Italian seismologists are to be jailed for six years for failing to predict the earthquake in l'Aquila.
'Actually you can never get it right,' says Robert van Gorcom, director of RIKILT. As an example he cites the mayor of New York's instruction to evacuate parts of the city in anticipation of tropical storm Sandy. 'There is a big chance that this sweeping precaution will prove unnecessary, just as there would be a fair chance of something happening if the mayor had done nothing.'
What considerations do you weigh up in your role as scientist?
'The main one is: have you, as a researcher, taken everything into account and have you pointed out the uncertainties? You have to be thorough when you give advice, because you cannot predict everything. Food that is 100 percent safe, for example, does not exist. That means that there will always be incidents involving food, like the recent case of salmonella in salmon. But it is still a difficult call to make. If you give nuanced advice with lots of ifs and buts, the client says: how does this help me? But if you give clear-cut advice that afterwards turns out to have been wrong, it seems you can end up in the clink.'
But should you as a researcher be held accountable if your advice is wrong? With food and hurricanes it can be a matter of life and death.
'The question is: did you do what you should? I think that you can hold researchers responsible for the quality of their risk analysis and output, which should be based on good models and reliable data. If the quality is not good enough, then you are liable.'