No-one Does Wrong Quite Like Lewandowsky


Uncertainty as Knowledge

Let’s say, for the sake of argument, that you wake up one morning with a burning desire to learn everything there is to know about uncertainty and its relationship to knowledge. And let’s say, for the sake of extending this rhetorical ploy, that you want to make sure that your desire is fulfilled without fear of being misinformed. Wary of Wikipedia, you turn instead to the Philosophical Transactions of the Royal Society – Mathematical, Physical and Engineering Sciences. To your delight, you find a special issue containing not one but a complete set of papers introduced under the title: ‘Uncertainty as Knowledge’. Finally, let us say, for the sake of dramatic irony, that you have never heard of the author, Stephan Lewandowsky, and so there is nothing to suggest that you have done anything other than stumble upon a mother lode of expert wisdom.

Now, for the sake of the unexpected twist, let’s try repeating the story but, this time, replace the eager student with a hard-bitten sceptic who had recently written an article bemoaning Peter Gleick’s treatment of the subject of uncertainty, and who was just looking to see where else such misdirection could be found. Surely not within the publications of the Philosophical Transactions of the Royal Society? Well let me tell you what I discovered so that you can judge for yourselves.

Experts on Parade

There are actually ten papers in the Royal Society’s set, not including the introductory paper written by Stephan Lewandowsky, Timothy Ballard and Richard D. Pancost. This means that there is far too much material for me to critique in just the one post. Consequently, I will concentrate here on the Lewandowsky et al introduction, together with the first of the papers it introduces – namely, a paper written by Mark Freeman et al, titled ‘Climate sensitivity uncertainty: when is good news bad?’ I may follow up by writing a review of the remaining nine papers but only if nothing more interesting distracts me in the coming few days.

In a nutshell, Lewandowsky’s introductory paper explains that, far from being an excuse to delay climate change action, uncertainty is the reason why one should act. The important relationship between uncertainty and knowledge is that the extent of the former provides the knowledgebase upon which we can formulate our risk management planning. To put it in the authors’ own words:

“Growing uncertainty about the future therefore ironically imbues us with the knowledge of what we can do to escape that uncertain future.”

A number of arguments, taken from ‘physical, economic and social perspectives’, are offered to back up this assertion. However, it is the Freeman et al paper that takes pride of place:

“Several articles in this issue expand on the relationship between uncertainty and knowledge. Perhaps the most formal and counterintuitive treatment is provided by Freeman et al. [2,3] whose article extends an initial analysis provided by Lewandowsky et al. [5,6].”

What we are faced with, therefore, is a phalanx of experts, lined up and waiting to set the record straight by providing a comprehensive and multi-stranded thesis, starting with (but my no means limited to) a mathematical analysis of uncertainty which builds upon that performed by none other than the great man himself. And all of this brought to you by that most august of authorities: The Philosophical Transactions of the Royal Society.

Such a shame, therefore, that the thesis falls flat on its face in the opening two sentences.

We’ve Been Here Before

I think we can all agree that if you are going to embark upon a lengthy dissertation on such a philosophically difficult subject as uncertainty and its relationship to knowledge, you should help your readers on their way by starting out with a clear and explicit statement as to what uncertainty is and where it comes from. Specifically, one should point out that uncertainty has two basic foundations: variability as an inherent feature of nature, and incertitude resulting from gaps in knowledge. Here is what Lewandowsky et al came up with:

“This issue of Philosophical Transactions examines the relationship between scientific uncertainty about climate change and knowledge. Uncertainty is an inherent feature of the climate system.”

Well, nature’s inherent variability certainly seems important to the authors, but what about incertitude resulting from gaps in knowledge? One might be forgiven for thinking that incertitude’s role as a foundation for uncertainty is covered by the first sentence since, after all, knowledge does get a mention. But don’t be fooled. It quickly transpires that the authors are not referring to incertitude as a source of uncertainty; rather, they go on to make the point that analysing the ‘scientific’ source of uncertainty (the climate’s inherent variability leading to a range of possibilities) provides the knowledge required to justify action, i.e. analysis of variability is the road to certitude. That is their idea of the relationship between uncertainty and knowledge – uncertainty is ‘a source of actionable knowledge’. Indeed, the very idea that lack of knowledge can lead to an uncertainty that obstructs good decision-making is soon dismissed as a scurrilous ploy:

“Although the climate community has sought to develop ways of dealing with the various forms of uncertainty (e.g. [1,2]), uncertainty has often been highlighted in public debates to preclude or delay political action (e.g. [3,4]). Appeals to uncertainty are so pervasive in political and lobbying circles that they have attracted scholarly attention under the name ‘scientific certainty argumentation methods’, or ‘SCAMs’ for short [3].”

If one follows the links, one discovers that the so-called scholarly attention to which the authors refer is little more than the well-used strawman argument that sceptics misunderstand the scientific method by demanding the removal of all incertitude. I don’t intend chasing that squirrel here, because the real problem is that an anxiety to downplay the important role of incertitude in the decision-making process has caused individuals such as Lewandowsky to completely overlook its role in undermining any proposed analysis of inherent variability. This, indeed, is the same mistake that Peter Gleick makes when he talks of ‘scientific uncertainty’ as if it relates only to the quantification of a range of possibilities. It’s almost as if these individuals actually believe that ‘scientific uncertainty’ can be quantified without considering the weight and heterogeneity of evidence.

The classic expression of this basic mistake is to attempt to analyze the uncertainty regarding equilibrium climate sensitivity (ECS) as if it were simply an exercise in measurement theory, conveniently furnished with probability distributions for the purposes of statistical analysis. In point of fact, the frequency distribution of ECS values provided by the collected outputs of climate model ensembles (and proxy reconstructions for that matter) is an expression of deterministic incertitude as much as stochastic climate variability. Its shape is an artifact of the current state of knowledge and it hides all manner of selection effects and evidential issues. It certainly is not just a measurement spread resulting from inherent natural variability. You can’t therefore treat it as a straightforward probability distribution, and to do so is a grave error. Everyone knows this, don’t they? So surely a gaffe such as this would not feature in the Philosophical Transactions of the Royal Society — would it?

Introducing the First Paper

Lewandowsky proudly points out that the first paper in the set (‘Climate sensitivity uncertainty: when is good news bad?’, by Freeman et al) builds upon a mathematical demonstration of his own, in which he shows that increased uncertainty regarding equilibrium climate sensitivity unavoidably implies increased risk, provided (as seems reasonable) that the posited damage function is convex. From this observation, Lewandowsky had concluded that uncertainty is not the sceptics’ friend. The Freeman et al paper extends this finding by further pointing out that, even when the increase in uncertainty involves only the lowering of the lower bound, the effect is still to increase the imperative for action. I won’t go into the details of the mathematics used to prove this point, other than to point out that theirs is basically an ordinal argument. It all looks very impressive and extends to a number of pages. Nevertheless, there is one basic problem that both the Lewandowsky and Freeman et al papers share: they both misunderstand the foundation of the ECS uncertainty and so make the mistake of assuming that the ECS frequency distribution can be treated as a simple probability distribution. At the end of the day, none of the authors involved seems to be particularly concerned about the difference between aleatory and epistemic uncertainty and no-one seems to appreciate the difficulties encountered when one uses the analytical methods devised for the former to analyse the latter. They are using the wrong mathematics and so are proving nothing – ordinal or otherwise. The result seems impressive but is based upon the false premise that the shape of the distribution accurately reflects the scale of the uncertainty.

You might think it harsh of me to suggest that Lewandowsky and his fellow contributors are guilty of an egregious error. After all, they are hardly alone in attempting to use frequentist statistical analysis to investigate incertitude. They are simply following a well-established tradition and extending it for all it is worth – which, to be honest, may not be so much. Furthermore, there are actually circumstances where the incertitude of the masses does act stochastically. Even so, there is no basis for assuming this applies in the example of the ECS frequency distribution. When dealing with evidential weight, consensus is a dodgy surrogate.

Traditional or not, just because so many have made the same mistake, that does not make it right (or perhaps it does if you follow Lewandowsky’s logic). Lewandowsky et al only attract my opprobrium because they make the error whilst mocking those who fail to do so (SCAMs indeed!). It is in that sense that no-one does wrong quite like Lewandowsky.

Ignorance as Uncertainty

I’d like to say that things get better after the Freeman et al paper, but they don’t. Far from being an aberration in an otherwise excellent set of papers, Freeman et al rather sets the standards for those that follow. The headline thesis is clear enough: Greater uncertainty implies a greater range of adverse possibilities and so contributes to the list of recognized perils. But nowhere within any of the papers is there a recognition that an increase in uncertainty may also undermine confidence in the range calculation, or even render meaningless the very idea of interpreting a frequency distribution as a probability curve. In fact, nowhere is there any serious discussion of epistemic issues such as evidential weight or the non-complementarity of evidence. Consequently, whilst I searched eagerly for any papers that covered the likes of fuzzy logic, imprecise probabilities, possibility theory, Dempster-Shafer theory, Info Gap decision theory, Bayesian belief networks or causal inference, all I found in the set was a loosely connected bunch of papers telling me that the more uncertain I am, the more I know how frightened I should be. This was a thesis that worked very well for me when I was a child afraid of the dark. Now I like to think I am more enlightened and I can appreciate that knowing what may possibly exist is a long way from knowing what necessarily exists. As a result, I am no longer a hostage to my imagination.

For the moment, I’ll leave it there. As I have said, there is too much material in this particular issue of Philosophical Transactions of the Royal Society for me to review in one post. I may or may not resume the critique in a future article. But, if I do, you shouldn’t expect there to be any great surprises. Whilst all of this stuff goes under the banner of ‘Uncertainty as Knowledge’ it also suffers from a lack of knowledge of what uncertainty is.


It has been suggested that this article is making a pedantic technical point, since no amount of quibbling over the foundation of uncertainty alters the fact that reducing carbon dioxide emission will reduce the risk levels. On the contrary, I maintain that a firm grasp of uncertainty’s conceptual foundation is essential if one wishes to comment upon how uncertainty bears upon the decision-making process. For that reason, it is important to understand that an over-reliance upon standard probability theory, particularly combined with the view that scientific uncertainty is purely a manifestation of inherent variability, is a fundamental impediment to a full understanding. It may be an overstatement to refer to this as a ‘grave error’, nevertheless, there are already alternative approaches that are being explored, even within climate and environmental science. I list below some further reading that I trust hints at this relatively unexplored potential:

Using fuzzy logic to determine the vulnerability of marine species to climate change

Implementing fuzzy decision making technique in analyzing the Nile Delta resilience to climate change

Climate change impact assessment: Uncertainty modeling with imprecise probability

Using Dempster-Shafer Theory to model uncertainty in climate change and environmental impact assessments

Addressing ambiguity in probabilistic assessments of future coastal flooding using possibility distributions

Info-Gap Decision Theory for Assessing the Management of Catchments for Timber Production and Urban Water Supply

Finally, it is worth reminding the reader once more of the importance of robust decision-making:

Climate change adaptation planning, research and practice

129 thoughts on “No-one Does Wrong Quite Like Lewandowsky

  1. Very helpful overview John. And in the nick of time I see:

    From 1 September 2020 you’ll need a subscription, usually through an academic library, to read some of our articles. If you do not have access you can purchase using pay-per-view, or contact your librarian. If you are a researcher or librarian and need help with access issues please email access@royalsociety.org.

    Liked by 1 person

  2. Richard,

    Well spotted. I had fully intended to leave posting this article until after the bank holiday weekend but decided I needed to post now because of the paywall threat. Even so, the two papers of greatest interest are available elsewhere. The Lewandowsky paper covering convex damage functions is externally linked from the Transactions and, subsequently, also from my own article. The Freeman et al paper first appeared here as a non-peer-reviewed discussion paper:


    It appears to be identical to the version that appears in the Philosophical Transactions.


