Grant McDermott bio photo

Grant McDermott

Data. Economics. Environment.

Email Twitter LinkedIn   Scholar   ORCID GitHub
So says Ivo Vegter in his latest column, calling up the spirit of Karl Popper.

I drop by in several places in the comments, saying that this pretty much utter nonsense according to any reasonable definition of the falsifiability criteria. For example:
[...]As it happens, Ivo has also beautifully misconstrued the application of Popper's insights, by consistently conflating the actual science with the response to that science by various parties.

[...]At its heart, falsification is about using an underlying set of theories to make predictions that can be tested against the relevant evidence. The theory of (man-made) climate change satisfies the falsifiability criteria across multiple dimensions. The fundamental pretexts of climate science are rooted in physics that has been understood and tested since the early contributions of Fourier, Tyndall and Arrenhuis in the nineteenth century. For their part, modern day climate models absolutely meet the Popperian standards. Not only do they forecast how climate phenomena will evolve under conditions of continued emissions, but (crucially) they have been extremely successful in hindcasting previous changes to the climate. Indeed, this accurate replication of past events is how they are selected in the first place…
As part of his reply to this comment, Ivo made the strange assertion that the ability to hindcast models is weakened by the fact that these models can be "tweaked" to accurately match the observed data. I respond:
How is it a case against climate scientists that they were able to benchmark their theories against against a historic record with "perfect hindsight"? Speaking as someone with a fairly considerable amount of scientific training, that actually sounds pretty ideal. Models were back-tested, refined and then discarded in favour of ones that better fit the data... pretty much in exact accordance with the scientific method.
For the record, I am well aware that Popper's falsification doctrine is not without its problems. Still, and at the very least, if someone is going to invoke his authority to make specific claims, then they should at least make sure they are accurate within that framework.

UPDATE: Oh God, it gets worse. Someone responds to me with the immortal line: "I have not done the research but from my own basic calculations..." *headvice*

UPDATE 2: See this excellent 2005 post by NASA's Gavin Schmidt: Is Climate Modelling Science? It concludes: "So, in summary, the model results are compared to data, and if there is a mismatch, both the data and the models are re-examined. Sometimes the models can be improved, sometimes the data was mis-interpreted. Every time this happens and we get improved matches between them, we have a little more confidence in their projections for the future, and we go out and look for better tests. That is in fact pretty close to the textbook definition of science."