Guest essay by Eric Worrall
How do you falsify a climate model? Australian National University Climate scientist Sophie Lewis acknowledges that climate models are not falsifiable – yet claims we should trust them anyway.
Climate change has changed the way I think about science. Here’s why
Research fellow, Australian National University
August 10, 2017 3.30pm AEST
I’ve wanted to be a scientist since I was five years old.
My idea of a scientist was someone in a lab, making hypotheses and testing theories. We often think of science only as a linear, objective process. This is also the way that science is presented in peer reviewed journal articles – a study begins with a research question or hypothesis, followed by methods, results and conclusions.
It turns out that my work now as a climate scientist doesn’t quite gel with the way we typically talk about science and how science works.
…
1. Methods aren’t always necessarily falsifiable
Falsifiability is the idea that an assertion can be shown to be false by an experiment or an observation, and is critical to distinctions between “true science” and “pseudoscience”.
Climate models are important and complex tools for understanding the climate system. Are climate models falsifiable? Are they science? A test of falsifiability requires a model test or climate observation that shows global warming caused by increased human-produced greenhouse gases is untrue. It is difficult to propose a test of climate models in advance that is falsifiable.
Science is complicated – and doesn’t always fit the simplified version we learn as children.
This difficulty doesn’t mean that climate models or climate science are invalid or untrustworthy. Climate models are carefully developed and evaluated based on their ability to accurately reproduce observed climate trends and processes. This is why climatologists have confidence in them as scientific tools, not because of ideas around falsifiability.
…
Read more: http://theconversation.com/climate-change-has-changed-the-way-i-think-about-science-heres-why-82314
The problem with Sophie’s position is that fitting a model to past observations is not a test of whether the model is right; all fitting the model tells you is that you have found a way to fit the model. What counts is the ability of the model to predict the future – to accommodate observations which were unknown at the time the model was created.
“Careful development” just means current prejudices are carefully applied. But there are many more ways to be wrong than right – especially about something as complex as the global climate.
Climate scientists are desperate for their educated guesses to be accepted as science; so desperate that at least some climate scientists openly challenge the very keystone of science, the requirement that scientific theories must provide a means by which they can be falsified.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Come on every one knows that it is the cost of postage stamps that causes global average temperatures up. Jo Nova proved this years ago. In a daring departure from proven science she proved her theory with at graph.
http://joannenova.com.au/2009/05/shock-global-temperatures-driven-by-us-postal-charges/
An questions 🙂
A ‘Climate Scientist’ has stated the obvious. She admits that climate models do not work but we have to believe that they do. Just confirms in my mindset that ‘Clmate Science’ is a new religion rather than science.
“I’ve wanted to be a scientist since I was five years old.”
Well I admire her persistence, but perhaps it is time to choose a new goal, one which is actually reachable.
Blast, beat me to it!
It’s a worthy goal to be sure. One day, perhaps?
It’s always been a matter of two questions: First, does anthropogenic CO2 cause warming, and second, will the warming result in climate catastrophe? Even if they prove the first is occurring, then you have to prove the second…
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.
I think they’ve reached the point where they don’t even bother rationalizing anymore. .
We’ve had 30 years to observe the accuracy of the predictions of the “CO2 controls the climate global circulation models” … and the evidence strongly supports my belief that CO2 does not control the climate, so the fixation on CO2, and demonization of CO2, is junk science.
The grossly inaccurate GCM predictions / projections / simulations can be described by a scientific term I learned during my second and last year of engineering school: A pile of steaming farm animal digestive waste products with a cherry on top!
Of course a 100 year climate prediction can’t be falsified in a single lifetime.
But then a wild guess of the climate in 100 years was never real science to begin with,
especially when the causes of climate change are not known, and are still being debated.
Also, there are no real climate models — just prototype models that have failed miserably to predict even ten years into the future — we do have failed models — we don’t have real models of the climate change process that work.
And there can be no real models until the causes of climate change are well known — and even then a model may only be able to predict the future climate if the causes of climate change are cyclical, rather than random.
“Climate scientists are desperate for their educated guesses to be accepted as science …”
Wild guesses, not educated guesses!
