Thursday, September 16, 2010

categorisation of uncertainty

Interesting essay and comment thread about Doubt on climate scientist Judith Curry's new blog.

She proposes an Italian flag model inviting participants to cast their percentages wrt focus questions or hypotheses on their opinions about evidence for (green), evidence against (red) and uncertainty / unknowns (white):

She outlines a couple of hypotheses, one from the IPCC and one of her own about climate in the 21st Century and provides her own percentage estimates in the different categories for these. Her position is even handed in attributing climate change equally to human and natural factors with the uncertainty (white area) being dominant in her judgment.

Her main aim is to emphasise the uncertainty of much of the science in contrast to the name calling between "alarmists" and "deniers" that spoils much of the debate.

Roger Pielke snr left a comment there drawing attention to his hypotheses 1, 2a and 2b, which is an alternative way forward for the discussion. I have blogged about the Pielke snr hypotheses earlier

One of the critical comments, from Dan Hughes, concluded:
I don’t get the litmus-test question or the Italian-flag approach. I think we need to get a much better handle on many of the critically important aspects. While a top-down problem can be specified based on very little truly understood information, and even mis-information, and even not-applicable information, successful solutions are always, and I do mean always, based on extremely well-understood bottom-up information
I think that's correct from a scientific point of view but Judith is making an attempt to at least dampen down an acrimonious discussion which has got out of hand.

I left a comment there drawing attention to other ignorance categorisations, which I have blogged about in the past:
Five Orders of Ignorance
we don't know what we don't know

Actually, my main thought wrt the Italian flag categories was that it could be applied to other areas outside of climate science, such as political economy. In response to this focus question:
Will the current economic crisis develop into something akin to the Great Depression?
My guesses on this one are:
Green (evidence for): 20 percent
White (unknowns, uncertainties) 60 percent
Red (evidence against) 20 percent

I say guesses because the distinction b/w evidence and belief is not clear to me. Evidence is evidence that you believe in.

No comments: