Saturday
Jan152011
by 
Bishop Hill
Bishop Hill Uncertainty is good
Jan 15, 2011 Climate: Statistics That seems to be the message from an opinion-taking exercise run by the Edge magazine. The respondents are a group of leading scientists, philosophers and others. The results are discussed in the Guardian.
Being comfortable with uncertainty, knowing the limits of what science can tell us, and understanding the worth of failure are all valuable tools that would improve people's lives, according to some of the world's leading thinkers.
Reader Comments (49)
Luckily, when we run into the 'limits of what science can tell us', we can resort to a combination of post-normal science and our own agendas to guide us.
I'm not certain about this. It worries me.
Alexander is right to be uncertain, and right to worry. The article quotes the physicist Lawrence Krauss of Arizona State University as saying: "In the public parlance, uncertainty is a bad thing, implying a lack of rigour and predictability. The fact that global warming estimates are uncertain, for example, has been used by many to argue against any action at the present time." and Neil Gershenfeld, director of the Massachusetts Institute of Technology's Centre for Bits and Atoms wants everyone to know that "truth" is just a model. "The most common misunderstanding about science is that scientists seek and find truth. They don't – they make and test models," he said.
The Guardian files the article under “science - mathematics”, but its hard to believe the sub-editor didn’t have another science in mind when he coined the headline: “We must learn to love uncertainty and failure, say leading thinkers”
The proper recognition of uncertainty in theories and observations is readily overlooked - see for example the history of error bars around major constants in physics such as the speed of light, where the currently accepted 'best' value is outside some of the limits judged by earlier investigators to span the truth. Computer models provide extra risk of false assurance since they are essentially deterministic, and require deliberate effort to come up with plausible uncertainties about their outputs. I suspect when the dust has settled on the shameful shenanigans built upon excessive confidence in computer models set up to illustrate some aspects of some theories of the role of CO2 in the atmospere and of possible feedback loops added to get extra spice in those outputs, much of the opprobrium will be linked to this remarkable 'confidence'. It led many players to overlook their basic responsibilities as professional adults to assess and report wisely on possible dangers and threats. Instead we have been subject to a spectacle of what I must assume to be weak personalities like Schneider, evangelical ones such as Houghton and Hansen (in their different ways, one religious, one political), demonstrating the harm of shouting 'Fire!' in a crowded theatre. Wiser heads have looked for the fire, in all the obvious places, and found no cause for alarm. But their voices have been drowned out by a very vocal section of the audience pretending to be hysterical, a section comprised of many environmentalists, eminences-grises such as Strong, and the local mini-Strong in the UK, Napier of the Met, and highly visible eminences such as Gore in the States.
So in a sea of troubled punters pushing and shoving to get to the exits, a relative handful of folks such as the Bishop with his writings are quietly asking everyone to pause, to not panic, to look around, and study more about this 'Fire!'. If it exists at all , it seems not to be an imminent threat, and we can handle it in a more steady, and steadfast and rational, manner than the most vocal of the crowd are demanding, based on assertions which are beyond their expertise, and anyone else's, to be so confident about.
Michael Shermer strikes a chord with;
"Bottom up reasoning is counterintuitive(sic). This is why so many people believe that life was designed from the top down, and why so many think that economies must be designed and that countries should be ruled from the top down."
Which to my mind defines both the problem with and scale of the AGW debate.
Uncertainty is certainly good at the Telegraph. Geoffrey lean has it that 'more frequent and intense downpours expected with CLIMATE CHANGE' while his colleague Louise Gray repeats the advice from the RHS 'prepare for CLIMATE CHANGE by planting drought resistant flowers and cacti'
Mind you the RHS gave this advice 7 years ago as well.
For good measure the Guardian has deer 'rutting' early as a result of CLIMATE CHANGE.
I don't understand this thing about uncertainty with climate models.
I think I get why physical observations give rise to uncertainties but models?
They produce output projections based upon a myriad of interacting parameters. They are fitted to past observations and make guesses at future outcomes of complex and poorly understood chaotic processes.