An educated guess of the climate in 100 years would require, at the least, a thorough understanding of exactly what variables cause climate change, and those variables would have to be cyclical, in a regular way that was predictable. Even then, common sense tells us, or should, that predictions of the future are likely to be wrong, as they have been throughout recorded history
Climate change blog for non-scientists
http://www.elOnionBloggle.Blogspot.com
‘Bottom up’ GCMs = epic fail; ‘top down’ (emergent structures analysis) works. http://globalclimatedrivers2.blogspot.com
Average global water vapor is increasing at 1.5% per decade which is more than twice as fast as it would if dependent only on feedback (increase in water temperature).
You can’t make up stuff like this – which is why the average person on the street has a hard time swallowing that, yes, it is really going on.
“Science needs to be falsifiable?” Sorry, but that train left the station over a century ago.
There were many who pretended that the train was still in the station, and acted accordingly, But already certain theories that are not falsifiable were declared to be “science”, therefore “science” no longer needs to be falsifiable.
The traditional definition of science, and the scientific method, that I was taught in secular state universities has means built right into the method that can falsify theories. All textbooks and professors unanimously averred that what I was taught was the correct scientific method, but then they didn’t follow it. By their practice, they showed that they no longer agreed that falsifiability is necessary for science.
For example, later this month there will be a total eclipse of the sun over the U.S. Will anyone recreate the 1919 experiment that declared Einstein’s relativity theory to be accurate, using the better, sharper lenses we have today, along with updated information concerning all the variables to be used in the formulae? Don’t count on it. That’s “settled science”. It doesn’t need to be falsified. And that’s just one theory.
That woman is correct, theories don’t need to be falsifiable in order to be scientific.
It would have been more convincing if she had mentioned other areas of science where you don’t need falsifiable hypotheses.
Here is a testable prediction method and forecasts
Climate is controlled by natural cycles. Earth is just past the 2003+/- peak of a millennial cycle and the current cooling trend will likely continue until the next Little Ice Age minimum at about 2650.See the Energy and Environment paper at http://journals.sagepub.com/doi/full/10.1177/0958305X16686488
and an earlier accessible blog version at http://climatesense-norpag.blogspot.com/2017/02/the-coming-cooling-usefully-accurate_17.html
Here is the abstract for convenience :
“ABSTRACT
This paper argues that the methods used by the establishment climate science community are not fit for purpose and that a new forecasting paradigm should be adopted. Earth’s climate is the result of resonances and beats between various quasi-cyclic processes of varying wavelengths. It is not possible to forecast the future unless we have a good understanding of where the earth is in time in relation to the current phases of those different interacting natural quasi periodicities. Evidence is presented specifying the timing and amplitude of the natural 60+/- year and, more importantly, 1,000 year periodicities (observed emergent behaviors) that are so obvious in the temperature record. Data related to the solar climate driver is discussed and the solar cycle 22 low in the neutron count (high solar activity) in 1991 is identified as a solar activity millennial peak and correlated with the millennial peak -inversion point – in the UAH6 temperature trend in about 2003. The cyclic trends are projected forward and predict a probable general temperature decline in the coming decades and centuries. Estimates of the timing and amplitude of the coming cooling are made. If the real climate outcomes follow a trend which approaches the near term forecasts of this working hypothesis, the divergence between the IPCC forecasts and those projected by this paper will be so large by 2021 as to make the current, supposedly actionable, level of confidence in the IPCC forecasts untenable.””