  3. There’s been some posts on this topic at Climate Etc over the years, starting from about 2012 when Lewandowsky was first developing this approach, if one may grace it as such, on his blog “Shaping Tomorrow’s World”, which didn’t go unnoticed. First out with a counter-argument was Ben of this parish, as recorded here: https://judithcurry.com/2012/06/06/uncertainty-is-not-your-friend/

    …but Judith and others soon came to similar conclusions as yourself, starting indeed with: “In fact, nowhere is there any serious discussion of epistemic issues…” It was bad enough back then, what is even more problematical is that time and perspective appear to have resulted in consensus sources having made little or no effort to address such issues. Meaning that years later, narrative policing is still very much a thing.

    Liked by 4 people

  4. Here’s a new Lewandowsky paper that uses ‘complexity science’* to argue that citizens’ assemblies can overturn populism:


    It lacks the cojonescomplexity-scientifics to actually use the words ‘populism’ or ‘populist’, but that’s what it’s all about.

    On the plus side, this new Lew paper doesn’t cite any old Lews in support. A first?

    *’Complexity science’:


    As you can see, it’s not just sociologists wanking off to the word ‘science’. There’s far more to it than that. It’s complex.

    Liked by 3 people

  5. In point of fact, the frequency distribution of ECS values provided by the collected outputs of climate model ensembles (and proxy reconstructions for that matter) is an expression of deterministic incertitude and not stochastic climate variability. Its shape is an artifact of the current state of knowledge and it hides all manner of selection effects and evidential issues. It certainly is not a measurement spread resulting from inherent natural variability. You can’t therefore treat it as a probability distribution, and to do so is a grave error.

    I think this is the point I was trying to make at a discussion of this subject at Amman’s blog many years ago, where I got roundly criticised by a number of highly qualified scientists (Willard was one) whose logical demolition of my arguments involved comparing me to a monkey, among other things.

    My point was not just that Lewandowsky was wrong, but that he was out with the fairies, or rather with the astrologers and numerologists, since his argument implies the possibility of deducing something about the probability of events happening in the real world from variations in the shape of a graph representing frequency distribution of opinions.

    Lewandowsky’s argument was that if you take a skewed distribution of estimates of ECS values and alter it to express the idea of greater uncertainty, this results in a lowering of the peak value (the mode, or most widely held opinion) and a bulge in the higher values off to the right, with the median (the midway opinion) and mean (average of all opinions) following. In other words, greater ignorance results in a flattening of the “peak” (representing consensus) which in turn results – mathematically – in a greater probability of a higher climate sensitivity.

    My argument was that it doesn’t matter whether the opinions are based on the best estimates of ECS values by the world’s finest climatologists or the patterns formed by chicken entrails, you can’t do it. There’s no way that you can get from a change in the opinions of mortals (which are the only data in L’s graph) to a change in the real world. Lewandowsky is in effect claiming that you can change temperatures by becoming more ignorant about them.

    His error is in assuming that estimates of climate sensitivity can’t take a negative value, and that therefore his probability distribution is bounded at zero. (Even the IPCC denies that, since their playing around with summing different feedbacks envisages the possibility of a negative warming effect.) But anyway, it can’t be bounded at zero, because the graph is not a measure of climate sensitivities, but of estimates of climate sensitivities, which only exist in the mind, where anything is possible. If the values are not bounded at zero, a collapse of the probability peak (i.e. a lapse into the fog of uncertainty) will result in peak consensus values rolling off to left and right like a collapsed soufflé, and not in an inevitable surge to the right, resulting in higher mean and median temperature estimates.

    Lewandowsky’s argument fails because his initial premiss that ECS estimates are bounded at zero is just wrong. But that’s just a mistake. What follows is far more serious, because it shows a psychologist who thinks that changes in a bar chart demonstrating how many people think certain things can change events in the real world. That’s numerology.

    Imagine a bar chart of people’s estimates of Trump’s IQ – certainly bounded at zero – with a peak (mode) at – say – 45. (Democrats tend to converge round this low figure while Republicans’ opinions are more spread out in a fat tail trailing off to genius level.) Imagine that we discover that we know nothing about Trump’s IQ. This will cause opinions to spread. Democrats’ guesses can hardly get much lower, but the peak at 45 is eroded by the realisation of our ignorance, so the spread will be towards the higher end. Lewandowsky would argue that due to our greater ignorance, it has become more probable that Trump is more intelligent than we thought. This is false. Or even that our ignorance has made him more intelligent than he was. This is insane.

    Liked by 1 person

  6. At Amman’s blog I used the example of bra cup sizes. (I wanted some kind of measurable quantity that couldn’t take a negative value. No doubt staring at all those skewed bell curves affected me unconsciously.)

    Take a number of estimates of a lady’s bra size made by domain experts, running from A to D+++, but peaking at the low value of B and tailing off in the higher reaches of male fantasy. Now introduce an ignorance factor by having the lady don a chunky sweater. Certainty is destroyed, and the consensus around B shattered. Few lower their estimates to A (why should they?) but uncertainty leads to more wild guesses at the higher end. Lewandowsky would argue that by donning a sweater, the lady increased her probable breast size. Wrapped as he is in the thick fog he has knitted for himself, he sometimes seems to be saying that her breasts got larger when she put the sweater on.

    I got a lot of insults for this argument at Amman’s blog, but Lucia Liljegren was supportive.

    Liked by 2 people

  7. Geoff: One’s tempted to say Lucia has balls. Whatever, she’s a wonderful human being.


    I’m relatively new to this denier thing and still playing catchup

    Yet even now, No-one Does Denialism Quite Like Ridgway?

    Liked by 2 people

  8. …Yes I know it’s ironic that the climate has inherent vulnerability and we don’t know what we don’t know, including ECS, so it might seem counter-intuitive to plunge in and destroy the economy on the basis of an uncertain solution for je ne sais quoi, to which I say, no, charge on, go boldly where no one has gone before, trust our climate modellers and our papers in The Philosophical Transactions of The Royal Society. Lew.

    Liked by 1 person

  9. Generally science manages to converge towards a value for a measurable thing. It is one of the many oddities of climate science that this tends not to happen for any important measurement. Indeed measurements themselves are changed retroactively.

    Forty years on, and our uncertainty is not decreasing. Wise heads might think that is a problem.

    They require rescuing from this impasse, and Lew is already ready for a challenge.

    Liked by 1 person

  10. Never mind Lew and his fellow science-denying loonies. We’ve lost Richard Betts. He’s gone over to the Dark Side and is now publicly endorsing XR’s ‘science’. Climate science has finally jumped the shark. Betts is the canary in the coal mine. It is no longer a scientific discipline – or should I say, it no longer even bothers to keep up the pretence of being a scientific discipline.

    Liked by 2 people

  11. Chester,

    “Forty years on, and our uncertainty is not decreasing. Wise heads might think that is a problem.”

    I’ll rephrase that, if you don’t mind. Forty years on, and despite considerable reduction in our epistemic uncertainty, the shape of the ECS frequency distribution has not substantially changed. Wise heads might think this a problem. In fact, wise heads would realize that this is only possible because the distribution is not one of probability and it doesn’t fully capture the uncertainty. If we, just for the moment, accept that the distribution reflects aleatory uncertainty, then there is nothing in the rule book that says that when the epistemic uncertainties are reduced, the aleatory uncertainty will reveal itself as being less than previously thought. In fact, it may go up, down or stay the same. Staying the same is the least likely, so wise heads would be asking, “Isn’t it a remarkable coincidence that the initial numbers, plucked out of the air back in the seventies, are proving to be the correct figures no matter how much our understanding improves?”

    The fact is that the ECS distribution is not a measurement distribution at all, it is a range of accepted opinions. And we all should know that knowledge is by no means the only thing that determines that range!

    And before anyone comes on here to point out that the range is not just opinion because it includes the ECS values calculated from climate reconstructions, I have just one thing to say:


    Liked by 2 people

  12. Geoff,

    We can look at this from the perspective of imprecise probabilities. Is there any reason to expect that the values on the alleged probability distribution are equally reliable throughout the distribution? Actually, no. The further one goes along the fat tail of the distribution, the less confidence one should have in its shape. In fact, the epistemic uncertainty expands so much that one might as well stop talking about probability and risk aversion and just accept that uncertainty has become the dominant cause of aversion. That is why, despite Lewandowsky’s clever use of ordinal arguments to avoid invocation of the precautionary principle (whilst still advocating precaution) I say he is still just invoking the precautionary principle albeit incognito.

    Or, as you charmingly put it, who really knows what is going on under the cardigan? A gentleman doesn’t ask.


  13. Richard,

    I like to think I do my bit. Here’s another observation to ponder:

    If one can talk of ignorance as uncertainty (i.e. epistemic uncertainty), and Lewandowsky wants to talk of ‘uncertainty as knowledge’, then how long should it take for some wag to put them together and talk of ‘ignorance as knowledge’? I think Lew has some explaining to do.

    Liked by 1 person

  14. Vinny,

    “As you can see, it’s not just sociologists wanking off to the word ‘science’. There’s far more to it than that. It’s complex.”

    I’ve just taken a quick look at the Lewandowsky paper you linked to. Nine authors to write 2,500 words! That was not a paper, that was an orgy with someone taking notes.

    Liked by 1 person

  15. An interesting aspect of variability is are we going to learn anything about it after it has played out over half a century or so. If disaster doesn’t pan out, will we know how lucky we were?


  16. John Ridgway, I confess to having been a bit tiddly last night. Lew’s Nature comment seemed very long, so I didn’t bother reading much of it.

    But, as you say, it’s actually very short and has an improbable number of authors.

    Most of whom are noisily anti-Brexit.

    Which makes this sentence very welcome:

    The conventional assumption amongst political scientists was that achieving democracy is a one-way ratchet.

    Anti-Brexiters acknowledging the convention of a one-way political ratchet is a good start; acknowledging its fallibility is even better. Though there’s obviously a long way to go.

    For decades, people like me who have opposed the EEC/EU’s irreversible increases in power via the Monnet Method’s tiny ratchets have been dismissed as racists and fascists. Lew and co will no doubt continue to insist that we are racists and fascists, but the ratchet itself is now there in the ‘scientific’ literature, and Lew’s name is attached, so it must be true.

    Also, I was very wrong about this Lew paper not being self-citing. It cites ‘Stability of democracies: a complex systems perspective’, a longer version of the same paper that was published in 2018 in The European Journal of Physics (?!) and was credited to the same nine authors plus Professor Alvin Birdi, a Bristol Uni economist who has said that few people trust academic enonomists and that they should be very careful of being accused of bias.

    Wise words, Birdi, wise words.

    Liked by 2 people

  17. Andy, this is not the Richard Betts I thought I knew. Spiteful. Vindictive. Sure, Piers often goes totally overboard with his theories, but I’d still trust him over any alarmist ‘climate scientist’ or XR freak.

    He was singled out by this dreadful government and slapped with an unjustifiable £10,000 fine. Why? Nobody else at the protest was. Piers is quoted as stating:

    “There is no justification in any terms for any of the anti-Covid measures. They are not to control Covid, they are there to control the population.”

    The facts, the science and the data back him up. There is nothing whatsoever conspiratorial in that statement.

    Liked by 1 person

  18. Jaime, indeed one should not be fined (especially, seriously) for having theories, however OTT. Nor for protesting with like-minded people, as long as there is no criminal damage, obstruction, etc. Considering that virtually the whole purpose of XR protests these days is to cause the latter, why aren’t they fined 10k each?

    Liked by 3 people

  19. It seems to me that you’re using some kind of pedantic technicality to arhue against what is essentially a very simple point. We don’t know how much we will warm if we emit a certain amount of CO2. It could be low enough to not be much of a problem, or it could be high enough to have impacts that are justifiably described as catastrophic. Given this, we might want to err on the side of being cautious and aiming to emit less than we might do if we could constrain climate sensitivity more precisely.


  20. ATTP,

    It seems to me that you’re using some kind of pedantic technicality to argue against what is essentially a very simple point. The persistent structural uncertainty in the field of climate science does not permit scientists to quantify (within reasonable scientific bounds of error) the amount of warming which is projected to take place by the end of the century. They kid themselves that a probabilistic assessment of the amount of warming is a scientific one and they kid themselves (and us) even further by insisting that the ‘fat tail’ of that probabilistic warming curve demands that we take Thermageddon seriously and form policy which allows for the occurrence of a ‘realistic worst case scenario’. Given the very significant negatives associated with pursuing a zero carbon policy ‘just in case’, we might not want to err on the side of caution.

    Liked by 1 person

  21. Jaime,
    You can, of course, frame it that way. In practice, though, one could consider the uncertainties in both the impact of climate change, and the impact of climate policy. A point to consider is that the uncertainty in climate sensitivity is such that if we don’t limit how much we end up emitting, the resulting climate change could have impacts that would justifiably be regarded as catastrophic. Hence, we might want to reduce the chance of such impacts materialising, which would require limiting how much we end up emitting. My own view is that we are innovative enough to do so in ways that won’t lead to economic catastrophic, but maybe others think that this is simply not possible. If so, I do question what you think will happen once fossil fuels do indeed become actually limited (or very expensive to extract).


  22. Ken, uncertainty about the impact of any quantifiable amount of of global warming is of course not the same as uncertainty about the actual amount of warming. Impact uncertainty (in terms of actual physical changes and our ability to adapt to them) merely multiplies the existing inherent ‘amount’ uncertainty, further undermining the justification for ‘worst case scenario’ policy measures (i.e. very rapid decarbonisation), the social and economic impact of which are rather more well defined.

    Liked by 1 person

  23. aTTP,

    My objection to your way of putting things (although I’m getting slightly off topic in terms of this discussion) , is that you again talk about “we” without defining who you mean.

    The simple fact remains that “we” in the UK, or in Europe, or even in the whole of the western world, could reduce our GHG emissions to zero, and it won’t make any discernible difference to the climate if the rest of the “we” (the rest of the world, in other words) don’t do the same. This is a rather fundamental point that the likes of XR and their supporters at the Guardian and the BBC (and you, I seem to remember, in terms of signing a letter of support) seem to overlook.

    There’s also the question of timescales. Do we accept the XR/Guardian/BBC view of this? If we do (carbon neutral by 2025 or 2030 or whatever), then your comment that this “…would require limiting how much we end up emitting. My own view is that we are innovative enough to do so in ways that won’t lead to economic catastrophic [sic]…” would also be highly questionable, IMO.

    Liked by 2 people

  24. ATTP,

    I woke up this morning in a more forgiving mood and so I am going to say now that it was perhaps hyperbole for me to refer to the use of a pdf to characterise ECS uncertainty is a ‘grave error’. Perhaps it would have been more temperate of me to say that it is perfectly understandable that everyone should attempt to capture the uncertainty in this manner and that, to an extent, it serves as an acceptable pretence to represent the uncertainty as if it were stochastically determined. However, I stand by my assertion that many of the underlying sources of uncertainty and biases are such that the pdf approach is nothing more than a convenient simplification that invites a number of false conjectures, including Lewandowsky’s. There’s just too much epistemic stuff going on for me to be comfortable with an ostensibly aleatory representation of the uncertainties.

    Is this a pedantic point? Well it might be if one were only discussing the current knowledgebase and how this informs our decision-making — you could argue that we already know enough to justify action, and I know that you do. However, arguments that are based upon a supposed relationship between uncertainty and knowledge, suggesting that the former can stand in for the latter, need to be based upon a much better conceptualisation of uncertainty than that offered by Lewandowsky.

    I was reading Knutti’s ‘Beyond Equilibrium Climate Sensitivity’ paper again the other day. There is enough in there to cause me to back away from the ‘pdf is nonsense’ line, but not enough for me to swallow the ‘uncertainty as knowledge’ one.


  25. John,
    I think you’re right that it’s not strictly correct to suggest that we can produce some kind of exact PDF for the uncertainties – some amount of judgement is required in order to quantify the uncertainty. That still doesn’t change that we are still unable, with high confidence, to rule out high climate sensitivity values. Hence, if we do want to give ourselves a good chance of avoiding the more serious impacts of climate change, we may want to take this into account when thinking about what future pathway we may want to follow, particularly in terms of emissions.


  26. Mark,
    By “we” I just mean our collective societies. Global warming is global. If we (I don’t know what other word to use here) want a good chance of avoiding the more severe impacts of climate change then there is a limit to how much more CO2 we can emit. We don’t need to do so, but the more that is emitted, the more likely it becomes that the impacts of climate change will be severe (to human societies and to other ecosystems).


  27. With the politicization of the global warming/climate change issue, the discourse is now located within the field of Sociology, in particular Environmental Sociology. The earth science datasets are all glass half-full, glass half-empty patterns, so the discussions are all about fears and imaginings and studies about which parts of the population have how much of which opinion. Of course, in an era of “post-normal science”, perception is reality (another sociological truism), and thus science is unhinged from objective observations. Truth is whatever the majority rules.

    Liked by 1 person

  28. ATTP,

    To be clear, here I am not arguing against application of the precautionary principle (there are reasons why I might, but that’s for another day). Instead, I am simply making the claim that arguments premised upon the assumption that a pdf is all you need to model the uncertainty may be flawed, because that is, in fact, an unsound assumption. In particular, I have pointed to Lewandowsky’s argument regarding the effect of convex damage functions (the one that says that increased uncertainty unavoidably implies increased risk). Remember that he proposed this mathematical trick as a way of avoiding the invocation of the PP, whilst still arguing that greater uncertainty means a greater imperative to act. I could accept that argument if he had not presumed that all of the uncertainties concerned lend themselves to a pdf treatment. I would actually prefer that he had just stuck to the PP and simply argued that being unable to rule out catastrophic worst cases is reason enough to justify action.

    In summary, I don’t think that worrying about the distinction between aleatoric and epistemic uncertainty, or the limitations of probabilistic approaches, is a ‘pedantic technical point’, though I agree it has nothing to do with arguments for and against the PP. On a constructive note, I finish by offering below an example of the non-probabilistic analyses I would like to see more of in climate science:



  29. aTTP at 11.03 am – thank you for the clarification.

    My problem with Greta Thunberg, XR, and indeed the Paris Climate agreements, is that they don’t express things the way you just did. They demand that the west takes action, immediately, while making no such demands of the rest of the world. So long as the GHG emissions of most of the world continue to grow, there’s nothing of any use we in the west can do, even assuming that growing GHG emissions might prove to be very dangerous to humankind on this planet.

    That being the case, I argue – and I think I have always argued – that we should concentrate on adaptation, given that we cannot in any meaningful way mitigate on our own in the west.

    Why don’t the UN, Greta, XR and many other of the loudest voices, demand that China, Indonesia, Russia, Iran, the Middle Eastern oil states, India, and many, many others take the same action that they demand we take? Their failure to do so doesn’t sit easily with your comment that:

    “Global warming is global. If we (I don’t know what other word to use here) want a good chance of avoiding the more severe impacts of climate change then there is a limit to how much more CO2 we can emit. We don’t need to do so, but the more that is emitted, the more likely it becomes that the impacts of climate change will be severe (to human societies and to other ecosystems).”

    Take this headline in the Guardian, for instance:

    “China’s European charm offensive disrupted by activists
    Protests against treatment of Hongkongers and Uighurs accompany foreign minister’s continental visit”


    I agree with the protests against Chinese treatment of the Uighurs and Hong Kong (though I would also add Tibet, and indeed their treatment of their own people). Nowhere in the article, or indeed in the protests, are climate change and GHG emissions even mentioned. On the contrary, we get this:

    “Merkel is pinning hopes of a German recovery partly on exports to China and more broadly sees China as a natural technological partner in the modernisation of German industry.”

    The irony, of course, is that here in the UK, far from making exports (of products or technology) to China, we have in large part exported to China our jobs and GHG emissions.

    Until climate alarmists wake up to geopolitical reality, and actually demand a global response, instead of giving 3/4 of the world a free pass, while demanding that the west commits financial suicide, then I will continue not to take them seriously.

    Liked by 1 person

  30. Mark,
    A few comments. Firstly, how much we will eventually warm will depend on how much we end up emitting in total. The developed world still dominates in this metric, and hence we will probably have contributed most to global warming. It’s maybe not unreasonable that we start making emission reductions sooner than parts of the developing world. As a consequence, many people feel uncomfortable telling people who are – on average – poorer than us that they need to make sacrifices now as a consequence of things we’ve mostly done.

    Secondly, there is essentially a global agreement. The global agreement is to get net emissions to zero about mid-century. I don’t think it’s correct that other countries are getting a free pass, even if they are getting a bit more time than we might be expected to have.

    Thirdly, I don’t really see why we should use other countries as an excuse to not do something. There are consequences to emitting GHGs into the atmosphere. We can, of course, decide that we’re going to continue doing so despite these consequences, but I don’t think we should then use other countries as our excuse for doing so. We should take responsibility for our decisions.

    Fourthly, I think there are opportunities here. At some point in the future we will move away from fossil fuels, even if we don’t do so because of climate change. There are presumably going to be economic benefits to being a country that has lead the way, rather than sitting back and waiting for others to do so.

    Fifthly, of course we will have to do some amount of adaptation. It’s not, in my view, a debate between mitigation and adaptation, it’s a debate between how much mitigation, versus how much adaptation.


  31. Some of the feedback received, particularly from ATTP, has prompted me to add a postscript with some further reading.


  32. aTTP at 7.53 am, thanks for the reasoned response, and apologies for the delay in responding. However, it’s been such a nice day, and we have such a bumper bramble harvest this year (in spite of, or because, of global warming?) that I’ve been out and about taking advantage.

    Dealing with your 5 points in turn:

    1. Of course, many in the developing world have barely emitted any GHGs at all, but the same cannot be said for much of the developing world, at least that part of the world that is defined as “developing” by the Paris Climate Agreements. It is difficult to obtain totally up-to-date information, and such information as we do have will no doubt be (temporarily) affected by the impact of Covid-19 on the global economy, but there is no doubt that the second-largest single national emitter of GHGs in the world to date is China (with the USA in first place). However, while (even under Trump) the USA’s emissions trends are reducing, China’s (and those of much of the rest of the world) are intensifying. Also worth noting is that China’s per capita emissions are around 50% higher than, say, those of the UK – which explains why China, a country with 20 x the population of the UK is responsible annually for around 30 x the emissions of the UK.

    2. There is a global agreement, yes, but the great myth is that the agreement amounts to more than a row of beans (it doesn’t). If you study (as I have) the INDCs submitted under the Paris Agreement, it is immediately apparent that the vast majority of countries are planning to – and are – increasing their emissions, and the much-vaunted goal of reducing emissions to zero by mid-century is unachievable unless something dramatic changes. It’s worth noting that the Paris Agreement contains no enforcement mechanism, so is legally a dead duck, and it also provides for self-certification of emissions. Does anyone still trust anything China says?

    3. It’s not really a case of using other countries as an excuse for not doing anything, rather it’s a case of not denying geopolitical reality. If we in the UK reduced our emissions to zero, that would amount to a few days of China’s increase in emissions, by way of example. And, in any case, what ARE we achieving. I repeat that we are outsourcing our jobs and our emissions to countries with lower environmental (and other) standards than our own. Big deal. Here’s the BBC reporting Greta Thunberg, with whom I assume you agree:

    “She also described the UK’s carbon emissions reduction as the result of “very creative” accounting.

