In what is setting an unbelievably bad precedent, several Italian scientists are on trial for manslaughter. Their “crime”? Failing to predict a 2009 earthquake in L’Aquila, Italy, which killed over 300 people. The scientists were on an Italian government panel whose task it was to assign potential seismic risks. After a series of very small earthquakes in L’Aquila, the scientists were asked for an assessment of the risks of a major earthquake. The scientists generally proclaimed the risk of a major earthquake to be very small. One week after the proclamation, a 6.9 earthquake hit and leveled the town.
Let’s get this straight…earthquake prediction just simply isn’t possible. We can assign probabilities for earthquakes. For example, my beloved U.S. Geological Survey estimates a 25-40% chance of a 6.0 or greater earthquake in the New Madrid fault zone (southeast Missouri, northeast Arkansas, etc.) over the next 50 years. They estimate a 7-10% chance of a 7.5 to 8.0 earthquake over the next 50 years. But in science, especially in such an incredibly inexact science such as earthquake prediction, uncertainty in those estimates must always be considered.
When scientists publish statistical estimates in peer-reviewed literature, they nearly always publish uncertainty figures for those estimates. I won’t even venture to guess what the uncertainties are for earthquake prediction, except to say that I’m sure they are incredibly large. Even for a general assignment of future earthquake risk like the New Madrid example above, I’m sure the uncertainty of that very rough estimate is incredibly high. That uncertainty increases significantly as you try to tighten estimates of when an earthquake may occur, or how strong it may be. In a case like the L’Aquila earthquake, similar earthquake swarms occur all the time without being followed by a major earthquake. The scientists involved made their best estimate on the risk of a major earthquake occurring, using the best available science and information.
Should they be put on trial for being wrong? Should it be a CRIMINAL OFFENSE for a scientist to be wrong? I map land-cover change and try to predict future land-cover change for a living. Should I be thrown in jail if a map I make of land-cover change isn’t 100% correct? Should my career and very life be ruined if I (GASP!) am WRONG?!?! SCIENCE IS BUILT ON SOMETIMES BEING WRONG, about theories being disproven, just as much as it is on theories being reinforced by empirical evidence. When we are proven wrong, we learn. We adapt our theories. We improve our state of scientific knowledge.
This is a unique case, and it’s extremely unlikely something similar would happen in the U.S. But it still sets an incredibly bad precedent, and not just from a legal standpoint. Putting a scientist on trial for being wrong is the same as putting the concept of science ITSELF on trial. Such an action completely misrepresents what science is about, something the general public already seems to have a difficult time comprehending.