Here is a testable prediction method and forecasts
Climate is controlled by natural cyclesHe. Earth is just past the 2003+/- peak of a millennial cycle and the current cooling trend will likely continue until the next Little Ice Age minimum at about 2650.See the Energy and Environment paper at http://journals.sagepub.com/doi/full/10.1177/0958305X16686488
and an earlier accessible blog version at http://climatesense-norpag.blogspot.com/2017/02/the-coming-cooling-usefully-accurate_17.html
Here is the abstract for convenience :
“ABSTRACT
This paper argues that the methods used by the establishment climate science community are not fit for purpose and that a new forecasting paradigm should be adopted. Earth’s climate is the result of resonances and beats between various quasi-cyclic processes of varying wavelengths. It is not possible to forecast the future unless we have a good understanding of where the earth is in time in relation to the current phases of those different interacting natural quasi periodicities. Evidence is presented specifying the timing and amplitude of the natural 60+/- year and, more importantly, 1,000 year periodicities (observed emergent behaviors) that are so obvious in the temperature record. Data related to the solar climate driver is discussed and the solar cycle 22 low in the neutron count (high solar activity) in 1991 is identified as a solar activity millennial peak and correlated with the millennial peak -inversion point – in the UAH6 temperature trend in about 2003. The cyclic trends are projected forward and predict a probable general temperature decline in the coming decades and centuries. Estimates of the timing and amplitude of the coming cooling are made. If the real climate outcomes follow a trend which approaches the near term forecasts of this working hypothesis, the divergence between the IPCC forecasts and those projected by this paper will be so large by 2021 as to make the current, supposedly actionable, level of confidence in the IPCC forecasts untenable.””
Bullspit. Climate change models are one of the few things about climate change alarmism that ARE falsifiable. All you have to do is take the predictions of the models and compare them to observed data i nthe prediction period (which does require you to wait a number of years, but certainly is not impossible). And now that we those years have elapsed, and we are well within the prediction period of most of those models, the models most certainly falsifiable. And not only that, they have been falsiFIED.
Now, there’s a lot about climate change that is not falsifiable. Because alarmists simultaneously claim that global warming causes opposite outcomes. If global warming, in theory, can cause both more arctic ice and less arctic ice, more extreme weather and less extreme weather, widespread crop failure and crop abundance, more rainfall and less rainfall, more snow and less snow (and the list of contradictions goes on and on), then whatever happens, it can be blamed on global warming. That’s the problem with global warming theory. You can’t pin the sons of bitches down on what effects are. They are covering all possible outcomes with respect to every factor, and literally nothing can occur that will falsify the theory.
This is a good example of how people write when they know that they are wrong. She knows that she is not using the scientific method and she knows that what she does really qualifies as pseudoscience. She just can’t bring herself to admit it, so she hides behind the word complicated. She should admit to herself that she is being controlled by her emotions not her brain.
The Popperian scientific method is a *heuristic* and a tool. It is a good heuristic, and a good tool, but it not definitional as “science.” Philosophers of science have acknowledged for a long time that the strict application of the scientific method has a lot of problems, most specfiicially being paradigm-bound. Failing a prediction does not necessarily mean that a theory is “wrong;” it may just need tweaking or evaluation of how measurements were made. Passing a predictive test does not mean a theory is “right,” it just means that it has passed a test that may not be “risky” enough.
Most real scientific inference uses dialectic rather than the “scientific method” because measurements have error and complex theories have lots of things in them that can be adjusted.. Most interestingly, the “scientific method” has never been applied to the “scientific method.”
People who claim that the Popperian scientific method is the only way to find truth have turned a heuristic tool into a religious dogme
I agree, billo. Before there was science, there was the philosophy of science. Beginning with Aristotle and accelerating with Galileo, Francis Bacon, and the Age of Reason, by the time of Newton the modern view of the scientific method was already well established. It was never ‘Popperian’ in the sense that scientists ever thought one could only demonstrate a hypothesis to be wrong, but never right, and it still isn’t or we wouldn’t place our lives in the hands of those who practice applied science, which we do daily. Aristotle recognized that if an observation contradicted a proposition that there had to be something wrong with either one or both of them, and the falsification principle really says little more, except it insists that to be scientific in the modern sense, a hypothesis (theory, “projection”) must be able specify (in advance) what kind of observation would show it to be false. That falsifiability is simple doesn’t stop it from being the sine qua non of good science, but it is hardly the sum of good science, which, for example, calls for unsimple qualities in a theory, such as being able to account for a large number of output events with only a few causal principles — but that only scratches the surface. That’s why not every idea that can be falsified counts a a scientific idea!
Nevertheless, if it can’t be falsified, while it may conceivably be true, it can’t be a scientific truth.
(I’m afraid I’m going to have to disagree with you on a number of points. First, the claim that the “scientific method” as used today was a dominant heuristic before Popper is simply wrong. Historically, “science” did not rest on falsifiability until the time of Popper. It rested on “verification,” or essentially the equivalent of parisomious cover theory in AI terms — how well and how simply it explained the existing data. While the idea of falsification is old, it’s incorrect to try to say that the “scientific method” as it is used by scientific method religtionists was formalized in a meaningful way before Popper. Certainly incongruities counted against the strength of verification, but the dominant heuristic was verification.
Second, you are wrong in implying that somehow the philosophy of science ended (or maybe just became obsolete) at the time of Newton. One of the thrusts of modern philosophy of science has been to show the problems with Popperian scientific method. I’ll point to Lakatos, Kuhn, and Feyerabend as people who have poked substantial holes in the conceit that the “scientific method” is the only path to scientific truth. And, again, it’s not that the scientific method is bad, but as a heuristic, it simply does not cover all paths to scientific truth, or as Feyerabend noted:
“My intention is not to replace one set of general rules by another such set: my intention is, rather, to convince the reader that all methodologies, even the most obvious ones, have their limits. The best way to show this is to demonstrate the limits and even the irrationality of some rules which she, or he, is likely to regard as basic.” (Paul Feyerabend, Against Method, 1975).
The very act of falsification has severe limitations, simply based on the fact that the technology and the methodological approach to falsification is limited by the paradigm in which the issue arrives (the Duhem-Quine problem)
And, in fact, the claim that falsification is the only way to find scientific truth is a dogmatic one, and is a particularly amusing dogma, considering that the scientific method has never been subjected to the scientific method, and is actually one of the “scientific” truths that is immune to itself. As one writer notes, when comparing verification with falsification:
“…Popper has put forward the general principle that attempts at falsification are most lilkely to prove scientifically fruitful. However, this is an empirical question and Popper does not provide any evidence to support his contention. Furthermore, if verification is not a logic and, therefore cannot be at the root of science as a heuristic tool, then it must be also accepted that falsification as employed in practice is not a logic either and can’t function as a heuristic tool. However, if striving for falsification is allowed not because it is a strict logic, but because it functions as a good heuristic tool for science, then the same must be allowed for verification, since it too is not a logic, but is claimed (by non-Popperians) to function as a good heuristic tool.” (Pearce N, Crawford-Brown D. Critical discussion in epidemiology: Problems with the Popperian approach. J Clin Epidemiol. 1989; 42(3):177-184).
In fact, one of the great failures of advocates of the “scientific method” as the only path to scientific truth is that the separation between scientific and non-scientific truth (often called the “demarcation” problem), is that its superiority is simply asserted as axiomatic. As another author wrote:
“To put it bluntly: because Popper has failed to provide any kind of rationale for the methodological rules he advocates, he has failed to give an adequate solution to the problem of demarcation, and to that version of the problem of induction which he would wish to claim he has solved, namely: What criteria ought to govern our selection of theories if out concern is to realize the fundamental aim of scientific enquiry? In addition he has failed to show that scientific enquiry can be viewed as a rational enterprise. In order to solve these three problems adequately it is essential to show that the advocated methodological rules give one a better hope of realizing the fundamental aim for science than any alternative methodological rules; and it is just this which Popper has failed to do.” ( Maxwell N. A critique of Popper’s views on scientific method. Phil Sci. 1972 39:131-152.)
In other words, the “scientific method” is not falsifiable, and has never actually been proven to be superior to verification. Religious dogma has replaced that kind of falsification about falsification.
In practice, while the scientific method often works, it is more often replaced by a dialectic method — because falsification requires faith in the test when falsification occurs and you think it shouldn’t, and that is usually not present. If I believe that gravity causes all things to accelerate to the ground at the same rate, and I go up on the Eiffel Tower and drop a bowling ball and a feather, they will not hit the ground at the same time. Why not? Because there are intervening issues (air resistance, wind, whatever). Thus, it would be wrong for me to consider the test as “falsifiying” gravity. Instead, I go back and try to remove more and more things that might be a problem, often *introducing* more issues in the process. At what point does adjusting for confounders make something no longer “scientific?” If saying that “Well, my prediction was wrong, but it’s not because my theory was wrong. I just didn’t set up my test right or measure things right, or whatever” makes something nonscientific, then very little “science” occurs in the real world.