    The country’s reported 37% reduction in emissions since 1990 was only 10% when aviation, shipping, imports and exports were counted, she said.”


    4. Your point here is, I think (in principle at least) a better one, and one I can in principle support. However, to what extent, and how exactly, are we leading the way? Here’s what the GMB thinks:


    “The bulk of the wind jackets our members will see from their living room windows will be manufactured in Indonesia for an Italian contractor, and transferred 7,000 miles on diesel burning ships back to the Fife coast where local people will pay for them through their electricity bills.”

    And much more in similar vein.

    5. We can adapt on our own. We can’t usefully mitigate on our own. If “we” are serious about reducing GHG emissions, then the Paris AGreement needs to be torn up and “we” need to start again.

    Liked by 1 person

  33. @Mark thanks for replying to these points – I had planned to do so but you saved me the trouble.

    Re: 3, I think we very much should use other countries’ inaction as an excuse for inaction. I would be happy to put UK Net Zero on the table for negotiation, iff the rest of the world did the same. I don’t think acting unilaterally here is good for our citizens, to whom our government is supposed to be beholden. In fact we don’t need Net Zero. We need a global per-capita CO2 ration, if that’s what is agreed. Thus Zambia etc can still increase their CO2 emissions under such a scheme.

    Re: 4, if we are going into the post-fossil-fuel era, then we ought not to do so while squandering money building wind turbines and solar panels. Cos if that’s all there is when fossil fuels run out, our goose is cooked.

    Re: 5, we obtain all the benefit from our adaptation and about 1% of the benefit of our unilateral CO2 cuts. (Based on population size.) Ergo, from a narrow nationalistic view, adaptation trumps CO2 cuts. (As mentioned, I am willing to offer such cuts in exchange for the same by other countries.)

    Liked by 1 person

  34. I note that the special issue of the Philosophical Transactions is still freely accessible despite the September 1st deadline for the introduction of the paywall. Therefore, I needn’t have rushed my article into publication. This is a pity, because I have since noticed that, in my haste, I had made a mistake.

    Where I wrote:

    “…and not stochastic climate variability.”

    I should have written:

    “…as much as stochastic climate variability.”

    The point is that there is an admixture of epistemic and aleatory uncertainty. This is not untypical, as I pointed out in my previous article (Gleick: What’s Not to Like?). The error didn’t affect the conclusions drawn in my article, but it is important that these details are correct, and it is important that I point out the error to you. The text has now been corrected.

    Liked by 1 person

  35. Ken,

    “A few comments. Firstly, how much we will eventually warm will depend on how much we end up emitting in total. The developed world still dominates in this metric, and hence we will probably have contributed most to global warming. It’s maybe not unreasonable that we start making emission reductions sooner than parts of the developing world. As a consequence, many people feel uncomfortable telling people who are – on average – poorer than us that they need to make sacrifices now as a consequence of things we’ve mostly done.”

    There are some huge logical flaws in that argument.

    1. The warming supposedly ‘baked in’ from previous developed world emissions will happen, regardless of what anybody does. It is thus irrelevant in any discussion of reducing emissions now.
    2. The vast majority of increases in emissions are occurring in the ‘developing’ world; the West is actively reducing and has greatly reduced its emissions.
    3. As a percentage of total GHG emissions, Europe, the US, Canada, and Australia account for about 25%

    You are saying that the nations which produce a quarter of emissions and which are actively reducing their share of emissions should do even more – thus putting themselves at a great economic disadvantage and further impoverishing their populations – whilst the rest of the world, which is increasing its current total share of 75% of GHG emissions, should not be expected to follow suit. Your justification for this demand is that, historically, the West has emitted more, but nothing whatsoever can be done to reduce any warming which may be ‘in the pipeline’ from historic emissions, therefore it seems to me that your proposal amounts more to a ‘punishment’ for Western nations for having the audacity to develop their economies using fossil fuels and bears no relation whatsoever to addressing the risk of future warming from emissions happening now. This anti-Western viewpoint is starkly illustrated in XR’s demands that the UK (which currently contributes only 1% and declining to GHG emissions) must get to net zero by 2025. I really do believe that they actually despise the UK for initiating the Industrial Revolution in the first place and they consider this our just punishment for doing so. This is actually the way they think. Social justice/climate justice trumps facts.

    I for one am not at all happy at the prospect of being ‘made an example of’ for the supposed ‘climate sins’ of my ancestors.

    Before you mention it, yes, I am aware that the UK outsources some of its emissions to places like India and China, but even accounting for those, we (and the West in general) are still not the main villains in this supposed ‘crime’ against the planet.

    Liked by 1 person

  36. Mark,

    Off topic? If so, it was ATTP who set the trend. Actually, I don’t think the ensuing debate was so much off the mark. My post is all about the importance of recognizing the implications of epistemic uncertainty and what this means for the decision-making process. It’s not good enough to dismiss such uncertainty as unscientific or imply that calling for its reduction amounts to a scam; nor is it good enough to act as though all the uncertainty is handed down to you by mother nature, leaving you with nothing else to do but to analyse it. The uncertainty runs deeper than that. If you follow the final link in my postscript you will find this:

    “Under deep uncertainty, decision-makers need to select robust options rather than best options. A focus on best options is more appropriate when it is possible to predict particular future states. However, when the future is characterised by uncertainty, a focus on best options may carry significant risks. In this context, Rosenhead et al. (1972) suggested the concept of robust decision making.”

    I think ATTP is focussing too much on what he believes is the best option. So maybe not so off-topic.

    Liked by 1 person

  37. ATTP is making a theological faith argument that wears sciencey looking lipstick.
    The faith based argument has to studiously ignore that there is no climate crisis except as declared by the faithful.. This assertion ignores that the link of CO2 to any given temperature or other weather event or climate trend is indistinguishable from pre-industrisl weather and climate.
    Set aside the circular faith/ magical thinking and look at the behavior: banning skeptical perspectives from the ATTP site even while making the derivative consensus arguments here. And then knowing that ATTP is actually a signatory to the anti-science, anti-civil society XR movement. And let us not forget as well his active support in a scurillous attack on scientist Susan Crockford.


  38. I’m sorry if I seem to be labouring the point somewhat, but here is another paper that I think may be useful in helping the reader understand the point I am trying to make:


    It is written by Professor Didier Dubois, a mathematician specialising in decision theory at the Institut de Recherche en Informatique de Toulouse. I would particularly like to draw attention to the following extract:

    “Risk can be defined as the combination of the likelihood of occurrence of an undesirable event and the severity of the damage that can be caused by this event. In the area of risk analysis, especially concerning environmental matters, it is crucial to account for variability and incomplete information separately, even if conjointly, in uncertainty propagation techniques.”

    If Didier Dubois thinks this is an essential point, I would be disinclined to dismiss it as an example of technical pedantry.

    Liked by 1 person

  39. Another paper for Lewandowsky and ATTP to consider:


    And the key quote is:

    “Techniques for decision making that rely on traditional probability theory have been extensively pursued to incorporate these inherent aleatory uncertainties. However, complex problems also typically include epistemic uncertainties that result from lack of knowledge. These problems are fundamentally different and cannot be addressed in the same fashion.”

    Note: “fundamentally” and not “pedantically”.


  40. John,
    I think you may be missing what I’m getting at. I’m certainly not suggesting that we shouldn’t be aware of, and take into account, that some of our uncertainty as to what may happen is due to a lack of knowledge. I also doubt that Stephan Lewandowsky would argue that the uncertainty is entirely determined by some kind of measurement uncertainty. My understanding of the basic point that he is making with his papers is that the overall uncertainty means that we can’t be sure as to the outcome and, hence, that we really can’t rule out outcomes that may universally be regarded as catastrophic. Hence, we should take this into account when considering what to do (which seems to be roughly what your sources are suggesting too).


  41. ATTP,

    > “I think you may be missing what I’m getting at. I’m certainly not suggesting that we shouldn’t be aware of, and take into account, that some of our uncertainty as to what may happen is due to a lack of knowledge.”

    I don’t think I am accusing you of making that suggestion. I am simply defending my claim that the distinction to be made between the two basic sources of uncertainty has more importance than the need to ensure that both types of uncertainty are included in an evaluation. The importance of the distinction also lies in the fact that the two types of uncertainty have to be treated differently mathematically, and the two types of uncertainty have different implications for the decision-making process, not the least of which is the fact that one can be reduced and the other can’t.

    > “I also doubt that Stephan Lewandowsky would argue that the uncertainty is entirely determined by some kind of measurement uncertainty.”

    Well one certainly would like to think so. But I have a problem here because, whilst it seems implausible that the likes of Lewandowsky and Gleick would think such a thing, they write as though they do. Take Lewandowsky, for example. His opening statement on this point, within his introductory paper, was:

    “Uncertainty is an inherent feature of the climate system.”

    He then goes on to endorse a mathematical approach that has its roots in the handling of aleatory uncertainty, inviting the reader to conjecture whether epistemic uncertainty has been overlooked entirely (as he does in his opening paragraph) or just fudged into his analysis. If the latter, there is at the heart of all of this an apparent belief that a probability distribution can capture Knightian uncertainty. All I can say is that it must be one thing to be aware of something and quite another to act accordingly.

    >”My understanding of the basic point that he is making with his papers is that the overall uncertainty means that we can’t be sure as to the outcome and, hence, that we really can’t rule out outcomes that may universally be regarded as catastrophic.”

    If this were true, then all Lewandowsky would be arguing for is the precautionary principle. And yet he claims he is deliberately avoiding the PP, preferring instead some simple mathematics that I suggest is not so much simple as simplistic. As I said, my article is not intended as a refutation of the PP, rather it raises concerns regarding the apparent failure to appreciate how epistemic uncertainty can undermine certain types of argument – such as Lewandowsky’s. Actually, he needs to stick to the PP because that, at least, doesn’t require a firm understanding of the implications of uncertainty’s foundations.

    >”Hence, we should take this into account when considering what to do (which seems to be roughly what your sources are suggesting too).”

    The supposed importance of taking into account that we can’t rule out catastrophic outcomes is not what my citations were about. I listed the sources since they were, in my opinion, offering a better way of determining the overall uncertainty. This they did by advocating non-probabilistic or nonstandard probabilistic approaches. They were not papers written for the purposes of supporting the PP any more than my article was written to denounce it. Whether or not any of the various authors then professed adherence to the PP is circumstantial. This is the bit were I think we’re going off topic.


  42. To defend Lewandowsky’s approach is to disagree with mainstream climate science, as represented by the IPCC AR5 technical papers. There is nothing wrong with doing so, of course, and indeed there is a small minority of ‘scientific catastrophists’ (i.e. they are actual climate scientists) who do exactly this, sometimes hooking their wagon to tipping-point theories and / or chaos theory, as well as minority approaches to uncertainty assessment. But by doing so they are essentially outside the scientific (not public) consensus, and indeed they claim that the IPCC is wrong by dint of being ‘too conservative’. However, it seems unlikely that AR6 will accommodate this viewpoint, hence they’ll remain outside the consensus. Lewandowsky is perfectly up-front about this clash with the IPCC / mainstream, for instance contributing his ideas as 1 of the 3 authors of a very clearly named paper: “IPCC is underselling climate change”. This is from last year, where the authors attempt to frame their beef as a failure of articulation rather than any failure of the science position (given they have to claim that something has failed within the mainstream process). Yet they explicitly state (in a quote from one of the other authors): “However, climatic uncertainties are nothing but an expression of the climate risks we face, and should inspire action rather than indifference.” Which exactly as John points out, does indeed summarize their argument as one that completely ignores (via ‘nothing but’) epistemic uncertainty.


