John Ridgway,
“There is something fascinating about science. One gets such wholesale returns of conjecture out of such a trifling investment of fact.”
Mark Twain, Life on the Mississippi
It is tempting to speculate what a resurrected Mark Twain would think of the current controversy surrounding the global warming debate. Some of the warnings being issued by the scientific community are certainly alarming, and I suspect that he would very much like to know the truth behind them. Moreover, the public have been informed that the science is settled and the time has come to take action. Nevertheless, even a superficial understanding of the challenges facing climatology is sufficient to appreciate that there are significant uncertainties still to be addressed. So, if he were alive today, I think Mr Twain might be wondering where the certitude is coming from.
The standard answer to this question is that the uncertainties have been evaluated and are deemed to be peripheral to the central point; we know it’s going to be bad and we are just uncertain as to how bad. However, the uncertainties to which I am alluding are too fundamental for this explanation to work. Instead, I suspect the problem is that climatologists are making predictions that cannot be readily falsified through appropriate experimentation. Knowledge gained from experiment is an important means by which epistemic uncertainty (that is to say, ignorance) may be reduced, and it is through such a reduced uncertainty that one would wish to achieve a convergence of opinion. But within climatology, consensus emerges principally through inference and disputation, in which logic and objectivity are in competition with rhetoric, ideology, policy and expedience. Significantly, it is through the sociology of science that one can establish certitude without having to reduce uncertainty.
Getting Back to Basics
No one would question that climatology is a science. Climatologists, therefore, benefit from the credence that any reasonable person would place in the proclamations made by scientists. This is why it seems so easy to label as ‘anti-scientific’ anyone who challenges the ‘overwhelming’ consensus amongst climatologists. After all, to quote Barak Obama, “The science is beyond dispute and the facts are clear”. The same assumption of unquestionable integrity lies behind the suggestion made by eco-psychologists of the University of South West England that anyone who questions the truth of Catastrophic Anthropogenic Global Warming (CAGW) should be treated as suffering from a mental disorder.1 Furthermore, the time wasted debating with such sceptics (or ‘Flat Earthers’, as Al Gore prefers to call them) could lead to critical delays in the introduction of necessary measures. Such scepticism, therefore, is every bit as irrational and immoral as the denial of the Holocaust, so there should really be a law against it, shouldn’t there? But before we get carried away (literally), we should remind ourselves why it is that we trust scientists, and ask if our understanding of where this trust comes from justifies being wary of the presupposed consensus amongst climatologists.
It is generally accepted that the principal strength behind the scientific method is the objectivity that results from restricting oneself to statements that are falsifiable, particularly through practical experimentation. At the very least, one should be making statements that include predictions that can be verified by reference to nature. This is how consensus should be achieved. Science is not a democracy – facts decide the issues, not a scientific electorate. But even in the purest of sciences one can occasionally find oneself in areas of speculation that are not obviously susceptible to the scrutiny of experimentation. When this happens it is almost inevitable that dispute prospers. After all, scientists are only human. If one adds into the mix a question of vital importance to the survival of the human race, then you can guarantee that politicians and the media will get involved, after which, the prospects of a calm, objective debate are virtually non-existent. More importantly, the lack of falsifiability sets the scene for the achievement of consensus by other means, resulting in a certitude that cannot be taken at face value. In the case of climatology, the first clue that the consensus may not be all it seems lies in its extraordinary significance. The consensus within climatology, we are told, reflects the achievement of a level of certainty that is unique within science. Apparently, climatology is the only scientific discipline in history for which the science has been settled!
Admittedly, the theory that CO2 is a greenhouse gas is easy to confirm in the laboratory, and the fact that, throughout industrialisation, there has been a significant rise in the amount of CO2 in the Earth’s atmosphere is readily confirmed by measurement. To this extent, the proposal that mankind is currently contributing to climate change is easy to accept. Nevertheless, this is not the issue. The question is whether current trends can be used to predict that there is a realistic prospect of irreversible and catastrophic environmental damage in the not-too-distant future.
Keeping in mind that a scientific proposition has to be falsifiable, is there a legitimate experiment we can perform to falsify the specific prophesies being made in the name of climatology? Well, I’m afraid the answer is no. The experiment would necessarily require taking a representative planet (planet Earth, for example), subjecting it to the sort of ongoing increases in CO2 that are concerning us, and then seeing what happens. Such an ‘experiment’ is underway, but it hasn’t been set up by the climatologists. As the 1988 ‘World Conference on the Changing Atmosphere’ put it, “Humanity is conducting an unintended, uncontrolled, globally pervasive experiment whose ultimate consequences could be second only to a global nuclear war”.
This is clearly not what we want to do, but in the absence of a responsible, intended and controlled experiment one has to be satisfied with theorising, based upon the current understanding of how the Earth’s climate system works. This entails field study, to garner information regarding current and historical climate change, combined with mathematical modelling to explore the range of future possibilities. Unfortunately, no matter how sensible this may be, it leaves a great deal of room for speculation.
Firstly, there is room to be cautious when attempting to discern the Earth’s history of temperature changes, particularly when this requires interpretation and manipulation of proxy indicators. This is an area mired in controversy; a point to which I will return later as we encounter the infamous Hockey Stick graph.
Secondly, there is plenty of room to be cautious regarding the nature of the mathematical models used to predict the evolution of the Earth’s climate system – a system that is known to be open, driven, non-linear, hugely complex and chaotic. The extent to which any credence can be placed in such models depends upon how well they capture the relevant factors at play and how realistically and accurately such factors have been parameterised. Furthermore, even for a comprehensive and accurate model, predictive skill rests upon the extent to which noise and the climate system’s dissipative forces can stabilize the climate on the large scale.2 Unfortunately, in view of the significant structural and parametric uncertainty within the models, it may be delusional to premise confidence levels upon the assumption of well-defined aleatoric uncertainty.3
None of the above, however, seems to have fazed the Intergovernmental Panel on Climate Change (IPCC). For those who required reassurance that the extant set of climate models were valid, the following assertion was displayed prominently within the IPCC’s Fourth Assessment Report:
“This confidence comes from the foundation of the models in accepted physical principles and from their ability to reproduce observed features of current climate and past climate changes.”