But, in fact, a lot of science does get done. It’s just not done on the basis of falsification, but instead in terms of a dialectic between aggregates of verification and falsification data, or the “weighing of evidence.” The problem is that this idea that the “scientific method” is the *only* path to scientific truth has become what many consider a cult (see, for instance, Buck C. Problems with the Popperian approach: A response to Pearce and Crawford-Brown. J Clin Epidemiol. 1989 42(3):185-187).
“The philosophy of science is important to scientists as ornithology is important to birds.” – Richard Feynman
Popper is poop. If you want to die of boredom, listen to philosophers of science.
Falsifiability means that the hypothesis matches reality. It is possible to create a marvelous hypothesis which has nothing to do with the real life.
That is actually quite common and popular. Let’s postulate that whether is determined by the holy spirits. Rain dance works in some occasions. Sometimes more human sacrifice is required. It sounds plausible and 97% of experts will support it. How can you deny it? Is it enough that randomized double blind test does not pass.
Thank you for your thorough scouring of my crude history of the scientific method, brillo. (https://wattsupwiththat.com/2017/08/10/claim-climate-science-does-not-have-to-be-falsifiable/#comment-2579029)
You write, “Historically, “science” did not rest on falsifiability until the time of Popper.” But I didn’t say falsifiability was the foundation of science but that it was the sine qua non of science. I certainly didn’t mean to imply that the Popperian concept of falsifiabilty was in use, what I was trying to say the the law of non-contradiction was in use, a law which is one of the primary tools of the dialectic method which you so ably defend. For example the method of reductio ad absurdum is very important in dialectics and rests on the axiom of non-contradiction.
Aristotle, who is sometimes said to do no experiments, wrote in his Meteorology:
“The action of this cause [the admixture of earthy residues] is continually making the sea more brackish, but some part of it is always being drawn up with the sweet water. This is less than the sweet water in the same ratio in which the salt and brackish element in rain is less than the sweet, and so the saltness of the sea remains constant on the whole. When it turns into vapour it becomes sweet, and the vapour does not form salt water when it condenses again. This I know by experiment. The same thing is true in every case of the kind: wine and all fluids that evaporate and condense back into a liquid state become water.”
So by experiment Aristotle shows that evaporated and re-condensed sea water is not salty. He apparently performs the same experiment with wine and other solutions/suspensions which adds to the quality of his verification. He does not specify what the falsifiability observation would be, but you can be sure that if you asked him he would agree that should you perform the same experiment and the water turned out to be salty, that either his claim that the vapor becomes sweet was mistaken or there was some mistake made in the second experiment. He would not think that both results could represent the truth about condensed salt water vapor. Why? Because we know he thinks that the same attribute cannot at the same time belong and not belong to the same subject in the same respect.
Perhaps this all seems too simple and obvious to mention, but after all, these kinds of considerations are at the foundation of science — Popperian or not. Next time I won’t make the mistake of using the Popperian-loaded term “falsifiability” when I mean something like “results not contradicted by other results.”
Incidentally, I defended verificationism against falsificationism right here on WUWT a couple years ago:
https://wattsupwiththat.com/2015/05/29/when-will-climate-scientists-say-they-were-wrong/#comment-1948671
I think my reply to Monkton of Benchley in that thread supports your comments on the dialectical method.
RalphB:
You said “…..the same attribute cannot at the same time belong and not belong to the same subject in the same respect”
I have a simple “model” that attributes climate change solely to the reduction of SO2 aerosols in the atmosphere, which has been empirically tested and validated multiple times (Google “Climate Change Deciphered”).
Does the validation of this model exclude all other attempts to explain climate change?
That is, is it possible that another validated model could be developed for the same subject (climate change)?
Nice post👍 Keep up th great work