Earthquake science was not on trial in Italy — it was about inadequate information and participation in decision-making, says Carina Fearnley.
The prison sentences for manslaughter given to six Italian scientists and one government official last week have led to an outpouring of outrage. The media have echoed the anger of numerous scientists that earthquakes are impossible to predict.
But few have drawn attention to the real facts of the trial. It was not about the failure of science to predict earthquakes — it was about missing and incorrect information about the risk prior to the 6.3 magnitude earthquake that hit L'Aquila, in central Italy, on 6 April 2009.
Some residents did not leave their homes because they felt unable to make informed choices, despite evacuation being a traditional decision made by generations before. Consequently, 309 people died, some of whom were persuaded by the scientists' information to stay at home. 
In comments over the last week, David Ropeik from Scientific American said poor science communication was to blame, stating that scientists have a responsibility to share their expertise so people at risk can make informed choices.  Stephan Faris, at Time, said the scientists should have consulted statistical earthquake models to provide more probabilistic information to help people decide what action to take. 
But there are much more serious questions about this incident: who is the 'expert' who makes decisions about action, and what information should be communicated to the authorities, the public and the media?
Limits of forecasting
At a meeting held on 31 March 2009 at L'Aquila's government office, scientists gathered to discuss the recent earthquake swarms (several small earthquakes in a short period). It appears that the scientists did not rule out the possibility that a large earthquake could occur.
But there was little discussion about what to do ahead of a major event, despite the vulnerability of old buildings.
At a press conference, Bernardo De Bernardinis, then vice-director of the Department of Civil Protection, said the situation in L'Aquila posed "no danger".
"The scientific community continue to assure me that, to the contrary, it's a favourable situation because of the continuous discharge of energy," he said. 
Is this a case of the messenger (De Bernardinis) simply getting it wrong? Or did the messenger not understand the science and uncertainties? Could the scientist alongside him, Franco Barberi, have corrected this error?
This is where the underlying problems emerge. A gap in expertise and decision-making capacity is something I have discovered time and again in my research on early warning systems, particularly those for volcanoes.
Scientists such as volcanologists work tirelessly on developing knowledge of a specific hazard, attempting to find better monitoring methods and possible forecasting tools. Note the term 'forecasting', not prediction — the former usually refers to an approximation, while the latter implies a level of precision not possible for many natural hazards.
The fact is that without the ability to provide scientific certainty, a scientist can struggle to provide a best forecast. Forecasting then becomes a bit of a gamble — and chance is seldom science.
Context for decisions
And such forecasts are only part of the problem. Officials responsible for public safety also have to combine this uncertain hazard information — which they are not an expert in — with their own field of expertise, the social, economic, political and cultural factors specific to their at-risk populations.
So it is no surprise that these decision-makers lean heavily on scientists for recommendations on actions to take before they face the public.
What this means is that science is really only part of the problem; understanding the context in which a disaster may happen is equally important if good advice is to be given.
As earthquake forecasting is not possible, precautions are required — the most common is making sure all buildings in earthquake regions adhere to strict building codes.
In L'Aquila, few buildings were safe, so traditionally people took the precaution of evacuation. Understanding such historical context is vital for any scientist or decision-maker providing guidance on assessing the risks involved.
Understanding the uncertainties
Scientific uncertainty does not mean crises cannot be well managed. When Mount Pinatubo in the Philippines erupted in 1991, extensive education and outreach helped people make informed choices about action, and may have saved about 250,000 people.
In L'Aquila, emergency authorities working alongside scientists gave no formal advice on precautionary actions prior to the earthquake. Without this information, which clarifies the uncertainties, decision-makers cannot make informed decisions to prevent loss of life. Early warning then becomes a leap of faith.
So who is the best expert to give this advice — the scientists or the government officials? And what about local people and their expertise?
While this is a constant battle in the risk management practices of many developed nations, the developing world is taking a lead with new ways of working together.
Countries such as the Philippines and Indonesia already bring together scientists, experts on vulnerable populations, and local experts to make a more holistic risk assessment, and there is growing momentum to adopt this approach.   The management of the 2010 eruptions at Mount Merapi in Indonesia is another good case study.
But the decision-makers are not just scientists or officials; in part, each person makes their own decisions about warnings.
What the L'Aquila trial shows is that rather than pointing the finger of blame, we should be improving the knowledge and understanding of natural hazards, by encouraging community resilience and education — and always stating the uncertainties. Developing countries are showing that this approach can be successful.
Carina Fearnley is lecturer in environmental hazards at Aberystwyth University, United Kingdom, and an affiliate of the Aon Benfield UCL Hazard Centre at University College London. Carina can be contacted at firstname.lastname@example.org
 Hall, S. Scientists on trial: At fault? Nature 477, 264–269 (2011)
 Ropeik, D. The L'Aquila verdict: A judgment not against science, but against a failure of science communication. (Scientific American, 2012)
 Faris, S. The Aquila earthquake verdict: Where the guilt may really lie. (Time, 2012)
 Surano. Strategy of geological hazard mitigation in Indonesia. (Centre for Volcanology and Geological Hazard Mitigation, Indonesia)