Search This Blog

Tuesday, October 23, 2012

Italian Earthquakes and "Responsible" Science

Italian earthquake destruction - from Ars Technica
So Ars Technica reports (as well as others) that several Italian scientists were convicted of manslaughter for not predicting an earthquake.

Now the Ars article talks about the superficial issues related to this but I think there is a lot more to think about.

The basic idea here is that the "government" contract with scientists to create a realistic assessment of danger - in this case an earthquake.  But it could be anything - a dam ready to burst, a faulty airplane part, etc.

No one, as far as I can tell, was asking for a specific time and date.  Instead there are the need for some kind of risk assessment in terms of "should people be evacuated."

This kind of thing happens all the time outside science - airplanes are grounded when "cracks" are found, for example.

Now if you are in a position as an expert I feel its your duty to ensure that, if danger or money is involved, you are as clear about risk as possible.

Its better to err on the side of caution.

So what about science?

Well, if I am in a lab running rats through a maze or spinning quantum quibits around in a vacuum chamber there is little danger, save for the apparatus exploding, of causing anyone harm.  If I publish reports of my findings, say that airplane wings made of spider silk test out well in my lab, its up to others to make realistic industrial tests to determine whether or not such an airplane wing is safe.

But there are other kinds of science for which things are, I think, less clear.

One is the "flu shot" model.

Now flu exists regardless of science and people will get it regardless.  No one knows what flu will strike in any given year so only an educated guess can be used to create a vaccine for the next winter's flu cycle.

Statistically it can be argued that even a bad guess as to what flu will strike might save lives even if the vaccine kills some number of recipients.

In this case victims (those that are harmed by the vaccine) have legal recourse against the maker if it harms them.

Like the earthquake model we have the "potential danger" model.

Here there are two flavors.

One is the "asteroid model" - scientists discover a large object that might impact earth but are unable to precisely calculate its path.  What do we do?

Here science has done "as much as it can" to alert us and has told us that precise knowledge is impossible to obtain.

My view is that these scientists are blameless whether the asteroid hits or not - they've done their best to make it clear that there is a potential danger of an unpredictable sort.

The second flavor is where scientists are asked to "rule" on some sort of public policy - either a specific danger as in the case of the earthquakes or on potential danger, e.g., climate change.

Here there is a specific "cost" associated with the "ruling."  In the case of the earthquake people will die depending on the ruling.  (If I rule there is danger people would flee and be saved.  If rule there is no danger people stay home and get killed.)

But the "cost" could be monetary instead of a life risk, e.g., what is the risk of building a plant near a fault line.

Again here industry and insurance play a role and the scientific advice is built into a larger "risk" model.

But the other side of this is something like "global climate change."

Here science claims there is a specific risk (and cost) associated with something.  Like the asteroid there is a claim that harm will come if "something is not done."

But there is a cost associated "doing something" and no guarantee that the risk is real.

Who bears this cost?

I think science must if it is science that puts out the risk.

For example, if "science" claims asteroid XYZ123 "will" hit the earth on December 21st, 2012 at 12:45 PM and it doesn't then the "science" must be held accountable.  If "science" claims a mere "risk" its different - others make decisions on the validity of the risk.

For too long many quasi-scientific groups from archeologists to climate scientists make claims about things based on "consensus" and not fact.

An asteroid has a trajectory which, if enough information is present, can be calculated precisely.  This is science.

If we can only guess at a trajectory (or whether CO2 is destroying the planet) this is "consensus" and not science.

But if we act as if a "consensus" is concrete science and we fail then I think science must be held accountable - which is what happened in Italy.

If I alone calculate a trajectory for an asteroid and shout "danger danger" I should be liable if the danger is not real.  Instead I should openly share my calculations.  If everyone computes the same results then it becomes a societal risk.

The reverse is also true.  If the "consensus" calculates a risk, not on mathematics but on agreement, and foists this off on society then those creating the "consensus" are responsible for it.

I guess it all comes down to objective facts and reproducibility.

An asteroid's path to hitting the earth is object and can be calculated if enough information is present.  Then the calculations are either right or wrong.  If they calculations are not certain then its not science but merely consensus.

In any case those creating the consensus should be responsible for the cost of their consensus right or wrong.

Which to me says that before you claim the "end of the world" you should be quite certain about it.

Today's science is mostly "consensus" building about things science collectively "feels are important" or are serious "social issues."

But going on TV and making claims is akin to shouting "fire" in a crowded theater.  However well intentioned the claim of "fire" might be there is still a cost to those in the theater if the one shouting "fire" is wrong.

Today's consensus science is given a free pass in this regard.

If we spend trillions of US dollars to stop global warming but it happens anyway will "science" step up and accept the responsibility?

These kinds of science are not "objective" and never can be - no one can know who made cave paintings or, in an open system like the earth's atmosphere, pin down objective specifics about why something is the way that it is.

As long as "'science" has no responsibility we cannot take it seriously.

No comments:

Post a Comment