Well it is reassuring to know that the models are founded in accepted physical principles, though this is a bit like saying that a witness is reliable because it is a real person rather than a Harry Potter character – it is hardly something worth boasting about. As for the reproduction of current and past climate changes, I have two concerns:
Firstly, the fact that a model can reproduce the past is only impressive if it is the only conceivable model that can do so, and since a diversity of models can be made to fit the record through judicious ‘tuning’, this is clearly not the case.4 Consequently, matching the climate record is not the true metric for confidence in a model. Instead, confidence comes from the plausibility and legitimacy of the parameter values to which the modellers resorted in order to achieve the match. Even the IPCC itself concedes, “If the model has been tuned to give a good representation of a particular observed quantity, then agreement with that observation cannot be used to build confidence in that model.” So, is such a strategy common practice? Who knows? The modellers don’t publish their tuning strategies.
Secondly, surely it is a logical non sequitur to suggest that a model that performs well for the purposes of hindsight will necessarily be reliable for the purposes of making predictions. For example, a model that incorrectly accounts for the effects of water vapour and cloud feedback might not reveal such a weakness until the effects gain significance.
I’ll leave the last words on this subject to a statement buried deep within the detail of the IPCC’s Fourth Assessment Report: “What does the accuracy of a climate model’s simulation of past or contemporary climate say about the accuracy of its projections of climate change? This question is just beginning to be addressed…”. So it turns out that the IPCC was already confident about its models even though the critical question was only ‘just beginning to be addressed’. Am I missing something here?
But above all my concerns, the key point I would wish to make is that using a mathematical model to make a prediction is a poor substitute for the running of an experiment. Climatologists may refer to each run of a climate model as a ‘mathematical experiment’ but in so doing they appear to be fooling themselves. In any other scientific discipline such activity would go by the name, ‘hypothesising’. It’s not an activity that establishes facts of nature (as would a true experiment), it simply enables climatologists to better understand the character of their speculations.
Let There be no Room for Doubt
So far, all I have offered to explain the coexistence of uncertainty and certitude within climatology is a somewhat churlish and semi-informed suspicion that the scientists concerned are guilty of professional misjudgement; that they are placing too much trust in the scientific arsenal available to them. However, one should not lose sight of the fact that climatology has been heavily politicised and that ideologies, as well as scientific understanding, are at stake. It is under such circumstances that the debate may be corrupted. So, if evidence were to emerge that uncertainties have been deliberately downplayed, one might argue that such corruption has taken place. I entrust this argument to the following two examples:
The first example is the accusation that the IPCC has been guilty of a simplistic presentation of the science, and that the certainty declared in its executive summaries does not reflect the uncertainty expressed by the scientists consulted. This concern was raised as early as the first of the IPCC’s assessment reports, published in 1990. Whilst a working group set up to advise the panel pointed towards uncertainties, these were not reflected in the executive summary. For example, the following statement appeared in the body of the working group report: “A global warming of larger size has almost certainly occurred at least once since the end of the last glaciation without any appreciable increase in greenhouse gasses. Because we do not understand the reasons for these past warming events, it is not yet possible to attribute a specific proportion of the recent, smaller warming to an increase of greenhouse gasses.”
Notwithstanding such reservations, the report’s executive summary was unequivocal in its conclusions and advised that the evidence for the potential damage from AGW was so strong that it, “requires immediate reductions in emissions from human activities of over 60% to stabilise their concentrations at current levels.” Furthermore, the executive summary said nothing regarding the crudity of the early mathematical models that were being used as the basis for the climate change predictions.
Without wishing to theorise as to why such downplaying of uncertainty should have happened, the fact that it was even possible speaks little of the confidence one can place in the consensus that the IPCC was seeking to portray. The IPCC has released a number of reports since, but the allegations of false certainty refuse to go away. For example, several passages within the IPCC’s Second Assessment Report, passages that had warned of uncertainty, were removed apparently after the peer review was supposed to have been concluded. Upon observing this, Professor Frederick Seitz stated, “But this report is not what it appears to be – it is not the version that was approved by the contributing scientists listed on the title page. In my more than 60 years as a member of the American scientific community, including service as president of both the National Academy of Sciences and the American Physical Society, I have never witnessed a more disturbing corruption of the peer-review process than the events that led up to this IPCC report.”5
For my second example of the promotion of false certainty, I take the notorious Hockey Stick graph produced by a team led by Dr Michael Mann, and used by the IPCC in its Third Assessment Report for the advancement of its cause. It famously showed a steady, uneventful decline in temperature over the last millennium, ending with a dramatic upturn coinciding with the onset of industrialisation (a shape of curve presupposed to resemble an ice hockey stick). As such, it was a hugely significant graph since it exorcised the pre-industrial warmings alluded to in the First Assessment Report; in particular, there was no sign of the so-called Medieval Warm Period. With the Hockey Stick to hand, the IPCC no longer needed to bury the uncertainty in the body of its reports, since the graph proved that the uncertainty simply didn’t exist. It showed that current temperatures are unprecedented and there is no evidence of any pre-industrial warming event in the Earth’s recent history, at least not on a global scale. The Hockey Stick was the smoking gun that proved the AGW thesis, and with it the science was settled. Or was it?