  43. Andy,

    I guess this all boils down to what scares you most — the known risk (aleatory uncertainty) or the unknown risk (epistemic uncertainty). When Lewandowsky says things like “However, climatic uncertainties are nothing but an expression of the climate risks we face, and should inspire action rather than indifference” he is lumping the two classes together so that, at the end of the day, it is still just risk aversion we are dealing with. But we shouldn’t do this because the unknown risk carries less epistemic clout. That is what Didier Dubois was warning against. However, the problem is that, when the unknown risk entails catastrophic scenarios, the epistemology takes a back seat and emotion takes over – the perception of risk becomes the decision-maker. You could just go along with this, but you could also recognise the role of epistemic uncertainty and the fact that it is reducible. As it is reduced, the perception becomes more accurate. Consequently, I see nothing wrong with the reduction of epistemic uncertainty as a strategy for risk management, though Lewandowsky thinks this a scam. He thinks the ‘true’ risk lies in the epistemic hinterlands.

    Liked by 1 person

  44. John,

    “However, the problem is that, when the unknown risk entails catastrophic scenarios, the epistemology takes a back seat and emotion takes over”.

    Did you misspell ‘epidemiology’? 🙂

    Liked by 2 people

  45. “However, the problem is that, when the unknown risk entails catastrophic scenarios, the epistemology takes a back seat and emotion takes over – the perception of risk becomes the decision-maker.”

    Indeed, as has long since occurred in the public domain. Inclusive of most of its authority sources such as political leaders, business leaders, religious leaders, economists, medical leaders, celebrities and influencers, whose calls to action are almost universally framed in highly emotive narratives of certain global catastrophe, so revealing the nature of their own commitment.

    Liked by 1 person

  46. Slightly relevant and prompted by the mention of fat tails, particularly in ECS. For the uninitiated the right skew of ECS probability distributions might seem a little strange. Why should such distributions not be normal? This perturbed me a bit when I first saw them. The fat tails meant that disaster could not be entirely discounted, even when the most likely value for ECS was in a tolerable range, or even net beneficial given the likely course of CO2 concentrations.

    The skew in ECS probability distributions and their fat tails arise as a natural consequence of the mathematics.

    ECS = 3.7*dT / (dF – dQ)

    Where 3.7 is the radiative forcing from doubling CO2, dT the measured temperature rise, dF the measured forcing change, & dQ the transfer of energy to the oceans. It’s actually easier to see what’s going on if you refer instead to TCR, the transient climate response, which ignores the ocean as follows:

    TCR = 3.7*dT / dF

    The change in temperature and the change in forcing are both estimates, and have probability distributions associated with them. (Maybe normal.) When you divide one distribution by the other, you get a ratio distribution, which inevitably has a fat tail. The tail gets fatter still if there is more uncertainty in dF than in dT, and dF is quite uncertain (it has a negative component due to aerosols, which to some sceptics is notoriously too large, reducing dF overall and leading to higher ECS/TCR). The point I am groping towards is that you can start with (as default) symmetrical distributions in the variables being estimated, and end up with a skewed distribution in the one being calculated.

    A long time ago (5 BC – before covid), Nic Lewis discussed the issue of fat tails at Climate Audit, in relation to Bayesian inference. His results provided evidence that the fat tails ought not to be so fat, and his energy budget stuff shifted the median of ECS seemingly firmly into the safe zone. Reading the comments back now I see one from Richard and one from myself – mine relating to the updated aerosol forcings that, by increasing net dF, seemed to “officially” shrink the estimate of ECS. In some ways this world is very different than it was in 5 BC. In other ways – climate alarm, despite a continual drip-feed of what one might term “sceptic hope”, this world is very much unchanged.

    Does wordpress automatically add tags these days? I guess I’ll find out in 3-2-1-


    Liked by 1 person

  47. The only thing that will be worse than a unilateral climate consensus policy will be a world-wide climate consensus regime.
    The climate consensus, like its ideological enforcers, is anti-scientific, anti-democratic and offers as cruel an imperialism as has ever been thought of.


  48. Hunterson7,

    I appreciate you sharing your views on the politics of the situation, but I’m afraid I am determined to stick to the nerdy issue of propagation of uncertainty and the importance of avoiding an entirely probabilistic approach. I’m surprised that ATTP hasn’t responded yet to my most recent efforts to explain myself. Even now, I read them back and feel I could do better. For example, I should perhaps explain where my pre-occupation with the false precision implied by a probability distribution comes from. I think Scott Ferson, professor of risk and uncertainty at Liverpool University explains it well when he writes:

    “Risk analysis is a new field of enquiry that is distinct from probability theory. The two disciplines address two different problems. Since the time of Laplace, probability theory has concentrated on making good estimates for single quantities whose values might be varying or might be uncertain. The typical problem in probability theory posits the existence of an answer that is a single real number. Risk analysis, on the other hand, focuses on fundamentally different questions. In risk analysis we are usually dealing with populations, and various potential magnitudes of risk. We are generally interested in the entire distribution rather than a single number. Whereas probability theory uses a distribution to characterise the uncertainty about a single scalar number, risk analysis requires some comparable device to express the uncertainty about the distribution.”

    So I think it is my background as a risk analysts, of sorts, that warns me that not all uncertainty is captured by a probability distribution until one has also taken into account the imprecision of that distribution. To do that, one has to step outside of standard probability theory. I see no recognition of this fact in the proclamations of Lewandowsky et al.

    Liked by 1 person

  49. Jaime,

    >”Did you misspell ‘epidemiology’?”

    Maybe I have used the wrong word – in my anxiety to deny all expert opinion I get so confused. DesmogUK has got us all banged to rights though when it says:

    “A close look at commentary on both COVID-19 and climate change reveals significant crossover between unqualified voices casting doubt on experts recommending action. Why?”

    Fortunately, Lewandowsky can explain it all:

    “There’s nothing mysterious about this. I think COVID is just climate change on steroids in a particle accelerator. The same forces are happening: you have the inevitability of a virus which is the same as the inevitability of the physics. And opposing that you have politics which motivates some people to deny the inevitables [sic] and instead resort to bizarre claims.”

    Bizarre claims? What? Like, ‘Uncertainty as knowledge’?

    Liked by 1 person

  50. @JIT 04 Sep 20 at 10:18 am

    i remember that thread & the lively back & forth comments from/with Pekka Pirilä

    end of this comment by him seems relevant – https://climateaudit.org/2015/04/13/pitfalls-in-climate-sensitivity-estimation-part-2/#comment-758941

    “Many people have written recently in various blogs that Nic gives the impression of being much more sure of his conclusions than most (or almost all) of the experienced climate scientists. I agree with those observations, that’s the impression that I get as well. That kind of situation is not justifiable.”

    it’s just a blog comment but…


  51. to be fair to Pekka Pirilä he says at the end – Posted May 8, 2015 at 10:27 AM | Permalink
    In my view the real conundrum is related to the attitudes to risk and, how that is combined with the uncertainties.

    On one side are the views that
    – risk aversion does not affect essentially the conclusions,
    – we can concentrate on what has been proven in straightforward ways, and
    – do very little until the proofs of severe consequences are more direct than those presented by the present climate science.

    On the other side are views that
    – emphasize risks and worst plausible scenarios, and
    – think that it’s too late to act, when the straightforward proofs are available, because reducing significantly further warming takes several decades. Some significant damages may be effectively irreversible by that time.

    The second approach results in giving strong weight on the high tail of the possible values of climate sensitivity. That weight factor means that arguments used to suppress the high tail based on subjective choices (like the prior of Nic, and also my subjective prior) must be applied with great care. Although I do think that my justification for such a prior is sound, I do allow for the possibility that I have misjudged the situation.

    If I’m asked to recommend policies, I may tell that my own subjective judgment leads to a relatively low probability for the high values of the climate sensitivity, but that others disagree and I cannot prove them wrong. Policies must be decided taking into account the diversity of views – and risk aversion leads to an extra weight for the more pessimistic views.

    My own risk attitudes are closer to the second alternative, but I do think that many overplay that argument. An extreme approach built on the related precautionary principle makes it virtually impossible to analyze the alternatives rationally, and that in turn is very likely to result in ineffective and wasteful policies.

    There’s enough, or too much conundrum in all of this, and it seems impossible to escape that.”

    Liked by 1 person

  52. John, from the attempted fabrication and subsequent promotion of a scientific consensus to outright lies spread by so called ‘experts’ claiming ‘it’s worse than we thought’ based on a representative sample of one, to relentless media hype I’ve got a whole list of the stunning parallels bewteen Covid and climate alarmism, but this thread (and probably not this blog) is not the place to air them. But I will collect my thoughts and impressions, elsewhere.

    Lew, predictably, has obligingly made himself available as one of the more risible focal points for such nonsense and no doubt soon he will be publishing a study on the parallels between Covid and climate contrarianism and how moon-landing conspiracy theorists have infiltrated both ‘far right’ movements which are opposing the necessity for societal change in response to the twin crises of climate change and Covid.

    Liked by 2 people

  53. One dissimilarity of course is the time we’ve had to examine the two ‘crises’. A 32 year old being the twin of a six month year old is I’m sure the kind of detail Dr Lew could abstract out through his advanced maths but we can use it to our advantage:

    I saw Jaime’s comment here then looked at WUWT and listened to physicist Will Happer, with his estimation of an ECS of around 1°C.

    Happer also mentions Covid but, strangely, not so much the parallels. I continue to think the relationship between the two is complex and variegated and we should strive to reflect that.

    Liked by 1 person

  54. DFHunter,

    Yes, it is about attitudes to risk but it is also about attitudes to uncertainty. One of the most common misconceptions regarding uncertainty is to treat it as equivalent to risk. As a result, uncertainty aversion is miss-sold as risk aversion. The basis for this error is a cognitive bias, so it is ironic when a professor of psychology falls for it and then attempts to prove his errant idea by pushing naïve and simplistic attempts at uncertainty analysis – a field that is clearly outside of his sphere of expertise.

    As I have said a number of times on this thread, I am not arguing for or against the precautionary principle (not here, at least). Whether to base a decision upon uncertainty aversion or risk aversion is down to the decision-maker. However, when analysing a situation prior to making a decision it is very important that the analysis is cognisant of the subjectivity of the uncertainty and does not misrepresent it in any way. Subjective probability may seem a good idea but it carries that risk. As Didier Dubois wrote:

    “The idea is to propagate information as faithfully as is technically feasible, but on the other hand to introduce subjectivity at the decision-making stage only, so that decision-makers remain aware of the degree of imprecision regarding available information. Then introducing subjectivity factors at the final step would clearly only reflect the attitude of the decision-maker in front of risk, while the use of subjective probability inputs comes down to assigning part of the decision power to information suppliers.”


  55. Consensus enforcement:

    The Covid-19 ‘crisis’ has gifted us with a temporal microcosm of the stuff which has been going on in climate science for years and Lew has been at the vanguard of the movement to deligitimise sceptics via smear tactics.

    Last comment from me John. Promise. Don’t want to derail the thread.

    Liked by 1 person

  56. Jaime:

    The Covid-19 ‘crisis’ has gifted us with a temporal microcosm of the stuff which has been going on in climate science for years and Lew has been at the vanguard of the movement to deligitimise sceptics via smear tactics.

    The use of the term “covid deniers” in this context looks like it’s a really strong parallel.

    The BBC headline of course says: Tests ‘could be picking up dead virus’. Note the could.

    Uncertainty and how to treat it. The original poster can delete me for saying so but I think this is on topic.

    Liked by 1 person

  57. Richard,

    >”The original poster can delete me for saying so but I think this is on topic.”