If we can't create a model that predicts the first three past the post at the 3:30 at Doncaster or the share price of BP this coming October then what chance do we have of assigning meaningful uncertainties to decadal climate prognostications?
Modelling chaotic and non-linear systems and basing your future upon their output is wishful and foolish fantasy.
There are two different types of uncertainty being bandied about here as far as I can see. There is the 'scientific' uncertainty, which is in essence a mathematical description of the possible errors. That is actually a quantifiable and 'known' value (or it should be if the scientists are doing their job properly. It tells us what range we can be confident in for a given parameter.
There is also the 'uncertainty' of not knowing how/why things can happen, to a greater or lesser degree. This is not a mathematical term, but indicative of a lack of knowledge.
Making key decisions on the basis of well quantified uncertainty of the first type is rational and sensible as long as the parameters are known. This is the type of risk analysis that I think Roger Pielke Jr and Steve McIntyre refer to in a number of places.
Making expensive, critical and major economic decisions on the basis of the second case - lack of knowledge - especially when fuelled by dogma is downright dangerous.
It seems to me that that paper confuses the two.
I read a few of the posts in the Grauniad and it is obvious that several of the authors either did not read the article or do not understand 'uncertainty' as a concept in science. However, it is interesting that an article like this has appeared in a newspaper that normally publishes only propaganda.
I think the argument would be better understood if inconclusiveness was used instead of uncertainty.
The story of climate science is a story of people who have a felt need to rush to conclusions. The actual usefulness of whatever conclusion is reached is secondary to the urgency of arriving at it. And those who are more able to hold in-conclusion open (as a necessary means to understanding) are deemed to be dangerous obstacles to this project.
It is significant, I think, that climate science is quick to overwhelm and close-down the potentially inconclusive space with its elaborate (and familiar-sounding) end-of-the-world fantasies. It might be the same space where our non-secular ancestors once located the idea of 'God'... as a means of making it a tolerable, open, prerequisite to knowledge.
"We must learn to love uncertainty and failure, say leading thinkers"
Whomever came up with this title and decided that 'love' should be part of a scientific process is either misrepresenting what the 'leading thinkers' say, or, if it correctly captures what the 'leading thinkers' say, then the 'leading thinkers' are more than slightly confused. Somebody is mixing in some poetry, here.
Love is not within the reach of science.
Andrew
While both science and religion are branches of philosophy, they are separate and should be kept that way.
Unfortunately, Post Modern "Science" exists and apparently thriving.
Uncertainty is the lack of knowledge and should not be replaced with belief. Yet, there it is.
Sad.
Some notable quotes there:-
'Being able to quantify uncertainty, and incorporate it into models, is what makes science quantitative, rather than qualitative.'
'We should stop thinking that the largest future earthquake or market panic will be as large as the largest historical one; the longer a system persists, the likelier it is that an event twice as large as all previous ones is coming.'
'Failure is not something to be avoided but rather something to be cultivated.'
'All creative avenues yield the maximum when failures are embraced.'
And some that are not there:-
'There is nothing so absurd or ridiculous that has not at some time been said by some philosopher.'
Oliver Goldsmith
'From the sublime to the ridiculous is but a step.'
Napoleon
As any Gilbert and Sullivan afficionado can tell you, climate science's theme song was written by W.S. Gilbert in The Gondoliers (1889).
It goes like this:
"Of that there is no manner of doubt -
No probable, possible shadow of doubt -
No possible doubt whatever.
No possible doubt whatever."
Sound familiar?
On Exactitude in Science
. . . In time, those Unconscionable Maps no longer satisfied, and the Cartographers Guilds struck a Map of the Empire whose size was that of the Empire, and which coincided point for point with it. The following Generations, who were not so fond of the Study of Cartography as their Forebears had been, saw that that vast Map was Useless, and not without some Pitilessness was it, that they delivered it up to the Inclemencies of Sun and Winters. In the Deserts of the West, still today, there are Tattered Ruins of that Map, inhabited by Animals and Beggars; in all the Land there is no other Relic of the Disciplines of Geography.