Somewhat surprisingly, no one within climatology saw fit to attempt a reconstruction of the analysis undertaken by Dr Mann and his associates; this does not appear to be the ‘done thing’. Instead, it fell to a semi-retired mining consultant, Steve McIntyre, to attempt such an audit – and the results were damning. I do not intend in this article to go into detail; see instead A. W. Montford’s book, ‘The Hockey Stick Illusion‘, or indeed Dr Mann’s ardent defence, presented in his book, ‘The Hockey Stick and the Climate Wars’. To justify my concerns, I do not actually have to take sides in the debate; it is sufficient to note that the debate exists. Nevertheless, when one becomes aware of the statistical shenanigans that lay behind the Hockey Stick’s shape, it is difficult to avoid the conclusion that the data had been mercilessly tortured for a confession. Certitude emerged from the analysis, but only at the expense of integrity. Given the huge endorsement and publicity that the IPCC had initially given the study, its subsequent unravelling and ultimate debunking had to be a matter of severe embarrassment. Or, so one might have thought.
Of course, here I am taking the sceptic’s position. There are those who point to more recent studies, such as the PAGES 2K consortium, that appear to provide independent corroboration of the Hockey Stick. So perhaps, in the end, Dr Mann’s statistical cunning wasn’t necessary in order to stumble upon the correct answer. The trouble I find with this argument is that PAGES 2K seems to be a variation on a theme: Take lots of proxy data, mangle it to death with manipulation and statistical analysis until you get the answer you want, and hope that no one notices the uncertainties and false assumptions that lurk within the detail. Already, the PAGES 2K findings have had to be updated in the light of criticisms6 and, as each amendment is made, the low frequency variabilities in the temperature reconstruction return ever more prominently. The more that paleo-climatologists try to get rid of the Medieval Warm Period, the more they seem to expose the hazards of trying to reconstruct a reliable, global temperature record from collections of localised proxy measurements.
The real irony of all of this is that the Hockey Stick’s political importance was always far greater than its scientific relevance. The AGW hypothesis is not falsified by the existence of a Medieval Warm Period. However, in the politicised arena in which climatology is conducted, the temptation to claim that the science was settled was too strong for the IPCC and, as a consequence, it backed a horse that ended up failing its drugs test. That would have been bad enough, but the lengths then taken to defend the Hockey Stick exposed an even more worrying trait. Even in areas where falsification is eminently possible, there appears to be a culture within climatology that seems to be designed to frustrate it. Anyone who doubts this needs to pay greater attention to what Climategate revealed: the illegal deletion of emails that were subject of a Freedom of Information Act request, conspiracy to intimidate the editorial boards of scientific publications, corruption of the IPCC peer review process, persistent efforts to obstruct the release of data for open scrutiny, censorship of data that did not support the central message, and blatant misrepresentation of data to achieve a desired political impact. Yet all of this was subsequently dismissed as nothing more than the ‘robust’ dialogue to be expected between honest and capable scientists going about their daily business!
However, one did not need Climategate to reveal that an ugly culture of intimidation, bias and censorship surrounds climatology, since there are many within the field who can testify to the following: IPCC membership withdrawn for those who are deemed to be contrarian; having publication of one’s articles unreasonably refused; withdrawal of research funding or threats thereof; accusations that a particular source of funding undermines one’s scientific and personal integrity; accusations of incompetence and stupidity; and vilification through the use of derogatory terms, such as ‘denier’ or ‘science-speaking mercenary’. Many observers may claim such tactics are necessary for the robust defence of a legitimate scientific consensus that has been placed under attack by an army of ill-motivated doubt-mongers, though I prefer instead to point out that such behaviour does little to allay the currently vaunted crisis in science.7
It is perhaps only fair to mention that some of the certainty-mongers, such as Dr Mann, are equally aggrieved at the tone of the debate and like to present themselves as the real heroes, guardians of the truth, surrounded by frenzied and rabid naysayers. I dare say that Dr Mann has suffered considerable abuse, but the idea that he is a beacon of light in a wilderness of aggressive ignorance is a bit difficult to swallow when he also claims that all right-minded scientists agree with him. All I can say is that there must be a lot of wrong-minded scientists out there.
But enough of mudslinging. When it comes down to it, the real problem with climatology may have nothing to do with would-be conspiracies. The problem may be the extent to which the consensus is due to a straightforward selection effect. It is an uncontested fact that, from the very early days, governments throughout the world have invested heavily into researching the AGW hypothesis, often to the exclusion of potential alternatives. I am not here to question their motives; they are not important. All that matters is that, if funding for research is restricted in such a manner that it favours an interest in a specific area of conjecture, then one cannot be surprised when the scientific community one has employed for advice starts to speak with unanimity.
There is no need to resort to conspiracy here. I am quite prepared to accept that scientists are being sincere and honest in raising concerns regarding the extreme possibilities of AGW, and there is every chance that they could be correct. However, it could also be that they predominate only because those climatologists who might have entertained alternative views were either coerced into compliance or left the game – the days of the self-supporting gentleman scientist beavering away in his study are long gone. It is conceivable, therefore, that the consensus, rather than being a result of minds being changed during debate and inquiry, instead emerges following a form of sociological natural selection.
One may be tempted to reject this proposal as being fanciful. Surely, it is most unlikely that a whole community of scientists could blithely ignore significant avenues of enquiry. Well, it might seem unlikely, if it were not for the fact that it happens all the time. All that is needed is the right mix of conviction, social pressure and the inability to falsify through experimentation. Take, for example, a scientific discipline that one might have thought was a paragon of scientific integrity – high energy particle physics.
String Wars
In the late 1970s and early 1980s theoretical physicists started to develop ideas that were not only unfalsifiable within the immediate future, but also within any conceivable future. These ideas went by the name of string theory; soon to be replaced by its supersymetric equivalent, superstring theory. The essence of the idea is that the elementary particles, including the force-carrying particles of quantum field theory, are not point-like objects but are instead manifestations of the vibration of open strings or loops of energy. Each vibrational mode is detectable as a particle of a given fundamental type. The problem with string theory, however, is that the strings are so small (in the order of the so-called Planck Length of 10-35 m) that the energies required to explore such fine detail in nature are literally astronomical – it would require a particle accelerator so powerful that it would have to be the size of a medium-sized galaxy. It has therefore been alleged that the theory has so far failed to make a single prediction that is testable in the laboratory.
