Why the L’Aquila Case is NOT about Risk Communication
I’ve been trying to fit this into 140 characters for Twitter, but it doesn’t work.
Here is why I think it is appropriate to gloss the L’Aquila trial as being yes really about seismologists’ failure to predict the earthquake, even though the actual arguments in the case are all about whether or not the 6 seismologists and 1 civic official met their obligations to fully and effectively communicate what they knew about the seismic hazard in the region:
- The examples given for how the scientists’ bad communication led to citizen deaths are all of people who felt scared and might have left town or slept outside, but were persuaded to sleep in their own beds.
- The behavior the court claims constitutes criminal manslaughter is not a long-term pattern of cavalier risk assessment that persuaded people not to retrofit their houses. The crime at issue is negligent failure to communicate risk related to an earthquake swarm that turned out to be a precursor to the Big One.
I’m relying on English-language press reports of the trial here, rather than direct documentation of the trial itself – Stephen Hall’s writeup for Nature is a good one. If I’m wrong about either of the points above please correct me, because those are the bases for my conclusions here. Ditto if Italian law defines “manslaughter” in some quirky way such that “negligent miscommunication that led people to make bad decisions that foreseeably led to their deaths” is a bad summary of the crime.
If you are okay with the risk of sleeping in your bed in an earthquake-prone area like L’Aquila, the reasonable thing for you to do when there is a swarm of small earthquakes is to keep sleeping in your bed. There is likely a small increase in seismic risk after such swarms – we don’t know how small, because the science on this point is, to steal the judge’s phrase, “inexact, incomplete, and contradictory”. However, it’s general consensus that any added risk is small enough that we really can’t justify sending everyone out to sleep in tents every time we get a handful of small earthquakes. Indeed, when people do decide to sleep in tents because they’re scared of an earthquake prediction it is typically regarded as an education/communication failure (cf. New Madrid in 2020).
So here we have a case where scientists communicated in a particular way about short-term risk, and as a result, people did the right short-term thing. There was no scientifically reasonable short-term communication strategy seismologists could have pursued in L’Aquila that would have saved those people’s lives, because the only short-term action that would’ve saved everyone* would have been to evacuate, and evacuating was an irrational thing to do.
In other words: Six seismologists have been judged guilty of criminally negligent communication resulting in death, for communicating in a way that convinced people to do the scientifically reasonable thing under the circumstances. By definition, any improvements to the scientific communication would still have led people to the conclusion that there was probably no reason to sleep outside. Therefore, this case is not about a failure to communicate the science. It’s about the fact that science gave the wrong answer.
*Yes, a timely reminder of “drop, cover, hold” or the importance of bolting your furniture might have saved some lives without requiring years of engineering and construction work. If the stories making it to the press were about relatives who put the heavy mirror back up above their beds because they heard that the small earthquakes had eliminated all the hazard, I might have a more charitable opinion about this case. But there’s only so much you can do against a pile of unreinforced masonry so this would not have saved everyone.
Also, I’ve cheated a bit by not discussing the non-seismologist city official, who as far as I can tell was the only one of the seven who actually did say wrong things about the science. Saying wrong things is bad both on general principle and because it leads people to make the wrong long-term decisions about how to address seismic hazard… but as luck would have it his mistakes still pointed people in the direction of a scientifically sound short-term decision to sleep in their own beds on the night of the Big One. We can legitimately point to him as an example of terrible communication and we should still be outraged that he was convicted of manslaughter.
mastering astronomy | I’ve got your missing links right here (27 October 2020) | Not Exactly Rocket Science on 27 Oct 2020 at 10:45 am
[...] that the central issue is risk communication. But Maria Brumm makes a good argument for why the risk-communication & quake-prediction sides of l’Aquila aren’t separable. David Spiegelhalter also chips in on risk. [...]
Stuff we linked to on Twitter last week | Highly Allochthonous on 28 Oct 2020 at 2:51 pm
[...] On a similar theme, Maria Brumm explains why the L’Aquila case really isn’t about risk communication. /2020/10/22/why-the-laquila-case-is-not-about-risk-communication/ [...]