Anthropogenic Global Warming (AGW) alarm has been with us for a good while, now. The matter seems to become more contentious, rather than less, over time. Unhappily, as a result of the mediocre quality of science education, many people do not know how to evaluate either a scientific hypothesis in general, or AGW in particular -- and irrespective of whatever anyone might think, because of how it is framed and evaluated, AGW is no more than a hypothesis. 

Science is about ruling things out. Any good scientific hypothesis will make predictions about the natural world -- ideally, it will predict at least one natural effect whose existence cannot be caused by anything other than the hypothesis being tested. Observations are then made to acquire evidence, and the evidence is evaluated against the hypothesis’s predictions. Evidence can either rule the hypothesis out or not; if the evidence differs from the hypothesis’s predicted effects, then the hypothesis is wrong and is considered to be ruled out, or falsified. That which has not been ruled out by evidence remains possible. If enough confirmatory evidence is accumulated, the hypothesis is elevated to the status of a theory. Scientific Method is, conceptually, no more complicated than that.

Karl Popper, the great philosopher of science, used a simple observational experiment to illustrate the scientific method’s requirement of falsifiability -- the requirement that a hypothesis be stated in such a way as to allow its testing against evidence with a view towards ruling it out. He noted that most people had once assumed that all swans are white. This assumption was based on the observation, over time, of uncounted numbers of white swans -- and each such observation was taken as evidence supporting the assumption. However, there came a time when a black swan was found in Australia, and its discovery served to disprove the assumption that all swans are white. In generalizing from this discovery, Popper understood that you would not test the hypothesis that all swans are white by undertaking a search for white swans -- because no matter how many white swans you found, you would neither have proven, nor even properly tested, the hypothesis. Instead, you must mount an intensive search for a single non-white swan.  If you found even one of those, you would have ruled the hypothesis out. Alternatively, and without finding a non-white swan, it remained viable -- but because there remained the possibility of a single undetected non-white swan, it could not be regarded as proven.

Einstein's Theory of Special Relativity provides an excellent real-world scientific example of evaluation by falsifiability. The Special Theory makes unique predictions about gravity's effect on light's behavior in a vacuum that, as far as anyone knows, could be accounted for by no phenomenon other than that assumed in the theory. When specifically tested for during a total eclipse of the sun in 1919, the gravitational effect Einstein's theory predicted was both detected and measured to equal precisely his theory’s prediction. Special Relativity was hence verified -- although, again, it is not regarded as proven. Instead, it remains possible in the absence of having been falsified by evidence. Now, it is true that Special Relativity is, like other theories, commonly accepted, and spoken of, as having been proven. However, that is merely a shorthand way of saying that it currently has no credible competition as an explanation of the phenomenon it addresses.

The AGW hypothesis that so many people claim accounts for what is essentially pretend global warming has never been treated this way. Initially, its proponents engaged in a search for supporting evidence: Elevated average annual temperatures, local glacial retreats, elevated-temperature indicators in proxy systems such as tree-ring records, measurable coincident increases in atmospheric CO2 concentration, and so on -- a search for white swans. But these efforts ignored, and failed even to seek, either any alternative explanations or evidence that would have ruled the hypothesis out. AGW has failed the predictions test again and again; any true scientific hypothesis with so poor an evidence-based evaluation record would have been scrapped by now. Instead, its proponents elevated it to the status of a theory and, ignoring the fact that climate changes continually, renamed it “climate change.”

No other potential causes of AGW have ever been investigated and ruled out. There must be at least one, because evidence shows that there have been times in the pre-human geological past when conditions were warmer and there was no glaciation at all anywhere on Earth. We also know, as a result of ice-core studies, that CO2 has generally been a lagging indicator -- that is, atmospheric CO2 concentrations are documented to have increased after, rather than before, atmospheric temperature increases.

Nevertheless, its believers treat AGW as verified, and simply alter its components and predictions to conform to evidence. When the predicted warming did not occur and snows continued to fall during London winters even though it was predicted that they would fail, for example, or when polar ice sheets expanded even though the theory has predicted that they would melt away, the hypothesis should be considered to have been ruled out by evidence. 

However, its proponents still treat AGW as though it were true. Otherwise-reputable scientists employ variations on several approaches to their falsification conundrum. The first of these approaches, the use of models, is a legitimate tool in particular scientific applications. Others amount to attempting to fudge the hypothesis to make it match evidence in an unscientific rearguard action.

Models are essentially used as predictive tools, so they are only as good as the information upon which they are constructed. If there are any unknown components in the modeled system, then the model’s predictions will, almost by definition, be unreliable.  In the case of a system both as complex and incompletely understood as Earth’s atmosphere, the model’s construction will essentially be required to include untested, incomplete, and/or unproven function assumptions and data. In such a case, the problems and pitfalls of using these models to construct governing policies quickly become self-evident: People trying to rely on the models essentially cannot know what they are doing.  When, for example, their model does not predict their real-world observations, they tweak it until it does -- which introduces errors-by-expectation into both output and the policies based upon it. These errors increase in magnitude, and therefore in effect, in a non-linear fashion directly proportional both to the size of the system and to the modeled outputs. 

AGW’s predictions are not being reliably confirmed by observations. When stasis and/or cooling occur rather than warming -- as has been the case over the last decade-and-a-half -- atmospheric scientists fudge interpretations by saying that if it is cool, well, that is just weather; if it is warm, though, that is climate.  Alternatively, they claim AGW predicts the cooling -- as, for example, with the recent polar-vortex outbreaks.  However, a theory that predicts everything predicts nothing -- because a theory that predicts everything cannot be falsified through testing; nothing will serve to rule it out.

Scientists have also approached the unaccountable stasis and/or cooling by going around and searching for "the missing heat" that their theory assumes exists and claims has already built up.  But this is not a search that would test the theory. It is a search that assumes the theory to be true -- it begs the question. Further, if the search detects the sought evidence, no one tries to rule out any possible causes other than AGW, assuming instead that if the evidence exists, there are no other possible causes.

In short, the AGW -- cum -- “climate change” debate is not about a hypothesis -- cum -- theory.  Even though no one has investigated it with a view towards falsifying it, evidence has ruled it out repeatedly.  It has no useful scientific applications because it has been broadened to predict all possible observations -- thereby predicting nothing at all.

Betsy Gorisch is a professional geologist with an interest in current events.