This problem has led a number of observers to question whether string theory is actually a science at all. For example, Peter Woit, in his book and blog, both titled, ‘Not Even Wrong’, bemoans string theory’s lack of falsifiability, saying, “It is a striking fact that there is absolutely no evidence whatsoever for this complex and unattractive conjectural theory”. Nevertheless, this hasn’t stopped the theory from coming to dominate the halls of academe, to the chagrin of those who would propose alternative ideas for the fundamental structure of nature. The problem is that, once a school of thought has been embraced by those who are in a position to dictate the future direction of study, it is very difficult to dislodge it. In the early days, the enthusiasm for string theory was understandable because it appeared to carry considerable merit and promise.8 However, once a theory is ensconced, the enthusiasm for it can be just as well explained by a desire to protect the reputations and livelihoods of those individuals and organisations that have invested so much in its development. A less cynical though similar sentiment is expressed by the mathematical physicist, Roger Penrose, in his book, ‘The Road to Reality’, when he says, “The often frantic competitiveness… leads to ‘bandwagon’ effects, where researchers fear to be left behind if they do not join in.”
It is fashionable in this day and age to cry, ‘conspiracy theorist!’, whenever someone questions the legitimacy of the regnant creed. However, one does not have to appeal to subterfuge in order to explain string theory’s dominance. Such dominance is an emergent phenomenon that should be expected, given that faculty heads will naturally promote those around them who share their own theoretical inclinations. And, given the pressures involved, an aspiring scientist is unlikely to see any lack of integrity behind their choice of study. Nevertheless, the fact remains that in many universities there came a point when it was extremely difficult to get funding to study any unification theory other than string theory.9
As non-experts in a particular scientific discipline, the rest of us can only rely upon scientific consensus to determine where the truth probably lies. In this instance, if one were to take a poll of theoretical particle physicists and ask them which of the competing theories is the most promising candidate as a Theory of Everything, the poll would undoubtedly come down heavily in favour of string theory. However, this merely reflects the predominance of string theorists within the field.
Of course, we can also observe the practitioners in debate, and decide who appears to be winning the argument. Unfortunately, however, arguments can be won by means other than by reference to irrefutable facts. In the absence of compelling experimental evidence, such arguments often start with well-intended allusions to virtues such as mathematical elegance, before degenerating into the sort of slagging match that academics seem to excel in. And woe betide anyone who dares to question the integrity of the field of study concerned. In his book, ‘The Trouble With Physics: The Rise of String Theory, the Fall of Science and What Comes Next’, the theoretical physicist, Lee Smolin, provides a passionate plea for a fresh and honest appraisal of the state of affairs within theoretical particle physics. For his troubles, he received the following review from one Luboš Motl:10
“…the concentration of irrational statements and anti-scientific sentiments has exceeded my expectations…”
Anti-scientific? Where have I heard that one before? It seems to be a strange fact requiring explanation that anyone who calls for the re-affirmation of scientific principles ends up being the one accused of being anti-scientific. At least, as far as I am aware, no-one has been accused yet of being a ‘string denier’. Even so, the debate is heated enough for participants to have coined the term ‘String Wars’.
I should point out that even those who are most concerned regarding the state of theoretical physics are not proposing that string theory be abandoned. Rather, they simply draw attention to the problems of experimental verification and warn that this results in a freedom to speculate that is not entirely healthy. Regarding the survival of theories, when natural selection by experiment is removed, natural selection by sociological means will often take its place. The result is that consensus is no longer achieved through the reduction of epistemic uncertainty, but rather by means of a narrowing of focus.
Beware the Bias
So, climatologists who are in agreement over dire AGW predictions are no more (or less) misguided or disingenuous than the hordes of theoretical particle physicists studying and promoting string theory. Also, let us be honest here, when seen in isolation, there is strength in the AGW argument, despite the central role played by predictions that cannot be falsified within the timescales required. However, one would feel a lot more comfortable in accepting this strength at face value if one didn’t suspect that the opportunity for developing counterarguments had been thwarted by non-scientific means. It is very disconcerting to think that the AGW argument may have been artificially strengthened due to confirmation bias. In such circumstances, can we be sure that we are not ignorant of our ignorance?
Furthermore, there has been no other scientific controversy that has suffered so much political interference, or attracted nearly as much media coverage, as that existing within climatology. I say ‘suffered’ because, despite an abject lack of qualifications to do so, there appears to be no one within journalism or politics who cannot tell you, with certainty, which side of the argument is correct: All the dire warnings arising from AGW are true and anyone who doubts it is a loon. As responsible citizens we are all required to back the scientists (the real ones), so how fortunate it is, I say, that we have the media and politicians to guide us through the intricacies of the debate. Surely, we couldn’t hope to have two more reliable and trustworthy sectors of society to let us know the strength of the consensus and to point out who the cranks, amateurs and phoney sceptics are.
And, of course, thank God for the internet. What little part of it that isn’t dedicated to pornography seems to be dominated by supposedly irrefutable arguments for one side or the other of the ‘CAGW will kill us all’ debate. For anyone with an open mind trying to determine the truth, it is just too easy to give up in despair. As for the rest of us, we form our opinions based upon an emotional impulse and spend the rest of our lives engaging in post hoc rationalisation, seeking out the information that confirms our chosen bias. It is unfortunately the case with global warming that individuals of all persuasions will always be able to find the encouragement they are looking for.