    Actually, I quite agree that it is on topic. The term ‘denier’, when used properly, refers to the refusal to accept the existence of evidence, i.e. that which is evident. When the evidence is verified data that suggests (with uncertainty) a particular conclusion, one can use the term ‘denier’ to refer to someone who denies the existence of the data or the existence of the uncertainty, but not to someone who does not accept any particular conclusion drawn. I think the confusion is introduced when expert opinion (i.e. an authority’s acceptance of a conclusion) substitutes for evidence. Refusal to agree with experts is then assumed to be a denier’s refusal to accept that which is evident, rather than what it is – an insistence that the uncertainty should not be denied. Of course, as far as expert opinion is concerned, the uncertainty is epistemic. Epistemic uncertainty is not something to be brushed aside, settled by reference to CVs, or misrepresented as aleatoric. Nor should consideration of it be dismissed as ‘unscientific’.

    COVID-19 is providing plenty of examples of this confusion and I’m happy to see examples taken from any context.

    Liked by 1 person

  58. John: Just on ‘covid deniers’, as the later Wittgenstein would say “‘In most cases, the meaning of a word is its use”. Pretentious or what?

    I very much like how you’ve explained the word ‘when used properly’. Used improperly, though, there was a fateful sequence:

    1. holocaust deniers
    2. climate deniers (and its cognates: climate change deniers etc.)
    3. covid deniers

    I would argue that 2 got a great deal of its power from 1 but that 3 now rests on an established use in 2 and elsewhere and no longer really alludes to 1. It’s all pejorative of course.

    Thanks for fixing on the nerdy aspects of the underlying confusion.

    Liked by 1 person

  59. I agree Richard. The term ‘denier’ has now become common usage to insult anybody who questions the ‘official’ narrative and has lost much of its original gravitas in relation to the implied ideological connection between Holocaust denial and those who question ‘dangerous’ man-made climate change. Denier will soon become as worn out and over-used a term as racist probably. It’s not restricted to one side either. The consensus enforcers often get it thrown back in their face. I noticed someone using the term ‘herd immunity deniers’ today; in fact I think I might have used that phrase myself some time in the past! 🙂

    Liked by 2 people

  60. Jaime: I first noticed the term being turned with ‘deficit deniers’. However far back that was. I do try and watch out for such things. It’s rhetorical froth by now – of the kind you might get close to a dodgy chemical plant, due to its origins, at least in the hands of those who have never apologised for the initial, deadly smear.


  61. I’ve been thinking a lot about the arguments I have presented both within the post and subsequently in the commentary. Even now, after so many attempts, I think it may be worthwhile if I were to try just one more time to explain the problem, this time from a slightly different angle.

    Firstly, it is worth acknowledging that the mathematics behind Lewandowsky’s argument are valid as long as one accepts that a probability distribution is a complete expression of uncertainty, such that uncertainty can only be increased by flattening or extending the curve (the uncertainty can then be calculated using Shannon’s equation). Such a calculation is equally valid regardless of whether the probabilities are purely subjective or purely objective – Lewandowsky’s argument works equally well for both cases. However, a risk assessment based upon subjective probabilities does not carry the same epistemic clout as one that is based upon objective probabilities. Therefore, the decision-making importance of a probability distribution depends not only upon the shape of the curve but also the pattern of epistemic uncertainty that lies behind the calculation of the various points along the curve. For example, an extension of the higher range that relies heavily upon expert guesswork does not carry the same evidential importance as an extension of the lower range when it is based upon objective measurements. Mathematical arguments that do not take this effect into account are therefore based upon a potentially false premise. Put another way, the ‘uncertainty as knowledge’ argument is undermined by the ‘ignorance as uncertainty’ effect. One can tell that this is a problem for a given case if one observes that reduction of epistemic uncertainty fails to result in a change of the shape of the curve, as has been the case with ECS.

    The issue of evidential weighting is well-understood in the field of safety-related engineering. Questions of confidence in risk calculations are central to safety-related systems development and it is the reason why so much effort has been undertaken to improve the cogency of safety cases. It isn’t good enough to reduce risks as low as reasonably practicable (ALARP); one also has to be as confident as reasonably practicable (ACARP) that this reduction has been achieved. It’s all about the evidence of nature versus the nature of evidence.

    I think I may have just come up with the title for my next post.

    Liked by 1 person

  62. You still a proud member of XR?

    Technically, there is no way to answer this leading question. This is because I’ve never been a member of XR, proud or otherwise, so can’t really answer a question framed as “are you *still* a member?”


  63. So you didn’t sign a letter along with other like minded academics in support of XR?


  64. So you didn’t sign a letter along with other like minded academics in support of XR?

    I don’t recall having done so. It’s possible that I’ve forgotten, but I don’t sign very many such letters. If you think I did, feel free to point out where. I obviously think that we’re not doing enough about climate change, so broadly support what XR is trying to achieve (at least in the sense of them promoting more ambitious climate action). I don’t, however, agree with all of what they present, and I don’t agree with all of what they choose to do.


  65. Oh, that’s very good ATTP: You support XR’s aims, but not its methods (well not all of them anyway). And as for some of its avowed aims, like changing society undemocratically, well those we don’t speak about. Anyway your mind is failing and you can’t remember things, like signing letters. I’m sure someone here will rush to your rescue and remind you. Or perhaps your cronies on your website would oblige.

    Liked by 1 person

  66. Alan,
    Charming, as usual. I’m sure someone would rush to my rescue (for some odd definition of rescue). I simply do not recall signing letter in support of XR, but it is possible that I have and have simply forgotten.


  67. It took a little searching, but here it is:



    ” If you’d like to join us at future events, or to find out more, please join our Scientists for Extinction Rebellion Facebook group, follow us on Twitter @ScientistsX or email scientistsforxr@protonmail.com.”

    Signatory no. 11: “11. Professor Ken Rice, Institute for Astronomy, School of Physics and Astronomy, University of Edinburgh”

    Liked by 3 people

  68. The letter ended with these paragraphs:

    “As scientists, we have an obligation that extends beyond merely describing and understanding the natural world to taking an active part in helping to protect it. We note that the scientific community has already tried all conventional methods to draw attention to the crisis. We believe that the continued governmental inaction over the climate and ecological crisis now justifies peaceful and nonviolent protest and direct action, even if this goes beyond the bounds of the current law.

    We therefore support those who are rising up peacefully against governments around the world that are failing to act proportionately to the scale of the crisis.

    We believe it is our moral duty to act now, and we urge other scientists to join us in helping to protect humanity’s only home. “

    Liked by 1 person

  69. @Mark, thanks for the chuckles. It strikes me as odd that our rebellious friends demand governments control other people’s lives. Why, if their message is so urgent & so true, do they not merely ask other people to make the necessary changes voluntarily? I will never agree with anyone who tries to impose their will on me. If they can persuade me of the urgency of the situation, & I do things voluntarily, that is different.

    I don’t think it is wise to support direct action to force big brother to bully the public.

    Meanwhile the emissions of the climate assembly have been published, according to Harrabin at the BBC. His summary portrays it as an utter embarrassment, stuffed with ideas that two sixth formers could have jotted down on the back of an envelope in ten minutes. They are, unsurprisingly, in favour of things that won’t help (wind power) and against things that would (nuclear).

    108 volunteers, umpteen hours, experts, facilitators – and I doubt they wrote the 550 pages themselves – and by Harrabin’s account, it achieved exactly nothing.


  70. Meant to add: we need ideas to be generated by smart people, on which the ordinary people can then be educated, and vote on. That’s why I signed the petition to hold a referendum on net zero, which was about 5% of the way to getting a debate in parliament, and has now probably expired.

    Liked by 1 person

  71. Mark,
    Thanks. My apologies, I had indeed forgotten about that. I signed that one because I agree with what is said in the letter. I think I got demoted rapidly down to signatory number 1210 (at least that seems to be where I am on the list that I’ve just found).


  72. aTTP, thanks for the fair and honest climb down.

    I was surprised at the time, and I remain surprised that you, and so many other scientists signed up to this:

    “We believe that the continued governmental inaction over the climate and ecological crisis now justifies peaceful and nonviolent protest and direct action, even if this goes beyond the bounds of the current law.”

    That is basically an endorsement of XR and their often illegal behaviour, IMO. Also IMO, XR are a bunch of self-entitled science deniers, so to see scientists supporting them is strange to me, to say the least.

    Liked by 1 person

  73. I should have added that the idea that there has been “governmental inaction over the climate” all over the western world is simply not true. The letter also says “We therefore support those who are rising up peacefully against governments around the world that are failing to act proportionately to the scale of the crisis.”, but I don’t see much rising up against the CCP’s lack of significant climate action in China, do you?

    Liked by 1 person

  74. Speaking of China (sorry John, I’m taking this off topic I fear):

    “China isn’t Europe’s ‘partner’ on climate
    Treating Beijing as reliable undermines Europe’s ability to hold it to account.”


    A very good article, IMO, grounded in reality, unlike most of the nonsense spouted by climate hysterics who think the Paris Agreements work and who seem to think that requiring only the developed world to reduce GHG emissions might actually achieve something in relation to climate change.


  75. ATTP, you really don’t lie very well.
    And every premise in your backing of XR, like your “forgetting” to sign it, is also a lie.
    Government is naction on “climate change”? A literal lie.
    A “climate emergency”? A literal lie.
    Backing a group that wants to impose societal destruction in order to address “climate change”? A manifestation of a complete lack of reasoning.


  76. Mark,
    Climate change politics are as phony as the crisis climate change policies purport to address.
    There is no surprise that it is going down the wide highway of chaos.


  77. Hunterson7,

    I am sure you are not alone in wondering how anyone could forget that they had signed a document that advocated breaking the law but, as moderator for my own post, I would ask that the individual be given the benefit of the doubt rather than accuse them of lying. That said, you are entitled to your opinion no matter how severe.

    Liked by 1 person

  78. So to bring this back on the topic of just how many ways Lew can find to be wrong, didn’t he also sign off as one of the execrable academics supporting XR?

    Liked by 1 person

  79. John,
    I apologise to you for intervening in a much less than diplomatic way into the dialogue you have developed with ATTP. It was uncalled for. Your excellent essay on Lewandowsky and your success in getting Kwn to open up a bit are both remarkable.

    Liked by 2 people

  80. Hunterson7,

    This is not the first time I have invited discussion here regarding Lewandowsky’s take on uncertainty analysis. See, for example:


    Both then and now, the debate was something of a damp squib – I think I am struggling to get Ken to take my concerns seriously. He appears to genuinely believe that Lewandowsky’s argument is airtight and I suspect he doubts that the average Joe on the internet would have the necessary background to question it with any credibility. In an attempt to establish some respect, I have previously tried claiming competence born of professional experience, but that has backfired spectacularly. To a certain extent that was because I have been less than diplomatic in the way I have handled matters but, even so, I suspect the real reason is that there has never been anyone on the other side of the debate prepared to give a sceptic the benefit of the doubt. Stephen Mosher called it ‘unsubstantiated credentialism’, or, as Willard put it when talking to a fellow traveller over at ATTP:

    “He’s supposed to be big in risk management, but don’t sweat it.”

    It goes without saying that I have never claimed to be “big” in risk management, only that I have sufficient expertise to be treated seriously. I suggest that maybe, when it comes to matters of risk and uncertainty, it is someone with a professional background in the theory and practice of decision-making under uncertainty who should be considered an expert ahead of a cognitive psychologist turned climate activist.

    Anyway, the reason why I furrowed my brow earlier is because I fear impugning Ken’s integrity will only result in you being labelled as just another ‘nasty’ denier. But remember, this would be coming from people who are just as willing to impugn the integrity of others. So don’t sweat it.


