When Science And The Law Completely Fail To Understand Each Other

Contrary to many press reports, the Italian scientists were not prosecuted for failing to predict the L’Aquila earthquake. Effectively, they were charged with manslaughter by negligent or reckless omission, for failing to adequately warn people of the risk of an earthquake and for failing to correct the assertion by bureaucrat Bernardo De Bernardinis, deputy technical head of the Civil Protection Agency that there was no danger. Now a court has found them all guilty and sentenced them each to six years in jail.

Fucking insane? Science on trial yet again in the country which condemned Galileo as a heretic? What could we expect from the justice system of a country hardly synonymous with honesty and competence?

It’s not that simple.

Firstly, this is not a case of a bunch of ignorant lawyers and politicians putting science on trial. The citizens of L’Aquila brought the complaint. Under the Italian constitution, prosecutors are bound to act on it.

The lack of a demonstrably direct connection between the reckless advice and negligence in not correcting it and the subsequent deaths makes the verdict appear unsafe. Six years is very harsh considering the multiple contributory factors in the deaths. However, De Bernardinis and the scientists would clearly have some case to answer in most Western countries, even if only a civil suit (which would probably bankrupt them).

Here is a chronological summary of events surrounding the L’Aquila earthquake:

Giampaolo Giuliani, a lab technician and amateur earthquake predictor used measurements of radon gas to predict a major earthquake in the region on March 29, 2009. Groups of people went around advising residents to leave their homes and a panic ensued. The earthquake did not occur.

Giuliani is not a mere crank. He has a science background and is seriously attempting to study radon emissions as a predictor of earthquakes. The problem is that he relies too heavily on radon emissions and appears overly confident of their strength as a predictor of actual quakes, as distinct from predicting general changes in seismic activity.

Although radon gas emissions have potential in searching for minerals, oil and geothermal energy sites, as well as being possible predictors of seismic activity, the problem is that high or even changing radon levels can have multiple causes. Being a predictor of increases in seismic activity is not enough: one needs to be able to predict seismic activity above the dangerous threshold and with an epicentre sufficiently close to the surface that the tremors will damage property and pose a threat to life. Radon gas emissions are currently a poor predictor of the timing and location of actual earthquakes, as distinct from predicting an increase in low level seismic activity.

The University of Massachusetts has a good geology course with a lecture on earthquake prediction. There are many observables which have weak predictive power, but science currently just doesn’t know enough to make accurate forecasts.

The next day, March 30, a magnitude 4.0 quake was observed in the surrounding Abruzzo region. That’s classified as a minor to light quake. There are tens of thousands of this magnitude observed worldwide each year.

The Richter scale is logarithmic. It is the logarithm of the amplitude of seismic waves measured along a single dimension (by the readout on a seismograph). Thus, an increase of one point on the Richter scale represents a tenfold increase in amplitude in the direction of measurement and consequently a 10^3/2 = 31.6 fold increase in energy output. The 6.3 magnitude L’Aquila quake on April 6 therefore released around 3000 times the energy of the 4.0 tremor on March 30.

On March 31, a meeting of the Major Risks Committee was held in L’Aquila. The Committee is an expert group which advises the Italian Civil Protection Agency on the risks of natural disasters. It consists of Bernardo De Bernardinis and the six scientists charged.

The minutes of this meeting record the following comments:

"A major earthquake in the area is unlikely but cannot be ruled out." - Enzo Boschi, president of the National Institute for Geophysics and Vulcanology.

"In recent times some recent earthquakes have been preceded by minor shocks days or weeks beforehand, but on the other hand many seismic swarms did not result in a major event." - Giulio Selvaggi, director of the National Earthquake Centre.

"Because L'Aquila is in a high-risk zone it is impossible to say with certainty that there will be no large earthquake." - Claudio Eva, professor of earth physics at the University of Genoa.

"There is no reason to believe that a swarm of minor events is a sure predictor of a major shock." - Franco Barberi, a volcanologist at the University of Roma Tre.

All the participants agreed that buildings in the area should be monitored urgently, to assess their capacity to sustain a major shock.

So what happened next?