When the Stakes Get Higher
I have argued here that there is little to be found in the CAGW controversy that cannot be found elsewhere within science once the tight grip of experimental corroboration is loosened. Whenever this happens, bitter dispute breaks out between sides holding entrenched positions. Despite this, a consensus emerges, but the consensus is of questionable value since there is good reason to suspect that selection effects are significant. Individuals who question the consensus are labelled as anti-scientific, even when their scientific credentials are beyond dispute. Scientists feel under attack, though they would help themselves considerably by taking greater care to avoid unfalsifiable speculation.
There are differences between physics and climatology, of course. For instance, in climatology the situation is worsened by a politically motivated denial of uncertainties and a lack of commitment towards openness and the reproducibility of results. Also, the stakes are much higher, so it is in the nature of things that we are all invited to join in the bun fight. However, as with supersymetric string theory, I do not feel qualified to arbitrate. All I can say is that it seems to be the case that the arguments persist because all sides are so highly motivated and there are no experiments available to settle the issue. Under such circumstances, I would have thought having an open mind was a reasonable position for the layperson to take. Unfortunately, tolerance of scepticism is no longer the order of the day and anybody who questions scientific authority is assumed to be irrational. However, the real violation of scientific principles lies in the notion that science can be settled without having to wait for predictions to transpire.
Initially, politicians evaded the falsifiability problem by invoking the precautionary principle, in which the plausibility of an idea is sufficient for it to be treated as if it were confirmed. So, in climatology not only was the falsification of ideas technically difficult, it wasn’t even deemed necessary. But this was an overtly political position to take: If deferring a decision until all uncertainty is removed carries an existential risk, then there is political wisdom in not doing so. Nonetheless, the precautionary principle is notorious for being a self-defeating logic. When the price for taking action is potentially catastrophic (as it may be when one considers the drastic actions proposed by the CAGW protagonists), then the precautionary principle can also be invoked to argue against taking such action. Uncertainty cuts both ways, and perhaps this is why the denial of uncertainty seems to have usurped the precautionary principle as the favoured policy in many people’s minds.
Down on the Mississippi, Mark Twain understood better than most just how easy it is to get carried away with conjecture when the evidence is sparse, but I doubt that even he fully appreciated just how easily such conjecture can mysteriously transform into fact as the stakes get higher. It doesn’t require artifice to achieve this, though it is surprising what some advocates will resort to in order to ensure that the ‘righteous’ side of the argument wins. If it were up to me, however, I’d accept the uncertainties and invoke the dreaded precautionary principle. Although far from ideal, this is a better option than downplaying uncertainties to the extent that having an open mind in the face of a questionable consensus is taken as a sign of criminal stupidity. After all, scepticism used to be the compass for the scientific mind. So I ask once again: What happened to the science?
John Ridgway is a physics graduate who, until recently, worked in the UK as a software quality assurance manager and transport systems analyst. He is not a climate scientist or a member of the IPCC but feels he represents the many educated and rational onlookers who believe that the hysterical denouncement of lay scepticism is both unwarranted and counter-productive
Notes:
1 See http://www.spiked-online.com/newsite/article/6320#.WWu5mGeWzZ4
2 For further information, see Storch, Hans von; Zwiers, Francis (1999). Statistical Analysis in Climate Research. Cambridge University Press. ISBN 0 521 45071 3.
3 I should explain that aleatoric uncertainty results from random fluctuation. It reflects the variability of the real world and, as such, it contrasts with epistemic uncertainty (gaps in our knowledge) and ontological uncertainty (gaps in our knowledge of the gaps in our knowledge).
4 Those who dispute that the post hoc tuning of climate models is common practice should refer to Hourdin Frédéric, et al (2017). The Art and Science of Climate Model Tuning. American Meteorological Society, Journals Online.
5 Much has been made of the fact that the IPCC successfully appealed against the Washington Post article in which this quote appeared. The basis of the appeal was that the article failed to mention that Seitz was not a climatologist, not a member of the IPCC, was in pay of an oil company and that someone had previously accused him of being senile. Apparently, by not launching an ad hominem attack on its interviewee, the Washington Post failed to strike the journalistic balance demanded by the IPCC! More relevantly, the IPCC also successfully argued that the changes did not represent a corruption of their peer review process, which, of course, was a tacit acceptance that the changes had been made.
6 Steve McIntyre again. See the Climate Audit website.
7 Other than falsifiability, reproducibility of results stands as a principal requirement of the scientific method. It is on this score that many scientific disciplines are currently failing badly; hence the crisis. Most, if not all, of the explanations offered for this failure apply to climatology. On this occasion, I chose falsifiability rather than reproducibility of results as my main theme. Maybe next time…
8 In particular, superstring theory equations predict the existence of a spin-2 boson that has all of the properties expected of the force-carrying particle posited by a quantum field theoretic description of gravity (namely, the graviton). This prediction paves the way for an understanding of gravity that is unified with the so called Standard Model of particle physics.
9 For further information, see Woit, Peter (2006). Not Even Wrong: The Failure of String Theory and the Search for Unity in Physical Law. Basic Books. Ch. 16. ISBN 0-465-09275-6.
10 Motl, Luboš (2004). “Lee Smolin: The Trouble with Physics”: a review on The Reference Frame.
Nice article.
But string theory…
Look, there is a better way of looking at all this: All knowledge – human knowledge – is suspect. No statement about the objective world can be held to be intrinsically true.
What human beings seem to do is take a huge data flow of experience and model it into a model that they then call ‘real stuff wot’s happening’
And out of that model, come ‘raw facts’. The cat is sitting on the mat, we say. It’s irrefutable.
Until we start to probe the exact boundaries of what constitutes ‘the cat’, and ‘the mat’, and define what exactly we mean by ‘sitting’..
