"Our greatest responsibility is to be good ancestors."

-Jonas Salk

Thursday, September 1, 2011

Burden of Proof

One delayer tactic is to demand "proof" of "global warming" before advocating "hugely expensive actions". (*)

This comes down to "burden of proof" arguments. Much though we might wish people understood Bayesian reasoning better, it appears that people are primed to a sort of naive absolutism by the legal system:
I was mugged once some years ago and was called in for a line up. It was night, the mugger was wearing a hoodie and for most of the time during the mugging there was a street light behind him. When I went in to the line up, I said I was 90% sure it was number 5. The officer questioned my 90% saying "Are you sure you're just 90% sure? We know it's the guy". I stuck to my 90%. I simply wasn't 100% sure and explained why (night, hoodie, street light).

After doing the line up, I was taken into an office with a detective to make a statement. When I got to the 90% sure thing, the detective started to get pretty agitated.

So, he stops the tape recorder and starts grilling me about my insistence on not saying I was sure it was the mugger. He got really angry and then said, "Don't give me that bullshit! We know it's the guy. Just say it's him, damnit!". He turns the tape recorder on and again I say I was 90% sure. So, he gets visibly angry, turns off the tape recorder and storms out of the office telling me I can get the hell out of there.
Where did this idea of "scientific proof" even come from? Anyway, this expectation of waiting for "proof", convenient for the fossil fuel interests, seems tied into some logical absurdities in the legal system as well.

( * ) - though they never demand "proof" that the actions would be hugely expensive. They certainly would be hugely expensive to the fossil fuel interests, after all!


7 comments:

manuel moe g said...

Also folk risk analysis has a basic asymmetry of what is considered an intervention.

Imagine mankind had to reapply for permission to release billions of tons of carbon emissions every year, contingent on "scientific proof" that the added tons could be harmlessly absorbed and the current and past emissions were consistent with ecological stewardship for the benefit of future generations.

In this phrasing, the intervention is the yearly release of billions of tons of carbon emissions. Folk risk analysis *always* makes the intervention the act of limiting emissions.

Marion Delgado said...

I shudder for the day the denialists discover Bayesian/Frequentist disputes. Think what hay they could make in public. "Scientists don't even agree on how they know anything! Statistics are a lie!" "If you analyze this hypothesis with MLE ..."

BTW my big deal now is charity towards public explainers, on every axis. I think oversimplification is our friend.

rab said...

"naive absolutism" reminds me of Feynmann's account of investigating the shuttle disaster. He asked the engineer what was the probability of success of a mission, and engineer said "100%". When Feynmann looked shocked, he said, "Umm..., minus epsilon".

Anna Haynes said...

"Folk risk analysis" - I like that term. Thanks Moe.

Steve Bloom said...

Then there's the related problem of the public perceptions engendered by the lengthy scientific process of coming to firm conclusions, noting the well-known sociological phenomenon of refutations of incorrect first impressions tending to reinforce rather than correct them. The field of climate science (and indeed society as a whole) is now paying the price for scientists having been so open about the uncertainties for so long. In light of this, it's hard to feel very generous toward the likes of Curry.

rab said...

I agree with Bloom on Curry. As the general public understands only in black and white, scientists should come out and say with certainty that there are catastrophic consequences if we do not drastically reduce our use of fossil fuels. Taking the 2% risk that they are wrong is simply the correct and courageous thing to do.

susan said...

There is no certainty in life, only in death. And that's where we seem to be headed. How difficult to make it clear.
--
and to dilute the above, which I think might be important, but ... attacked in Cambridge, couldn't help, only guy that looked right was the policeman! Too bad, but they were very nice to me (white girl, of course). I screamed and screamed and the guy made tracks, but nobody at all came out, just moseyed on home. Made me think hard about observation!