Bernardo De Bernardinis and Franco Barberi fronted a press conference later that day, where De Bernardinis stated that:

"The scientific community tells us there is no danger, because there is an ongoing discharge of energy. The situation looks favourable".

As a consequence, people did not evacuate, including some who had been considering doing so. There was insufficient effort to identify particularly vulnerable buildings. On April 6, L’Aquila was hit by a 6.3 magnitude quake, killing 308 people.

On June 3, De Bernardinis and the six scientists were indicted for manslaughter, for failing to properly assess and communicate the risks, because people relied on their advice and subsequently died in the earthquake.

De Bernardinis was certainly going to find himself in trouble after spouting the above crap. It is clearly a complete misrepresentation of the comments minuted during the Risks Committee meeting.

Unfortunately, this kind of thing happens all the time when non-scientist bureaucrats, managers or “business analysts” attempt to produce an unholy fusion of communicating scientific findings and playing politics.

Many people had been panicked by Giampaolo Giuliani’s well meaning but incorrect prediction two days earlier. There was a political dimension to the press conference, since repeated scares with no ensuing quake would have resulted not just in public anger, but the possible ignoring of evacuation warnings when scientists had more data to support such a prediction.

De Bernardinis was clearly trying to juggle these demands with the inherent uncertainty of statistical analysis and particularly, earthquake prediction. The problem is that he is obviously a complete moron who does not understand scientific principles, as is usually the case when bureaucrats are attached to or worse, given oversight of scientific departments.

Clowns like De Bernardinis being placed in positions of (sometimes direct) authority over scientific departments or committees is a serious and common problem in both the corporate and public sectors. The bullshit they use to sell themselves is typically along the lines of: “My role is to interpret and implement the science in a way that’s practical for the business” or “that people can understand”.

The implication being made is that scientists are impractical, uncommercial boffins who can’t communicate with “ordinary people”. They might be good at hard sums, but it takes a special kind of professional (highly paid of course) who is comfortable in both camps and has the business acumen to implement the science in a practical way.

This self-marketing is a con job. In reality, subtleties which can materially affect decisions and outcomes are misunderstood and ignored as the science is misinterpreted and dumbed down for “the business” or government. When results are less than expected, including the occasional disaster, blame is of course sheeted onto the technical people, as it was in the GFC.

I have encountered many of these managerial types in the corporate sector. They abound in finance and IT. In my experience, the overwhelming majority are politically adept, but technically incompetent parasites whose survival strategy consists of a few simple tricks:

§        Cast yourself as “practical” and “business like” and the technical people as uncommercial and poor communicators.

§        Make regular lists of key statistics, brief task descriptions in project specs and topic headings in technical documents. Senior management or ministerial committee meetings rarely go into detail on individual topics. Asked for an update from your department? Easy, just read out the topic headings and your brief descriptions (all provided by someone else), then follow up with some stats and some general motherhood statements. You can even put them in a Power Point presentation (because everyone at the table is only capable of linear thought). Your technical people can create all the tables and graphs.

§        Learn buzz words. You only need a high level definition because you’ll only ever say them to “business” people or government ministers / staffers. Only drop buzz words sparingly and you’ll be seen as someone who can converse with the propeller heads, but is really “one of us”.

§        Take credit for others’ work. “My department delivers these results / products because I manage it for commercial outcomes.” Any overall project plans are of course your creation.

§        Take advantage of most people’s natural aversion to conflict. If any of the technical people show you up, paint them as “not a team player” and try to manage them out. Your superiors will help you. Corporate and government types want “team players” because committee decisions allow diffusion of responsibility. Someone who points out the Kool-Aid is actually poison is very threatening.

Apart from being a waste of time, space, money and oxygen, the more serious problem is that these idiots usually end up believing their own bullshit. Career success convinces them they CAN actually interpret and distil technical information and scientific results and communicate them, while at the same time balancing inconvenient uncertainties with political or business imperatives.

Of course, they can’t and so we end up with people like De Bernardinis telling everyone “there is no danger” or “risk officers” telling boards banks are sufficiently well capitalised.