From Jorge Luis Borges, Collected Fictions, Translated by Andrew Hurley
The fact that global warming estimates are uncertain, for example, has been used by many to argue against any action at the present time -Lawrence Krauss.
C'mon. If we don't know then why would we want to act? This could go in many directions. Police convict people all the time when they don't know. Many people have been exonerated of wrong-doing thanks to more certain results using DNA.
If we don't really know if the globe is warming, why would we want to act as though it is. It might be cooling. We could be wrong. Then we could be guilty of accelerating the cool down...and really be screwed.
Don Pablo, no body has proven that post-modernism is alive and well in science more than climate scientists who design hockey sticks.
I've been looking for some commentary from Physicist Alan Sokal who hoaxed the post-moderns in 1996 in their Social Text journal. He authored Fashionable Nonsense as a result of the hoax. I'd love to hear what he thinks of climate models that produce hockey sticks.
[Snip - namecalling]
"it is interesting that an article like this has appeared in a newspaper that normally publishes only propaganda..."
I'm cynical of Grauniad's motivation for publishing. I can feel a Monty Python script in all this:
Scientists come on stage dancing silly dances, holding up silly science statements. First the audience listens but then questions and then boos start. Scientists are shuffled off stage, replaced by a hypnotist. "Now we are going to learn about the uncertainty factor. On the count of ten, everyone here will become uncertain about what the scientists have said. You're fine with that? Yes, and you're going to become uncertain about even having seen them or heard them. Okay? Count to ten..."
Audience show clearly that they have become Uncertain.
Scientists come on as before, silly dances, waving silly placards... whose wording has been carefully changed. No mention of climate skeptics input for instance but UHI is now being "factored into models"... etc...
Audience behave like this is their first encounter with Scientists. Their memories have clearly been Wiped.
Hypnotist re-emerges. "See, the Uncertainty Principle works. We are adding it as Unequivocal, er, 90 percent probability, to our, er, Post-Scientific Science..."
Neil Gershenfeld, director of the Massachusetts Institute of Technology's Centre for Bits and Atoms wants everyone to know that "truth" is just a model. "The most common misunderstanding about science is that scientists seek and find truth. They don't – they make and test models," he said.
------------
That's true, science is a system of building models (not necessarily computer models) which are tested and either discarded when they are shown not to work or extended to fit new observations.
When science and politics combine, certainty is imposed (the pretend certainties of politics) and the question of science vs the science establishment arises.
I get the distinct impression with this article, that although observations may not fit with a desired theory (or bundle of theories) which key with a political purpose, we should gloss over the uncertainties as part of the natural process of science and have faith and go along with the politics. Lots of reasonable statements taken on their own, used to support an unreasonable conclusion.
Maybe it's just because it appears in The Guardian.
ZDB,
You have to distinguish between 'Climate science' and specific falsifiable hypotheses relating to climate. If a hypothesis isn't falsifiable, but turns out to be infinitely mutable, it isn't worth anything.
I'm just after leavin' off banjo pickin' and off to tend paw's still and shoot me a couple possums.
Cosmic
I have no problem with being called a Bishop Hillbilly. After all, we are all astrotufers being financed by Jed Clampett out of the fortune he is making from crude (oil that is, black gold, Texas tea).
Fade up banjos.
ZDB,
I suspect I know why you post here. It's refreshing for you to read the comments of free thinkers, who present their own thoughts, as opposed to the endless and now boring tribal chant of 'Global Warming'.
Andrew
Yee Haa! Dreadnought :)
UHI it is not so that because you measure windy and non-windy nights have the same trend that therefore there would be no UHI effect.
Of course, there is UHI we measure it, in the daily weather report. UHI is a fact.
Of course, the nr of UHI sites increases we have urbanisation ongoing since 70y.
the fallacious logic of windy non-windy analysis , is that presumption is made that a windy night should be cooler for a UHI site.(cooler as averaged over decennia)
No reason for that whatsoever.