  81. @John re: a suggestion about your next post. Some way above I mentioned the standard ECS equation and how it’s easy to obtain an estimate of ECS with a measurement of dF and dT (actually TCR, but). I then went on to say that after making a perhaps unwarranted assumption about the distribution of said estimates of dT and dF, you end up dividing one by t’other and getting a ratio distribution with a fat tail. This of course is uncertainty of the dice-rolly kind, except when it comes to the assumption of the distribution (is it valid to assume symmetry?).

    What I would find helpful (by implication, others also) if you ever carry on discussing uncertainty re: climate would be a post on other forms of uncertainty as they pertain to the estimation of ECS. Is the equation itself up for grabs? Maybe the value of ECS is not a constant, and varies according to temperature? There are a variety of ways of coming up with estimates of ECS, some more robust than others (energy budget studies must trump computer models). Is it even worth trying to combine such estimates?

    The final pdf is a product of a series of assumptions as well as some measured data. What pitfalls lay on the route? Some of these uncertainties may be Rumsfeldian, but concrete examples would be useful.

    Finally, if there was ever an ECS pdf you were happy with, how would you quantify the associated risk? Again the fat tail wags mightily, because here we have the region of (presumably) rapidly ramping up damages.

    To put it another way: what do you say if BoJo phones up and says, “John, hi, some guy told me you’re big in risk management – any idea how to work out what risks we face from CO2 if the world carries on as we are?”

    There is of course undue certainty all around. On lamp posts around here XR propaganda claims that “The Science” says we are due for 4 C temperature rise, storms, looting, refugees, wildfires… I can’t remember all of it but that’s the gist. If people really believe this stuff, we’re in trouble. Me, the only thing I find mildly concerning about global warming is the potential for stronger hurricanes. Everywhere else I’ve looked, I’ve drawn a blank on more than mild peril. The question is, do I believe that because I can’t handle the truth?


  82. John,

    He appears to genuinely believe that Lewandowsky’s argument is airtight and I suspect he doubts that the average Joe on the internet would have the necessary background to question it with any credibility.

    I don’t think it’s airtight. If anything, I think it’s quite simple. I think you’re finding some complicated reason to criticise an argument that I think is quite simple. We’re trying to decide how much we should do to deal with climate change. In a sense we’re considering how much should we do to limit future emissions (given the roughly linear relationship between emissions and warming, and the likely non-linear relationship between warming and impacts). The more uncertain we are about the resulting warming (and hence impact) the more we may want to do to limit emissions (i.e., if there is a chance that that climate sensitivity is very high, then maybe we should limit emissions more than if we could – with high confidence – rule out the high climate sensitivity values).

    As far the latest kerfuffle. I should have known better to respond to a bad faith question from an anonymous commenter. They managed to call me a liar, which was probably their goal all along.


  83. John,
    By the way, the believer assertion that lower CO2 levels will reduce the risk level is utter hogwash.
    It is like saying if more people go to mass regularly there will be less plague.
    There is no historical evidence at all to support the claim that lower CO2 = more benign climate or weather.


  84. ATTP,

    I fail to see what is so complicated. The “uncertainty as knowledge” proposition is undermined by the “ignorance as uncertainty” effect. As a result, the assertion that greater uncertainty implies higher risk fails by not fully taking into account the contributions of epistemic uncertainty. This is not over-complicating things, nor is it a pedantic point. It is actually a point of profound importance.

    As for the kerfuffle, I suppose you will just have to accept it as the inevitable consequence of peddling a truth that is so unbelievable.


  85. John,
    Hold on, whatever uncertainty you consider, we cannot (with high confidence) rule out high values of climate sensitivity. If you could do so, then you could publish something making this argument.

    I suppose you will just have to accept it as the inevitable consequence of peddling a truth that is so unbelievable.

    It was almost certainly inevitable.


  86. Jit. Surely the main problem associated with global warming (from whatever cause) is sea-level rise, given the multitudes that inhabit shoreline cities? Other severe problems are of entirely our own manufacture as we try to control climate by reducing CO2 emissions.


  87. ATTP,

    >”Hold on, whatever uncertainty you consider, we cannot (with high confidence) rule out high values of climate sensitivity.”

    But, with all due respect, you can’t accuse me of claiming otherwise. This is just the precautionary principle and, as I have said several times, this post is not an argument for or against the application of the PP. What I am saying, however, is that Lewandowsky’s argument that uncertainty is ‘actionable knowledge’ is unsound because he makes the assumption that all the uncertainty can be captured in the shape of a probability distribution. There are plenty of experts out there will tell you that probability is not good at delineating incertitude from variability and that this is a major problem for uncertainty analysis. Furthermore, Lewandowsky’s ordinal argument (which, remember, he is proposing as an alternative to the invocation of the PP) only works when the probabilities involved are uniformly objective or uniformly subjective. This is the only way he can get away with saying that increasing uncertainty necessarily results in an increase in risk. However, my main beef is the apparent treatment of uncertainty as something we don’t own but is out there in nature for us simply to accept. If we cannot (with high confidence) rule out high values of climate sensitivity then what are we to do about that? Proceeding on the assumption that the worst case applies is only one option, and not a particularly robust one.


  88. @Alan that must be a bait coming from a geologist…

    If the water is at our toes now, in 2100 it will be half-way up our calves. Any civilisation that can’t cope with that does not deserve the name. I see no problems for barrier islands or coral atolls or deltas (except those caused by unwise human interventions).

    We can deal with even the most excitable rates of sea level rise.

    Mind you, we ought, in theory, be able to plan for and deal with the storm surges caused by stronger hurricanes as well.


  89. Lewandowsky’s argument that uncertainty is ‘actionable knowledge’ is unsound because he makes the assumption that all the uncertainty can be captured in the shape of a probability distribution.

    I’m surprised you say this. In the opening article Lewandowsky, Ballard & Pancost explicitly say

    In some cases, these changes can be mathematically constrained by a wider probability distribution, as for example in a multi-step assessment of attributable risk that has been applied to the floods in the UK in the autumn of 2000 [14]. In others, deep uncertainty makes it difficult to assign any meaningful probabilities [15]. In yet other cases, we cannot even preclude completely unexpected consequences that are sufficiently surprising to sit outside conventional frameworks for expectation [16].

    Let me see if I can explain my understanding of their basic argument. Let’s consider the TCRE range, which is 1K to 2.5K per 1000 GtC. In other words, we’d probably warm by between 1K and 2.5K if we emit a total of 1000 GtC. Let’s also assume that we broadly agree that we really want to avoid warming by more than 4K. If the TCRE range was constrained with high confidence (i.e., a negligible chance of warming by more than 2.5K per 1000 GtC) then we could emit a total of 1600 GtC (i.e., about 2.5 times more than we’ve emitted to date) and be confident that we won’t warm by more than 4K. However, the TCRE range is (IIRC) a likely range (in IPCC speak, this means roughly a 66% chance of failiing within this range). Given this, if we really want to avoid warming by more than 4K, then we might want to end up emitting less than if this range were much more tightly constrained.

    The basic argument doesn’t require an exact mathematical description of the uncertainty, just a recognition that the less certain we are about the upper end of the range, the more effort we *may* want to put into emission reductions.


  90. ATTP,

    >”I’m surprised you say this.”

    No, what is surprising is that it is true. Despite the authors’ throwaway line in the introduction regarding unknown probabilities and deep uncertainty they still proceed to advocate arguments that only work with known probabilities and tractable uncertainty. Throughout the Freeman et al paper, and the Lewandowsky paper it builds upon, you will see nothing but arguments based upon the properties of probability distributions and well-behaved damage functions. Not an imprecise probability in sight.

    I did at one point think to include your quoted passage within my article, but only to embarrass the authors still further. Why mention deep uncertainty and then ignore it completely? I guess it is because there is no conceivable way of developing a notion of ‘uncertainty as knowledge’ when the uncertainty is deep.


  91. Jit,

    I’m flattered that you think that I might know enough about climate science to be able to write about specific sources of uncertainty in ECS calculations, but I don’t. My concerns are more generic than that and stem largely from experience in a completely different field, in which decision-making under uncertainty/risk does not automatically result in the invocation of the precautionary principle – in disguise or otherwise. I will, however, offer the following link, which does at least cover some of the areas you are interested in. There are many ways of surveying the subject, and this is as good as any:

    Click to access Uncertainty%20in%20Climate%20Modelling.pdf

    In the meantime, I have been thinking of taking a long rest from writing articles and now seems a suitable time, although it is perhaps a shame that I am doing so after the last ship I launched sank in the harbour. I’d prefer to be leaving on a high.


  92. @John my request was a bit dumb, not to mention presumptious – it’s just that I find arguments easier to parse when they are accompanied by a concrete example. The example need not have been ECS, but it did feature in the OP. I’ll read the pdf.

    I enjoy reading your articles, even when I Warnock them. As to sinking in the harbour, that alas applies to more-or-less anything I’ve written, whether fiction or fact. A long time ago I realised that I would be compelled to continue to write, even if marooned on a tiny island with only the sand and a stick for paper and pen. I hope you’ll continue too, tho’ if you have something better to spend your time on, no-one would begrudge it. (He said speaking for people he has no right to.)


  93. Jit,

    Thanks for that. I have to admit my main motive for contributing to this site was always the enjoyment of writing (there was never any prospect of changing the world, let alone changing ATTP’s opinions). Much of what I have written in the past has never seen the light of day, nor will it ever. Nevertheless, it was fun to write. The weird thing is that I never have been able to discern the key to success. There seems to be no relationship whatsoever between my work’s popularity and my self-assessment of its merit.

    Anyway, I still expect to be contributing comments and, in that vein, I offer the following link:


    It’s an example of the sort of forum discussion I had dreamt of encouraging on this site. My aspirations were, of course, wholly unreasonable!


  94. Jit,

    Looking out for a concrete example for you I found this:

    Click to access Bedi-A-2013-PhD-Thesis.pdf

    It is about rock engineering rather than concrete, but I thought that would be near enough 🙂

    As with any good PhD thesis, it provides a good level of background theory before moving on to the application. As such, it looks to me like a damned good piece of reference material.



  95. John,

    (there was never any prospect of changing the world, let alone changing ATTP’s opinions).

    I think we disagree less than you seem to think (I’m certainly not arguing that we shouldn’t consider all the uncertainties). However, I’m genuinely failing to see how doing so would substantively change the basic point – the less we can constrain the chance of catastrophic outcomes, the more we may want to do.


  96. ATTP,

    To be fair, my quip was very much tongue-in-cheek. Even so, there is still a fundamental difference in the way we are thinking about this. I note with interest that you write ‘constrain the chance’. This is, of course, very much the language of standard probability theory at the gaming table. However, when the nature and extent of uncertainty precludes even the determination of the chances, what then do we constrain?

    I sincerely believe that you are advocating uncertainty aversion and I believe I appreciate the logic behind that. However, I draw the line at arguments, such as Lewandowsky’s, that are inappropriately using precise probability distributions to claim that it isn’t uncertainty aversion we are dealing with but risk aversion after all. I think it is important that we do not delude ourselves because, if it is uncertainty aversion that we are dealing with, then the options for the way forward will require both robust and precautionary decision-making. Uncertainty management is a different discipline to risk management. The latter may be all about constraining chances but I do not believe the former is.


  97. Okay, I still feel that you’re mis-interpreting, or over-interpreting what is really being suggested. Or, we have a different view as to what is actually happening. The basics are straight-forward. Our emission of greenhouse gases is increasing atmospheric GHG concentrations. This leads to warming of the climate system, typically presented as an increase in surface temperatures. This will have impacts (sea level rise, increase in the frequency and intensity of extreme weather events, the loss of some ocean and land ecosystems, etc). We expect the impacts to increase non-linearly with warming (i.e., the impact of warming by 1C will be much smaller than the impact of warming from 1C to 2C). There is a level of warming we probably want to avoid (we probably can’t constrain this precisely, but most researchers seem to agree that warming beyond something like 4C will probably have substantial negative impacts).

