judges will restore some degree of certainty to the law when the appeal comes before them.”

Wayne Peck is a senior seismologist in the Seismology Research Centre at Environmental Systems & Services

“Communication of earthquake hazard probabilities to the wider public is a complex issue. Following this verdict, seismologists and other natural hazards experts are will find it even more difficult to achieve a balance between, communicating what the most likely outcomes are, whilst acknowledging the chance of low probability, extreme outcomes. To err in one direction leaves them open to being charged with being “too reassuring” but to err in the other leaves them open to being accused of being alarmist, either way minor nuances in the language used can be interpreted differently by different audiences leaving the experts in a no win situation.”

Our colleagues at the UK Science Media Centre gathered the following reaction from UK-based scientists.

Dr David Rothery, Senior Lecturer in Earth Sciences, Open University, said:

“I am glad that those found guilty will appeal. Earthquakes are inherently unpredictable, and prior to the magnitude 6.3 l’Aquila quake a ‘best estimate’ was given, which was that the current low level seismicity was not likely to herald a bigger earthquake. That turns out to have been wrong, but most of the time it would probably have been right.

“Maybe the message came across rather too complacently, but six years in jail for at worse poor communication skills seems to me totally disproportionate.”

“If civil protection is to be effective, rather than having jail sentences hanging over them and having to battle the legal system, the Italian seismic experts should be spending time studying and advising on earthquakes.

“I am far less concerned about the scientists’ ‘failure’ to communicate the risks clearly enough, than I am about the likelihood that building codes for seismic resilience may have been flouted, with the result that buildings that should have withstood the shaking collapsed.”

Prof Bill McGuire, Professor of Geophysical & Climate Hazards, UCL, said:

“This is an extremely alarming verdict. If this sets a precedent then national governments will find it impossible to persuade any scientist to sit on a natural hazard risk evaluation panel. In the longer term, then, this decision will cost lives not save them.

“We don’t have the ability to predict earthquakes, but what national governments need to so is spend time and money ensuring that the buildings in areas of potential earthquake risk are able to withstand expected earthquakes.”

Professor Bruce Malamud, King’s College London said:

“The scientists involved in the trial conveyed to the public the uncertainty and small probability of an earthquake occurring in L’Aquila based on accepted knowledge we as a community have accumulated over many years. The words ‘improbable’ and ‘unlikely to occur’ are often unfortunate translations from scientists of these small probabilities, as they convey to some in the public that a large magnitude event will never occur, when the scientists are trying to convey that there IS a possibility, just small and finite. But, that any year, there is a given chance of an earthquake of a given size or larger occurring. It would certainly benefit society if instead of prosecuting individual scientists for a perceived failure to communicate, it rather works on educating the average citizen, through the schools and examples, as to what is meant by uncertainty and low probability.”

Professor David Spiegelhalter, University of Cambridge, said:

“This bizarre verdict will chill anyone who gives scientific advice, and I hope they are freed on appeal. The lesson for me is that scientific advisors must try and retain control over how their work is communicated, and are properly trained to engage with the public.”

Sandy Steacy, Professor of Earthquake Physics, University of Ulster, said:

“If it stands, this verdict will have a chilling effect on earthquake science in Italy and throughout Europe. For instance, who would now be willing to serve on an earthquake hazard evaluation panel when getting it wrong could mean a conviction for manslaughter?

“And what will be the effect on the “impact” agenda? Here in the UK scientists are being challenged to ensure that their research has influence outside academia; this case suggests that such engagement can be very dangerous.”

Dr Roger Musson, British Geological Survey, said:

“This is a very sad business indeed, these are people I know, who were doing their best to give an accurate account of large earthquakes. It seems to be wrong that they should be prosecuted for offering scientific advice to the best of their ability”

“It will certainly make scientists less free in speaking out where perhaps their expertise are really needed”

“It’s not about them failing to predict earthquakes, it’s not about anything those 6 scientists said at the forum which they were asked to give their opinion, what those 6 scientists said was correct and any seismologists would support it”

Richard Walters of Oxford University’s Department of Earth Sciences said:

“I am very saddened to hear about the verdict. The issue here is about miscommunication of science, and we should not be putting responsible scientists who gave measured, scientifically accurate information in prison. This sets a very dangerous precedent and I fear it will discourage other scientists from offering their advice on natural hazards and trying to help society in this way.

“I have read the translated minutes of the meeting of the Grand Commission of High Risks on the 31st March, and the scientific information that was conveyed within that meeting was not inexact, incomplete or contradictory. It was clear, measured and scientifically accurate.

‘The prosecution have not distinguished between the different defendant’s actions or words. To be prosecuted for other people’s miscommunication of your scientific advice is a travesty.’

Dr John Elliott of Oxford University’s Department of Earth Sciences said:

‘This verdict is a sad end to a tragic series of events in L’Aquila. Earthquakes cannot be predicted, and these scientists should not even have been on trial accused of providing incomplete information, because it is unfair to have expected them to have provided an exact and complete warning of an earthquake in the first place – this is something which is not yet credibly possible for earthquake science.

‘This potentially sets back scientists’ desire and ability to engage openly with the public and authorities on the risks faced by society from natural hazards, particularly those involving seismic activity.’

Dr David Rothery, Senior Lecturer in Earth Sciences, Open University, said:

“I hope they will appeal. Earthquakes are inherently unpredictable. The best estimate at the time was that the low level seismicity was not likely to herald a bigger quake, but there are no certainties in this game.”

Prof Malcolm Sperrin, Director of Medical Physics, Royal Berkshire Hospital, Reading, said:

“Assuming that negligence and malpractice are not factors here then the prosecution, and now sentences, of the Italian seismologists comes as a considerable surprise. In seismology, as with many other branches of the pure and applied sciences, opinions are derived from observables and the application of experience and training. It is never the case that predictions are completely without uncertainty and any scientist will make this clear as well as an estimation of how accurate such predictions are.

“If the scientific community is to be penalised for making predictions that turn out to be incorrect, or for not accurately predicting an event that subsequently occurs, then scientific endeavour will be restricted to certainties only and the benefits that are associated with findings from medicine to physics will be stalled. It is worth pointing out that many of the valuable contributions made by scientists such as penicillin, radiobiology etc have stemmed from the enquiring mind rather than absolute certainty of success.”