There is a huge disjunct between what is happening at the quantum level, and what is happening in the classical world. Take one simple example. If the world is made up of quantum level entities constantly transforming and reforming one into another, how can we give the stable existence of ‘objects’ any credence at all in the classical view? Is my existence as matter of mere probability?
String theory isn’t right and it isn’t wrong just like every other model used in quantum physics. What theoretical physicists are doing is a very sophisticated form of curve fitting. They are looking for mathematical equations that seem to fit the broad shape of what their data seems to say is ‘happening’. Though even the concept of ‘happening’, itself, is open to question.
String theory is just one multi-dimensional model that seems to fit the facts. There are others, less fashionable.
Mutatis mutandis it doesn’t really matter which theory you use so long as it accurately predicts. The real philosophical problem today is that science is full of scientists who don’t actually understand the science they are doing.
They are still at the level of a three year old where they think that because a model predicts the right answer, the model is actually a picture of something that really exists. A lightning storm is the Gods fighting and every time the fire god, strikes the sound god, he throws back a sound bomb. This accurately predicts that thunder always follows lightning…It is a falsifiable model that has never been falsified. So it must be true…
All this confusion arises because of the human propensity for sloppy shorthand thinking, to regard the models they hold in their heads as reality itself, instead of just a picture of it.
Quantum physics reminds us of the same mantra a Buddhist might utter: No, a tree falling in a wood with no one to see it makes no sound, worse, unless someone sees it fall, it hasn’t actually fallen. Even worse, there is no tree and no wood. Nothing exists until we think it does, not even ourselves…except…the fact that magic thinking does not work, suggests that what we dream into being in the way of an external reality, is constrained by some force beyond us.
If we examine the whole panoply of sloppy Left style thinking in the political and popular arenas, this is really the underlying problem that pervades it.
The philosophy of the New Left, declares correctly that truth and reality itself is a social construct, which is true. Today we talk of electricity and sound waves, not the gods of light and sound, and live in a different reality than the one our forbears inhabited, BUT we have forgotten what the great thinkers of the past understood…Reality as we understand it may well be a social construct, relative to culture, but that does not mean we have the ultimate freedom to choose what it may be. Science was the great antidote to superstition, in the science sought to elucidate models that worked whether or not we believed in them. And it did work rather well, to the point where the hoi polloi started to believe that the models of science actually represented something real. Swiftly followed by the sort of third rate scientist who does thing like climate science.
Completely ignoring William of Occam, who is so well known and so completely misunderstood. He didn’t say ‘the simplest explanation is the correct one’ but ‘when acknowledging the impossibility of arriving at the Truth, it makes sense to pick the simplest approximation that works’.
And there in a nutshell is the problem. The public at at large and indeed considerable numbers, if not the majority of the so called scientific community itself, do not realise that science itself has no truth content. It is simply a collection of models that WORK.
Whilst the Left is equally benighted in that having realised that all knowledge is relative to culture as is any given truth, they think that by remodelling the way people think and what they believe, they can in the end change reality itself….And in the process has generated huge sewage works full of models that DON’T work.
But are comfortable to believe in …
The fact that some models work, in terms of predictive ability, doesn’t show us that they are true, but it does show us that there is a world beyond our conceptions, that will have its own way.
Both Science and the Left are too sure of their conceptions.
And when you mix them up and get Climate Science…and a host of other politically correct garbage…the result is – as my dear old granny used to say – a right buggers’ muddle.
Take e.g. racism. Science – real science – would have no problem in the proposition that human beings who have identifiably adapted to niche conditions around the world in unique ecological niches, would in fact be identifiably different. Different strengths, different weaknesses. But politically correct thinking goes beyond the avoidance of bigotry – that is making false and oversimplified assumption about race – to almost denying the existence of it, whilst reaffirming our need to do something about racism!
That is, in modern PC terminology, ‘race’ is deemed not to exist, only ‘racism’..how weird is that?
I have made this argument many times, that we need desperately to understand philosophy today, especially the philosophy of science, much better than we do.Because it sweeps away most of these arguments.
Science has to abandon its claim to truth content, and reaffirm its real strength, which is its utility, and thereby disarm the Left who are using its framework to justify what are in the end lies. Models that do not work.
Gravity does not ‘exist’, but the idea of it allows very useful predictions to be made. AGW does not ‘exist’ either, but the idea of it has allowed no useful predictions whatsoever to be made. And that is the ONLY difference, between science and non-science.
Our world is all made of models. From the moment we open our eyes and scream at the appalling cacophony of experience the world presents us with, we start making models of it. Some work, some don’t. Some are beautiful, but useless, others are ugly, but effective. Provided we dont believe in any that get us killed, and make it to procreation, the whole game staggers on. Science is a collection of mutually consistent models that we have found to be more or less reliable. Not true, just reliable.
String theory seems to give some right answers. But it has no more truth content than relativity does. It’s just a model. And others also give right answers.
AGW has failed to give any right answers yet. It is ‘metaphysical’.
Plenty of other excellent models are metaphysical. The idea that physical reality actually exists for example, and we are not simply plugged into the Matrix…
But AGW is not even useful in the prediction of climate.
So what is the point of it?
I maintain it is simply the product of Left thinking. Its purpose is commercial, political and sociological. It is a play on Guilt. Mankind, bereft of any God who might forgive him, now seeks to salve his conscience in Gaia, rejecting the Nietzschian alternative, and grovelling in terror at the power he actually thinks he has, and his sheer incompetence at deploying it.
Modern man thinks he has the power of a God, but the morals of an alley cat. Or rather the Left is comprised of people who think that Science has the power of a god, and that scientists are like them, with the morals of an alley cat.
Ergo science must be brought under moral and political control. Or better still, because they think everyone has the morals of an alley cat and can’t be trusted, destroyed altogether.
In a feudal or theocratic society, at least everyone knows where they stand…so let’s take the world there..
Most published research papers are wrong.