The measurement at the windy UHI site is not some air from a neighbouring non UHI site . It is air from further away and higher up. That air willl be 50% warmer as without the wind and 50% cooler as without the wind. Wind is not an influencing factor for the trend between UHI <-> non-UHI
I was in Innsbruck a while ago. minus 5degrees. Then a bit of wind came: it got 15degrees warmer in 12hours time. The famous Foehn wind (basically Saharah winds funelled through the Alps)
The non -UHI sites in the neighbouring mountains remained as cold as ever.
"...and use this to pick over the tiniest little ambiguities from climate scientists..."
Tiny little ambiguities indeed Z.
Good to see that you've identified ambiguities in Climate Science.
Could you be more specific, please.
You've also put your finger on the problem of determining the outcome of Chaotic systems.
Their hyper-sensitivity to tiny, little changes with initial conditions.
Climate is a wee bit tricky, Z.
Assigning meaningful confidence bands to the projections made by GCMs is meaningless.
If, on the other hand, predictions could be made rather than projections then uncertainty metrics may be meaningful.
The IPCC spells it out quite clearly that the models don't, and can't, predict future climate.
They can, and do, make predictions. Do you deny that the IPCC is wrong to be cautious?
@PeterS
Love your suggestion about replacing the word uncertainty with inconclusiveness in the given context.
It's spot on. Applying the term uncertainty to wishful guesswork about the future is 47.34 %, +/- 1 hand wave inconclusive.
On the other hand, projections about which way the climate will fall in future are 100% inconclusive, with a certainty of 100%, if Chaos Theory is to be believed.
Erratum:
My earlier post
"They can, and do, make predictions"
replace with
"They can, and do, make projections"
Best wishes
Roy
certainty is a thing to be claimed, not demonstrated.
Thanks for the link. I made the following post:
"The physicist Lawrence Krauss of Arizona State University agreed. "In the public parlance, uncertainty is a bad thing, implying a lack of rigour and predictability. The fact that global warming estimates are uncertain, for example, has been used by many to argue against any action at the present time," he said."
The comment above contains an equivocation on the word 'uncertain'. Scientific hypotheses are always uncertain in the sense that they are fallible; that is, new experience can prove them to be false. Einstein's hypotheses, now honored as laws, are uncertain in the sense that they are fallible. By contrast, critical claims of climate science, such as "increased concentrations of CO2 in the atmosphere cause more water in the atmosphere and, in turn, that raises temperatures" are uncertain in the sense that they have no reasonable record of confirmation at all. That is a very different meaning of the word 'uncertain' and no one should rely on unconfirmed hypotheses in making decisions about expenditures on reducing CO2 concentrations.
Yes this is all very well but didn't you know the 'science of climate' and the indubitable exactitude of great academic institutions like, the CRU of UEA and Penn State dictate: the 'science is settled'! MM CO2 will be the downfall of the polar bears, western civilisation, cause thermageddon and will tilt the axis of the terrestrial sphere causing further mayhem, even a flat earth or summat............. "It is written!" - the politicians all know it - aren't you bloody listening???
Or a revelation!!!!! Something all great scientists have known ere, since time began! Namely: "nothing in science [or under heaven and earth] is ever settled,".................here endeth the first lesson. Has the grauniad lost all it's marbles..............Or, did it ever possess any [collective editorial nous/brains/intellect......... from the outset? - NO!
"It ain't what we don't know that hurts us . . ."
". . . it's what we know that ain't so."
Cumbrian Lad,
You make an interesting point.
"There are two different types of uncertainty being bandied about here as far as I can see. There is the 'scientific' uncertainty, which is in essence a mathematical description of the possible errors. ... There is also the 'uncertainty' of not knowing how/why things can happen, to a greater or lesser degree. This is not a mathematical term, but indicative of a lack of knowledge."
Yet we do need to be able to make decisions in the face of either type of uncertainty. With respect to climate change warming can be estimated subject to the first type of uncertainty: maybe we will hit 6C in 2070, maybe 2100, maybe 2120. But it will happen. The myriad of impacts are harder to quantify. How bad will the droughts be? How fast will sea levels rise? What changes will lower ocean pH cause? Many of the mechanisms are still unknown. But consider that if the climate changes, we cannot change it back. If we develop alternatives to CO2 emitting energy sources, and then in light of better knowledge decide it is safe to emit CO2, we can go ahead and do so. Sooner or later we will run out of fossil fuel sources, so the effort won't have been a complete waste.