    If we had a really good understanding of how our emissions translated into increases in atmospheric GHG concentrations, warming, and then impacts, we could be more confident in what we should do than if we’re not sure how our emissions will impact concentrations, warming and then impacts. Given the latter, we may want to err on the side of caution and do more than if we were more confident in our estimates.

    What we’re not uncertain about is that emissions increase atmospheric CO2 concentrations, which then produces warming, which will then have impacts (and that these will become increasingly distruptive, the more we emit and the more we warm).


  98. ATTP,

    We both appear to be repeating ourselves now, so perhaps it might be a good idea to wrap this up. I’ve been thinking about what my concluding remark might be, and it goes a bit like this:

    I get what you are saying but you appear to be simply advocating a precautionary approach, and I am not trying to argue against that. The title of the article is ‘No-one Does Wrong Quite Like Lewandowsky’, and so it is an argument against what he is saying. If you go back to his original paper on the subject, he accepts that both a CBA approach and the precautionary principle are problematic before saying:

    “We therefore suggest that neither CBA nor the precautionary principle are sufficient to resolve the dilemma posed by figure 1… Our approach rests on developing ordinal constraints – i.e. constraints of the form “greater than” or “lesser than” – that derive from the functional form of the mapping between uncertainty and outcomes.”

    I am arguing that he fails in that enterprise because he misunderstands the “functional form of the mapping between uncertainty and outcomes.”

    And that’s all there is to it – apart from the suggestion that this is the sort of error that is made possible by failing to properly delineate epistemic and aleatoric uncertainty before attempting to propagate uncertainty. This is a common theme outside climate science and so I see no reason why it shouldn’t feature more within climate science.

    If you really want to know what I think about the precautionary principle, we need to talk more about risk efficiency and the principle of globally equivalent risk. I don’t want to do that simply because that would be taking me off topic for today.

    Liked by 1 person

  99. ATTP, other than the trivial fact that CO2 acts as a ghg, the rest of your assumptions about the outcomes are not supported by data. No loss of ecosystems, no slr changes, no increase in extreme weather, and most certainly no climate emergency.
    You cannot show the data to support that, only projections and rewrites of historic data.
    So I stead of confronting those inconveniences you, and Lewandowsky, dig in and distort everything around the issue, including how to measure risk.
    Your signing up with XR indicates how far you have been willing to go to distort everything to make your wildly simplistic excuse for reasoning stand up


  100. ATTP,

    I do not wish to get into any detailed critique regarding Hunterson7’s comments. However, I doubt whether he would be surprised to hear me say that I would rarely be as categoric as he. For example, I certainly wouldn’t say ‘no data’ and wouldn’t be as dismissive of your concerns. Also, I wouldn’t put you in the same camp as Lewandowsky when it comes to precautionary thinking. I think your position has more merit, but I would still take issue with it for reasons that have nothing to do with Lewandowsky’s thesis. That said, I should add that I disapprove of any scientist who is willing to publicly support the XR manifesto. They engage in a misplaced certitude that makes Hunterson7 look positively undecided in comparison. Furthermore, I would never give encouragement to a fundamentally undemocratic group, no matter how expedient it may seem for a ‘greater good’.

    There now, you have succeeded in getting me more off-topic than I could ever have imagined.


  101. John,
    Fair enough. I’m not trying to catch you out, though. I just don’t think that it’s really possible to have a serius discussion about this issue without discussing our current understanding of emissions, concentrations, warming, and impacts.


  102. Jit. “Alan that must be a bait coming from a geologist…”

    Coming from a geologist is the fact that during the last interglacial sea levels were 5 metres higher than they are now, yet temperatures were not as high as many climate projections imply. A major collapse of floating ice shelves allowing fast flow into the sea of land-based glaciers could easily cause a rapid rise of sea level of a metre by 2100. The melting rate would depend on how far equator-wards the floating ice moves. It’s not the physical rise in sea-level that is significant, but the increased effect of surges, tidal ranges or (especially) erosion rates. Not every coastal country can afford to do a Japan and seawall all its soft coasts.


  103. @Alan I referred there to what I consider a key sceptic principle, borrowed from 18th century geology, which is that tomorrow is going to be a lot like today and today is a lot like yesterday was. I look at the present rate of sea level rise, unchanged for a hundred years, and see no reason in speculative models or ifs, coulds, buts and maybes to subject that gentle slope to catastrophic swerves.

    To get to a metre in 80 years means a 4-5 fold increase in the rate of rise, starting tomorrow. If you want to wait a decade or two, your swerve necessarily has to get wilder.

    Liked by 1 person

  104. Jit. Uniformatarianism is all very well (and rescued geology from Werner’s Neptunism) but there always have been sporadic catastrophic events and present-day rates of change are not necessary immutable. Today is not necessarily a firm guide to tomorrow.


  105. ATTP,

    >”I just don’t think that it’s really possible to have a serious discussion about this issue without discussing our current understanding of emissions, concentrations, warming, and impacts.”

    Yes, of course, absolutely agreed. But by the same token, I’d hope you would agree, a serious discussion of a subject that entails decision-making under uncertainty would not be possible without a consideration of what uncertainty is and how it should be analysed. When you have people like Peter Gleick seeming to dismiss epistemic uncertainty as ‘colloquial’ and unscientific, and Lewandowsky et al concluding that uncertainty is ‘actionable knowledge’, but basing their views upon an analytical approach that was designed for aleatory uncertainty only, I think there is some justification for raising concern. Incidentally, I see that Lewandowsky and Freeman were up to the same mathematical wizardry when they collaborated on this paper:


    Interestingly, on this occasion they were also collaborating with every sceptic’s bête noir, Michael Mann. This is the guy who, speaking to the 2019 IGCC Summit in Sydney, denounced climate sceptics as “villainous and immoral”, adding:

    “Uncertainty is not our friend. We have to weigh on the side of assessing a potentially far greater risk than some of the simple linear models that are used might suggest. We really need to be thinking about the possibility that the impacts could be far greater than the models that are usually used to assess impacts and to assess risk.”

    If I am to be denounced by someone who lays claim to a superior understanding of uncertainty, I’d rather have hoped that his claim were true.


  106. John,

    But by the same token, I’d hope you would agree, a serious discussion of a subject that entails decision-making under uncertainty would not be possible without a consideration of what uncertainty is and how it should be analysed.

    Yes, but I don’t think you can start this discussion without first agreeing (or discussing) the basics. If we don’t agree in the fundamentals, then a detailed discussion about uncertainty, and decision making under uncertainty, isn’t really possible.


  107. ATTP,

    Now it’s my turn to be surprised by what the other person has to say. Of course it is possible to discuss the fundamentals of decision-making under uncertainty without having to agree upon the basics of a particular field of application. What I am trying to introduce into the debate is a concern that is not domain-specific — it is therefore the very definition of fundamental. Furthermore, it is the likes of Gleick, Lewandowsky and Mann who started all of this by pontificating upon the fundamental nature of uncertainty without first doing their homework. Their arguments would be just as applicable, and just as flawed, had they been applied to rock engineering. Added to that, their confidence in the correctitude of their thinking about uncertainty is so high, they feel justified in dismissing those who hold differing views about it as ‘villainous and immoral’. That alone makes the debate fundamental in my book.

    None of this detracts, of course, from the fact that the basics of the physics of the situation need to be agreed before drawing conclusions in a particular example. But when the extent of the uncertainty includes epistemic uncertainty regarding the basics of the physics (i.e. model structure), one cannot dismiss, as a detail, how such uncertainty should be propagated.

    I think the bottom line is that you want to talk about what you want to talk about, and the same goes for me. I have no wish to downplay the importance of your concerns and I would appreciate some reciprocation. If you are still hoping to convince me that I am being pedantic and quibbling over non-fundamental detail, then rest assured that you are flogging a dead horse.


  108. John,
    Okay, I agree one can have a discussion about the fuindamentals of uncertainty without needed to have a discussion about the fundamentals of this particular issue. However, the reason I’m focussing on the latter is because I was hoping it would help me to understand your views on the significance of uncertainties in this context.


  109. ATTP,

    Okay, well maybe the following may help to explain why I see the correct delineation of epistemic and aleatoric uncertainty as a key issue. The other day I was reading this interview with Gavin Schmidt and Kate Marvel, regarding the recent study that reduced the range of values posited for ECS:


    What particularly caught my eye was the reason why the range had reduced. Instead of attempting to create a probability distribution that took into account the full gamut of evidence types, they concentrated upon observation of climate history. Specifically, estimates based directly upon climate model outputs were not used. In Schmidt’s own words:

    “Climate models help frame the questions we are asking and can be examined to see how climate patterns in space and time connect to things we can directly observe. But we know that climate models have a lot of uncertainty related (for instance) to cloud processes, and so we didn’t use them directly to estimate sensitivity. You could, however, use our results to assess whether a climate model has a sensitivity that is within our independently constrained range.”

    By doing that, the team removed a significant source of epistemic uncertainty. This is a good thing since no decision under uncertainty is improved by increasing epistemic uncertainty. Sure, there was still a lot of epistemic uncertainty left in the measurement problem relating to paleoclimate, but even there the emphasis was upon using as much coherent, data-driven evidence as was possible to reduce the epistemic uncertainty. The other advantage in their approach was that the use of aleatoric methods to analyze the uncertainty increases in pertinence once the problem becomes focused upon a genuine measurement issue. The result of all of this was twofold: a) the ECS range was reduced, at long last; b) the confidence in the range calculation was not as undermined by epistemic uncertainty as it had previously been.

    This is a far cry from trying to persuade anyone that uncertainty is actionable knowledge. Had it been left to Mann, we would still be demonizing anyone who did not sign up to his exhortation to be “thinking about the possibility that the impacts could be far greater than the models that are usually used to assess impacts and to assess risk.” He was right about one thing, however. Uncertainty certainly wasn’t his friend – as can now be seen by the observational approach ruling out his cherished, fat tail fantasies.


  110. John,
    I think one has to be careful of what they mean by “not using climate models directly” means. The paper is 184 pages long, so I haven’t read it in detail. They are, however, clearly using GCMs. What I think they’re not doing is simply running a whole bunch of GCMs to equilibrium and then using the output of this to estimate the ECS. The problem with doing something like this is that you then don’t account for how well the different models match obervations (for example). They’re, instead (as I understand it) using various observations to constrain the ECS, but this still involves using GCMS (and other types of climate models). To me, this seems like a perfectly reasonable method for trying to constrain this. It doesn’t (as they themselve say) the higher levels of climate sensitivity, but it makes them more unlikely.


  111. ATTP,

    Yes, I understood that. They also made it clear that their limited reference to the models had the effect of eliminating a source of uncertainty.


  112. John,

    They also made it clear that their limited reference to the models had the effect of eliminating a source of uncertainty.

    Indeed, but this seems to simply be what scientists always do to try and better constrain some parameter. How, in your view, does this relate to decision making under uncertainty?


  113. ATTP,

    Except that hasn’t been happening with ECS, has it? It has been stubbornly anchored to it’s original guestimate for 40 years. Odd, don’t you think? Perhaps that isn’t so surprising when there are scientist who believe that uncertainty is actionable knowledge. With such ‘knowledge’ at one’s fingertips, who needs real knowledge to justify action. What was it that Lewandowsky said about the desire to reduce uncertainty before acting? That was it, he called it a SCAM.

    Liked by 1 person

  114. The corruption of science that Lew peddles in the nonsensical push to redefine risk and evidence was presented much better many years ago.


  115. beththeserf,
    Please check the link to what appears to be a very interesting video. It appears to not be functional.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.