The problem for the Italian scientists is that one of them, Franco Barberi, a volcanologist and then acting president of the Major Risks Committee, was present at the press conference and didn’t correct De Bernardinis. He was sitting right there while a man who is evidently a complete fool told everyone there was nothing to worry about. He knew this to be false, must have understood the possible consequences, yet failed to speak up and say: “No, hang on … that’s not true. There is a danger; we’re just not sure how much. Can I suggest …”

Unfortunately for scientists, engineers and technical people everywhere, by acting like a mushroom, Barberi has reinforced the unfair stereotype. Of course there exist technical people who are poor communicators and are hopelessly uncommercial. However, most are comfortable socially, capable of discussing their work and have a reasonable grasp of financial and political realities. That’s how they rose to senior academic or industry positions.

The other five scientists weren’t at the press conference, so there is some doubt as to whether they were aware of De Bernardinis’ statement. The prosecution’s position is that it was their duty to find out what was said and correct it if necessary. Since the reasoning behind judgements by Italian courts cannot be made public for three months after the verdict, we will have to wait to see what the judge thought of this argument, although the verdict and sentences give us a clue.

In many jurisdictions, inaction leading directly to death is viewed as tantamount to either murder or manslaughter where there is an undertaking to do something and to not do it would likely result in death. An example is if a heart surgeon walked out of a heart transplant in the middle of the operation and the patient died.

In New South Wales, Section 18 of the Crimes Act makes it clear that death as a result of the omission of action with “reckless indifference to human life” counts as murder or manslaughter.

This is what the prosecution is saying: You knew there was a risk, yet either said the opposite or effectively said it by failure to correct De Bernardinis' statement. As a result, people did not evacuate as they had planned. Some of them died. Your reckless statement or omission is taken to have caused their deaths.

I can’t see a problem with the element of recklessness. The problem is with cause.

By virtue of their positions, there was clearly an undertaking by the Risks Committee to provide advice as to the possibility of an earthquake strong enough to pose a threat to life. But that’s all they could have done: state that there was a risk, probability unknown.

Did neglecting to make this statement actually cause people to die?

The argument as to cause is: What would people have done differently had the scientists made the correct statement? How would have events differed? What is the likelihood the deaths and injuries would still have occurred?

The correct statement the Risks Committee should have made at the March 31 press conference is that there is some risk of a major quake, but we don’t know how much and for how long. They should then have given advice regarding identifying particularly at risk buildings and any mitigating actions which might be taken.

What would the people of L’Aquila have done had the correct statement been made and no earthquake occurred for five days? Would some have left town and then come back? Would people have avoided certain buildings? How long would they have waited?

What of the buildings which collapsed? Did they comply with building codes? Is that what really caused the deaths? Were some of the buildings which collapsed even known to not meet building standards? Would people have died had the buildings been properly constructed?

What of the older, heritage buildings which cannot be made to comply with modern standards? Would they have been evacuated? For how long?

Given the large number of uncertainties and unanswerable “ifs”, plus the very real contributory factor of building construction in at least some of the deaths, it is hard to see how a manslaughter charge based on reckless omission could be proven beyond reasonable doubt. I doubt even De Bernardinis’ foolish statement could have been proven to have led directly to deaths.

Had the truth been that there was a material probability of a large quake in say, the next week, then clearly defined evacuation measures could have been taken. This is what would occur in the case of a tsunami warning, for example.

However, the truth in this case was that the probability of a large quake was unknown and even if one did occur, its likely timing was also unknown.

Given this advice, people would soon have returned to their homes and lives, probably after only a few days, if they even left at all. Then they would have died.

It is not fair to hang a manslaughter charge on the scientists because they did not deliver a necessarily vague warning. Even the misrepresentations in De Bernardinis’ statement would have mattered little.

Being in positions of public trust, the scientists should have had a far better understanding of the political and legal issues surrounding their statements. They have done all scientists a great disservice, because what will be remembered is: “Scientists failed to communicate risks”. That De Bernardinis was the chief culprit and is not a scientist is probably not even known to many people and will be largely forgotten.

If scientists are able be prosecuted for “failing to communicate risks”, then lawyers and politicians (many of whom are lawyers) need to understand scientific principles. In particular, they need to clearly grasp the inherent uncertainty in many scientific fields and what science can and cannot reasonably say.