Yup, Stan. So true.
A lesson that has been totally forgotten.
CO2 must be the villain of the piece, we know nothing of any other actor so it must be it!
Through ignorance comes certainty and from ignorance consensus rules supreme.
The debate is non-debatable, to do otherwise is to accept the possibility of infallibility.
Deniers are the scum who refuse to die off and let us tell you how things should be done!
Mike, your certainty is a clearly held belief. That you appear to hold little doubt speaks magnitudes more about your instincts than about your scepticism.
OK, we are f**ked, that's what you believe.
I don't. On what grounds do you believe that were we are doomed?
I really want to know.
I'm guessing now, but I'll carry on, that we've spent similar amounts of time in assessing the scenario.
The more that I've read the less I believe the doom and gloom sayings. What made us so different that we interpreted the same messages so differently?
Mike's
seems a bit too certain to me.I just hope that we get 6C of warming rather than 6C of cooling. From what I've read from interpretations of ice-core data, changes of that magnitude have occurred in the past, quite naturally, in as short a span of time as a couple of years or so. The changes in temperature haven't been the result of changes in CO2 concentration.
The problem is that people want to talk about uncertainty but use the argument from knowledge. As for instance, when discussing homeopathy, one sometimes hears, well, we don't know everything, there are lots of things that remain to be discovered.
Yes, and the problem is that we do not know what they are, not that we do not know what they are, and the efficacy of homeopathic remedies is one of these things that we do not yet know, but which is true.
Much of the precautionary principle argument about CO2 goes like this. We do not know if its harmful therefore we should restrict it. But of course it applies to everything. We do not, to the same level of certainty, know whether erecting windmills may have catastrophic effects, so don't lets do it. We do not know whether sitting in the lotus position may cure him, so he had better do it.
Its incredible what contortions people go through in order to argue that we should or should not do things for which there is no evidence whatever, including arguing that the lack of evidence is itself a reason. What has happened to education?
michel
The problem is that if you increase the price of electricity to the developed world and deny electricity to the developing world, people will certainly die as a result.
Donald Rumsfeld got a lot of (undeserved) stick in the press for this statement.
There are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns – the ones we don't know we don't know.
Although it is logically perfect, and in Climate science the Unknown Unknowns are the ones that mess up the models in a big way, and the unknown unknowns just mess them up a bit. We can all think of known unknowns. This should be one of the joys science and research, finding an unknown unknown and making it a known unknown then with further effort making it a known known.
@ Scottie, Jan 15, 2011 at 8:32 PM:
Kudos for the Gilbert&Sullivan quote!
Seems the belly-aching about doubt and uncertainty wasn't exactly unknown to humanity before climate 'scientists' evolved ...
Btw - if post-modern science is dispensing with 'truth', preferring to debate the various degrees of uncertainties, what happened to axioms? Have they now also been junked?
[Snip. Namecalling]
[Snip - response to previous]
Mike
Yes, we need to make decisions despite lack of full knowledge, but that pre-supposes that we have 'enough' knowledge. We certainly have no basis for asserting that temperature will reach 6C above current, at any of your chosen timelines. Recent trends suggest warming is slowing, and even the best case models are being undershot. That strongly hints that adaptation rather than actively trying to 'manage' CO2 makes more sense socially and economically. To your questions 'how bad will...' all indications are that they'll be as bad as ever they were and no worse. I really can't take seriously the various 'tipping point' ideas, these are based on weak science and paranoia. In business we are used to taking decisions on incomplete information, but the risk has to be measured against the outcome. Shackling the worlds economies for the benefit of very few actually makes it difficult to adapt, and therefore imperils the greater part of humanity, Such a policy can only appeal to those who really believe Man is a pest, and needs to be eradicated. I believe Man is a much more noble beast.
