Thursday, December 11, 2014

The Perils of Earthquake Prediction and Political Hubris


Six seismologists were finally acquitted last month after they made rash comments before Aquila earthquake of 2009.  They were victims of unreasonable expectations and not scientific ignorance. 
At 3am on 6 April 2009, an earthquake measuring 6.2 on the Richter scale devastated the medieval city of L’Aquila in the Apennine Mountains of central Italy.  Over three hundred people died and the city’s cultural treasures were left in a parlous state.  But it is events that unfolded shortly before the quake that have continued to attract worldwide attention. 

Six days before the disaster, a government committee of six seismologists and a public official tried to dampen down fears that an earthquake was imminent.  In particular, the one member of the committee who was not a scientist, Bernardo De Bernardinis, stated that there was “no danger”.  In 2012, a local court convicted the committee members of involuntary manslaughter.  When they were first charged, numerous professors, decorated with the weightiest of credentials, wrote letters attacking the prosecutors.  Putting these men on trial was an affront to the dignity of science, they cried.  When the seven were found guilty, the cacophony of outrage doubled in volume.  The Aquila seven joined Galileo as paradigms of scientific martyrdom. 
The wheels of Italian justice turn extremely slowly and only now have appeals against the decision of the local court been handed down.  The six scientists have had their convictions quashed, but that of De Bernardinis, who said there was no danger, was upheld.  Further appeals are still possible.

So what was really going on?  The world’s media misreported the 2012 trial with an even greater level on ineptitude than usual.  No prosecutor had alleged that failing to predict the earthquake was a criminal offence.  This was because predicting an earthquake is impossible.  The record of failure is long and inglorious.  We’ve only recently found out why earthquakes happen at all.  Aristotle thought they were a result of vapours escaping from the soil.  In the eighteenth century, some theorists blamed lightning strikes.  The development of plate tectonics in the early-twentieth century means that we now understand what causes the ground to shake. But mainstream geologists long derided this theory and it only achieved widespread acceptance in the 1950s.
Prediction remains a pipedream.  Studies of animals have found that, while they can act strangely before a major quake, plenty of more innocent occasions set off the same behaviour.  A retired engineer claimed that an earthquake was looming at L’Aquila because he was picking up higher readings of a radioactive gas.  But again, this also happens when no earthquake is due.  Foreshocks, such as those felt at L’Aquila, occur before about half of large quakes.  In contrast, large quakes only follow foreshocks about one occasion in fifty.  Thus, major seismic events do give some warning signs.  It’s just that those warnings don’t usually presage a serious earthquake.  In the jargon, “false positives” are far more common than true predictors.  Just imagine if scientists demanded the evacuation of Los Angeles, promising the big one was around the corner, and then nothing happened.  That is the most likely outcome given the current state of knowledge.

So, the seven Italians were not convicted of failing to predict the disaster in L’Aquila.  Rather they were accused of going about their duties negligently.  And negligence that causes death is often characterised as manslaughter.  Given De Benardinis assured the public they were completely safe, the failure of his appeal seems fair.  But the seismologists were in the impossible position of not knowing what the risk was, just that it probably wasn’t very great.

One question the case raises is the extent to which scientists should be held accountable.  The implication of many of the L’Aquila seven’s defenders is that scientists should be given carte blanche to say what they like.  Anything else would obstruct free enquiry.  But that can’t be right.  A scientist who carried out their work without due care or made off-the-cuff pronouncements would surely be culpable.  Given that we cannot predict earthquakes, a confident statement that no earthquake was due would be as bad as saying that it was imminent.  As in many other fields, an honest mistake is a defence, but negligence is not.  Scientists enjoy the status of latter-day sages.  Many imagine that their methods provide the only road to truth, not only in physics and biology, but in the social sciences as well.  So perhaps the message from L’Aquila is that in making those claims, scientists unintentionally erect expectations that they cannot possibly meet.
Still, there is another way of looking at the case of L’Aquila.  The government set up a committee to advise on earthquakes and people of the city felt betrayed when it failed to protect them.  No matter that the state could no more control the ground than Canute could the tide.  Like so many westerners, the citizens of L’Aquila thought that their government was an indomitable Leviathan.  The media fuels this mood with its constant refrain that “something must be done” even when there is patently nothing that can be.  So grandiose has the rhetoric of the state become, that people imagine the reality should match the words.  Even in a country like Italy where the government is so self-evidently incompetent, it is still expected to be in control of events that are intrinsically beyond control.  Extending the scope of a government’s tasks to scientifically impossible tasks such as earthquake prediction is only a small step further from expecting it to achieve the economically impossible by “kick-starting the economy” or the mathematically illogical task of preserving generous entitlements without raising taxes or cutting other spending. 

We have come to expect too much from our politicians and our scientists.  What it needed is for both professions to become more humble. Otherwise, they can expect to be severely punished if they can’t live up to their rhetoric.
Discuss this post at the Quodlibeta Forum