"What most people want is not knowledge, but certainty."
It's an apt quote of philosopher Bertrand Russell that Italy's Professor Massimiano Bucchi borrows to sum up people's expectation of what science should be able to tell us before earthquakes happen.
And this demand has only grown as science has advanced over recent decades, said Professor Bucchi, a professor of Science in Society at the University of Trento and a keynote speaker at this week's Science Communicators Association of New Zealand conference in Christchurch.
"If you look at recent policy documents and discussions, the only argument given for supporting and strengthening research is its practical use - technology, innovation, development," he told the Herald.
"In the long run, this dangerously raises expectations towards science and poses huge problems when they cannot be fulfilled."
Also, he said, scientific expertise no longer appeared any more in public as a unanimous, consensual voice.
Sir Peter Gluckman, the Prime Minister's chief science adviser, saw this happen in the aftermath of the Christchurch earthquake.
"In the early days, GNS had lots of different spokesmen after every quake - but once the community had a consistent spokesman, the story was more coherent, and I think that helped a lot."
Sir Peter was also concerned to see technical debates playing out on the front pages of newspapers.
He was preparing a report to be delivered to the Government this year, addressing issues around risk and miscommunication.
An extreme example of the gravity and danger that came with publicly forecasting the probability of earthquakes was seen in Italy last year, when six scientists and officials were convicted for downplaying the likelihood of a major shake six days before the 2009 L'Aquila earthquake struck and killed about 300 people.
Reacting to their convictions, GNS Science released a statement arguing the case had been about science communication, and not quake prediction.
"The main lesson from the Italian case is that a great deal of care must be taken by scientists in communicating what they know and don't know about the likelihood of future earthquake occurrences, both to public authorities and to the public," GNS principal scientist Dr David Rhoades told the Herald.
The expert committee were aware the risk in l'Aquila was much above normal because of the swarm that was going on at the time, but they also knew that, based on similar occurrences in past, the probability of an imminent disastrous earthquake was low.
But Dr Rhodes said the message given by one member of the committee emphasised only the second point and not the first, probably because they were trying to reassure the public and counter the predictions of controversial researcher Giampaolo Giuliani.
Victoria University's Dr Tim Stern said such forecasting was especially tricky in Western democracies.
"You can damage property values and create distrust by crying wolf too often."