Maybe the new call to express uncertainty is because the original AGW forecasts are seen to be wrong, for want of a better word. If you now express great uncertainties you then fit in with the new AGW is connected to all bad weather events, so a forecast of warmer snow free winters in the UK can now fit in with 3 cold winters in a row as the band of uncertainty covers both eventuallities.
But how far before the "band of uncertainty" turns into "we don't have a ****ing clue!"?
Because that seems, to this uneducated but not unintelligent observer, to be just about where we are now. The only caveat being that "we" are still trying to pretend that we know what we are talking about. A poster at Booker's site is ripping TAR and AR4 to shreds as far as Australia is concerned and demonstrating that when you boil it all down to simple English the message is "it will continue to be very dry. And very wet. But probably not at the same time." With a bit of doom-mongering thrown in because, we must never forget, the IPCC is a political organisation tasked solely with investigating what to do about man-made global warming
In essence, the IPCC appear to be actually getting the message right (by and large). It's the Summary for Policymakers where the advocates and the eco-warrior NGOs are being given their head and either cherry-picking or distorting the actual findings to drip poison into the ear of government.
And then, of course, there's the Team with their own axes to grind. But what we are starting to hear (if we listen carefully enough) is that they no longer have the answers.
For those interested there is good analysis of uncertainty in climate science by Judith Curry here:
http://judithcurry.com/2010/11/28/waving-the-italian-flag-part-i-uncertainty-and-pedigree/
http://judithcurry.com/2010/09/22/the-uncertainty-monster/
Recommended.
The problems of modelling and of uncertainty are further explored by Roger Pielke Snr here:
http://pielkeclimatesci.wordpress.com/2007/06/18/comment-on-the-nature-weblog-by-kevin-trenberth-entitled-predictions-of-climate/
Pielke examines the well-known statement by Kevin Trenberth regarding modelling. Those over-invested with confidence on the solidity of climate science would do well to consider what he says carefully. And yes, I do know that he goes on to rehearse the orthodox position, but that is not relevant to what he says here:
Mike Post wrote: "The problem is that if you increase the price of electricity to the developed world and deny electricity to the developing world, people will certainly die as a result."
No one is proposing the latter. People are being killed by more intense floods and heat waves now. All projections indicate things will get much worse. Invading countries to get their oil tends to kill people. Mining coal is a bit dicey too. Cap and trade isn't priceless, but it not going to kill anyone.
Phillip Bratby wrote: "The changes in temperature haven't been the result of changes in CO2 concentration."
There is no physical or experimental basis for the theory that doubling or tripling CO2 concentrations don't impact climate. In the past sometimes orbit changes drove up temperature which then drove up CO2 and H2O which further warmed the planet.Other times volcanic activity drove up CO2 which then drove up temperatures which then drove up atmospheric CO2 & H2O which amplified the warming. Now, we are pumping CO2 into the atmosphere rapidly. There is only one possible direction this can push us in. The only questions are how fast will this happen and how bad will it get. Given that the ~0.5C increase is already causing some problems, the palioclimate data and yes computer modeling studies, the damage we are doing is likely to exceed the cost of changing our energy industries - something we will have to do sooner or later anyway.
Mike Post wrote: "The problem is that if you increase the price of electricity to the developed world and deny electricity to the developing world, people will certainly die as a result."
"No one is proposing the latter." No but energy pricing tends to follow global trends, when the oil price goes up coal and gas prices follow, it may be an unintended consequence but a consequence it will be.
"People are being killed by more intense floods and heat waves now." Yes but not because of CO2 or because they are more intense, Austrailia is beacause of a cooling La Nina, Brazil is do to building on valleysides displacing stabilising trees and vegatation. All show increased deaths where population eg Pakistan are incresing rapidly
"All projections indicate things will get much worse." But what where and when is never said or if it is its wrong eg drought ridden Australia or snow free UK.
Mike
Some of what you say is true. Some is supposition (hypothesis) presented as fact. This is not the same as truth.
It is vital to distinguish clearly between the two, especially when supposedly presenting a scientific argument.
You do not do so above, which greatly weakens your position.