Drug companies mine the literature looking for interesting results that they can develop into profitable drugs. When they find something the first thing they do is attempt to reproduce and replicate the results. In the overwhelming majority of cases, they can not do so. In most cases, the original researchers can’t even reproduce their own results. link
Medical science is unique because there is a concerted effort to reproduce and replicate results. That means that the other sciences are not being properly scrutinized. Even so, the replication crisis has been noted in all disciplines, even engineering.
Scientists are not inherently trustworthy, their inability to replicate even their own research is damning. Folks like Dr. Michael Mann are an example of expert overclaiming on steroids.
Key word “profitable”. Easily corrupts science.
This cuts both ways. Big Pharma has corrupted science, in particular by burying papers that show no or negative effects from their drugs. link
On the other hand, the profit driven pharma companies have exposed the miserable quality of academic research.
In my view the climate models have been falsified completely, because none of them has predicted the current pause of warming. Falsification of a model is simply data which shows that the model is not substantially accurate, and that is certainly the case. Running the models for millions of years ago also does not predict the climate correctly, so they are again falsified. This is all covered up by nonsense statements of accuracy and exactness of results which are simply false.
Don’t get me into string theory, one dimensional strings with extension into 3 dimensions? Etymological nonsense.
When you’re dealing with statistical models, the proof is a bit less black and white than with “hard” science. How close is the model to what is happening? What’s the acceptable margin of error? I think in the case of models not predicting the pause, that’s a serious problem. However, one can gain wiggle room by changing statistical methods and changing the “pause”. It’s all very fuzzy. Not something to base a worldwide change on.
String theory and quantum mechanics are models that work. They make sense and aid in the study of the universe. No the macro one, of course. Since 99.99999% of the population will never understand the theory and it has very little effect on their lives, a model that works is all that is needed. It may turn out to be etymological nonsense. Right now, it really doesn’t matter except to those working in the field. To everyone else, it’s just a fascinating theory that they cannot really relate to.
back to basics: http://www.davdata.nl/math/mentalclimate.html
about science and religion
Indeed : let’s grow up and realize that we don’t know the future and our ancestors never did.
Progress is improving the presence.
The banality of climate “science” on display.
The author is so confident in the specialness of climate “science” that the lack of science is simply rationalized away.
For years I and others have pointed out that climate “science” is a pernicious social movement that uses science words and claims to justify the prejudice and politics of its true believers.
The author of the essay this blog post stems from is so confident in those beliefs that he is comfortable with setting aside the pretense of science altogether.
@Hunter;
Are you replying to the correct post?!? He is saying pretty much the opposite.
Bravo, Mr. Ridgway, bravo !!
This is an extremely well-written synopsis of the matter at hand.
“No one would question that climatology is a science.”
I would. Until someone can provide a scientific definition of climate, I wouldn’t call it science. It hasn’t been demonstrated that climate science is anything more than squiggology.
Andrew
Science is what may be quantitized, measured and researched. See the dietary recommendations which vary over time. Honest scientists always mention uncertenties ans assumptions. Some things we know pretty well (mechanics, energy..) other things are subject to ongoing investigation like the climate and many biochemical processes.
But climate is a conceptual derivation of weather. It’s doesn’t exist until someone decides to call a section of weather history ‘climate’. It’s the weather system that is being studied. Climate is imaginary.
Andrew
climate is just weather averaged over a period of time. I think the climate can be researched on itself being an average as well as the “amplitudes” (deviatations from average) of weather conditions.
Maybe it’s wise to investigate regional climates. No one cares about global temperatures. (put your head in the oven and your buttocks in the fridge and you’re fine?)
“climate is just weather”
Exactly.
Andrew
So David, what you can take from this brief discussion with me is that when you see someone using ‘climate’, you know they are already projecting their beliefs on the weather.
Andrew
In other words, ‘climate’ is a way to smuggle your beliefs about the weather into the conversation.
Andrew
“No one would question that
climatologyclimate science is a science.”As somebody once said, if a subject has the word “science” in its title, that is a pretty good indication that it is probably not science.
If science were to be honest with itself, it would admit that its very foundations are built upon conjecture.
And yet, our cell phones still work…
But will they always work?
I came: Maybe. Maybe physics will break down. It’s all conjecture.
” To this extent, the proposal that mankind is currently contributing to climate change is easy to accept.”
Not via CO2 as you suggest.
When I first heard of climate modellers “tuning” their models by using historical data, I was aghast
at the thought, based on my studies of statistics in grad school. I remember an experiment designed to illimunate the problem : create a written psychological test by administering a large number of questions to a group and then , using correlations between each item and the criterion scores, select the best 20 items for a new test. You can show that this test correlates very highly with the criterion, much higher than a test composed of all of the items tested. Now bring in a new group of test subjects and administer this test composed of the “best ” items. And when you do, you will ALWAYS find that in this group, this wonderful test is not very good. The problem is that when you selected those “best” test items you capitalized on random statistical fluctuations
(the error component of the items). Retesting test items in another group is called cross validation.
Modifying climate model parameters because you have independent reason to choose a “better’ value is OK, but altering them simply to make them agree with criteria data is a BIG no-no – there is no theoretical rationale for the alteration, so you will be simply tuning to agree with (mostly) static in the data. Even worse would be to claim that the “tuned” model is excellent because “Look how closely it models this past data.” This is not rocket science – everyone who has ever studied statistics and experimental method knows not to capitalize on random flutuations of data. If one DID tune the model, then they should at least cross validate that tuned model by having it postdict several other past periods of time and vary the length, etc. Of course, this still doesn’t really validate the model unless you can show that the Earth’s climate is consistent over time in the way it behaves. Let’s see anyone prove THAT.
” I was aghast at the thought, based on my studies of statistics in grad school”
But it isn’t statistics. It’s solving physics-based equations, conservation of mass, momentum, energy etc. Discretised, there are a huge number of them, and a huge number of unknowns. There is the usual requirement that the number of equations has to match the number of unknowns. But with a big ill-conditioned system, that gets fuzzy.
There are things that weren’t explicit equations, like TOA balance, that we still have reasons for believing should be true. Solving the initial equations doesn’t pin that down tightly enough. So you could add it as an extra equation. But that means you need an extra degree of freedom (unknown). That can be a parameter which was an input, but which you didn’t know very well. It should be related to the extra constraint – maybe cloud reflectivity or some such. The extra constraint (TOA) lets you pin that down. For the original equations, we had a direct calculation (time stepping) that led to values for the many discretised unknowns. Here we probably don’t have that, and have to work by trial. In de’s you might call that a shooting method. Here it is called tuning. But it is not different in principle.
Climate models are all statistical models. There may be direct measurements, but without statistics, it’s just the raw data. The minute you draw a trend line you’ve crossed into models and statistics.
“The minute you draw a trend line you’ve crossed into models and statistics.”
They don’t draw trend lines in GCMs.
You are making it way more complicated than it is. Observations -the pause, paleoclimate record, the available untampered instrumental temperature records, the satellite observations…ect…- disproved the theory that global warming was a function of CO2 concentration. There were three basic alternative courses of action to take from there:
A) Form theories with CO2 concentration not related to temperature and test.
B) Form theories that CO2 concentration, among many factors, is only a relation to temperature, but its significance is overridden by natural variation, and the ESC to CO2 concentration is of limited overall significance, and retest.
C) Fiddle with the data to make them fit the theory needed by the political driven agendas.
They chose option C to their everlasting shame. It is that simple.
“But even in the purest of sciences one can occasionally find oneself in areas of speculation that are not obviously susceptible to the scrutiny of experimentation.” then you are no longer practicing science but a religion …
How much ale must one drink to eliminate (or create) aleatoric uncertainty?
Admittedly?! … NOT by moi.
Experiments – I love ’em ! I started messing about with things when I was about 8. By the age of 12 I had taught myself enough about electronics to be able to repair radios, tv, and similar (mostly valve stuff then). At school we did lots of experiments and even had to devise our own investigation of something and write it up for “A” level physics. Engineering at University was practical too, lots of lab work. I may have gone a “bit too far” when I got a rather heavy motor generator set jump off the floor!
I’m retired now but I am working on an experiment to clear something up about climate science. I am testing the theory of the gravito-thermal effect. This may well turn out to be complete and utter BS. We know Anthony thinks so as do other moderators and contributors here. Those who wrote the “slaying book” think otherwise and I do believe Richard Feynman thought is was correct.
So we observe 2 groups of people with opposite views, lots of talk and lots of theoretical discussion. Not getting it sorted though is it? Presumably (??) no one can object to an experiment to decide the issue.
I am lucky enough to have secured a suitable building which will allow me to erect a vertical 10m column for the experiments. This is to be a scale up of the work of R.Graeff. I plan to use pure argon as one of the gaseous mediums to be tested as we think this would result in a 3K temperature difference between top and bottom. I expect either to find a result near 3K or zero which would decide the issue one way or the other provided we do everything properly.
What happened to science? Is not the answer that the definition of science has been changed? Yet even the proponents of that change don’t want to admit that the definition has been changed?
When I studied science in secular state universities, I was taught that science could study only repeatably observable phenomena. Yet there are theories that dealt with unobservable phenomena that are declared as “scientific facts”. That’s a self-contradiction, the most basic of all logical fallacies. Basically, the scientists gave lip service to the definition of science that they had inherited, but act according to a different definition.
It appears that the definition of science follows the ideas found among the intelligentsia of a society. What we call “modern science” had its roots in the Protestant Reformation. Ideas such as that observations could override the theories of “experts” was already percolating in theology before the Reformation broke out, in fact helped guide Luther and the other Reformers into starting the Reformation. Science followed theology In making repeatable observation the basis of science, from which observations hypotheses and theories are derived, and against which hypotheses and theories are tested and sometimes falsified. In this use of science, it was recognized that certain beliefs are forever outside the realm of science. Science was intended to study only a limited part of total reality, namely that which deals with observable, physical phenomena.
About two centuries ago, a different spirit infected Europe’s intelligentsia. They wanted a different narrative to be told, and they wanted it to be “science”. So scientists who agreed with the beliefs common among the intelligentsia complied, even though those beliefs deal with phenomena that cannot be observed. Yet they couldn’t admit that they were changing the definition of science, because then the game would be up. This practice among scientists started over a century ago.
The example of high energy particle physics is a good example. It’s based on a mathematical derivation ultimately of Einstein’s relativity theory. In this upcoming total solar eclipse, will anyone recreate the 1919 experiment that “proved” Einstein’s theory using the improved lenses and cameras that we have today, as well as updated information concerning the variables used in the formulae? I doubt it. What if they show that the 1919 experiment went awry? What if a recreation shows that Einstein’s theory didn’t accurately predict the bending of starlight, but Newtonian physics does? What then? What will happen to a century of physics development? Would such a result lead to a crisis in physics? What would happen to thousands of careers and reputations based on the reported results of the 1919 experiment? It is precisely for those reasons that I predict that a recreation of the 1919 experiment won’t happen. It is precisely for those reasons that it should be done.
We already have a clue what might happen. Tesla was an experimental physicist. He took the results of predictions based on Einstein’s theory and quantum mechanics, took them to the lab and tested those predictions against experiments, and found that they differed. In modern science, that would count as evidence against those theories. In practice, Tesla was denounced as old fashioned and out of touch.
In closing, there’s a crisis in science, and it’s all of science.
OK, the 1919 experiment looks like being the next one I will have a go at. Maybe I will have better luck funding it than the gravito-thermal one!