Jumping To Conclusions

Peter overlooks a possibility; it may be that Christina Vrachnos is more subtle than the reporters at the New York Times. She might have been attempting irony which went over the reporter's head. Knowing the media's penchant for hyping the global warming issue, I often sarcastically blame the weather conditions -- whatever they are -- on global warming. Cold outside? Maybe it's because of global warming. There was an earthquake? Maybe it was due to global warming. The Cubs are in first place? Maybe it's global warming. I can certainly picture an overly earnest New York Times reporter overhearing such a comment and eagerly repeating it in a column. They want to believe it, after all.

And on the subject of global warming, there's an excellent post by Iain Murray, generally at the Edge of England's Sword, but guest-blogging over at the Volokh Conspiracy, in which he points out what a truly objective media would:

I've been dealing with climate science issues in detail for approaching a month today. As a result, I am amazed whenever I hear anyone say that "science shows" anything in the climate change debate. The plain fact is that normal scientific methods simply aren't applicable in the climate science area. Normally, you come up with a hypothesis and run experiments to check it. The trouble is that you can't run experiments with the climate. We have no other Earth to act as a control (anyone who points to Venus as an example is showing his ignorance there and then). The science can therefore only progress by building models, which, if acceptably accurate, might predict what will happen. But those models are based on theory. If they cannot predict what is currently happening (as we know from observation) accurately, there is something wrong with them and/or the underlying theory. Theorists, however, are often wedded to their theories.

Exactly; if you read popular coverage of the global warming issue, all too often scientific models and scientific laws are confused. We hear talk that newly-collected data has enabled scientists to refine their models, and shows such-and-such about the future, when what is really the case is that plugging the newly-collected data into the model shows such-and-such, if that model is accurate. But a newly refined model can never be said to be accurate; only by seeing if the model accurately predicts the future can we make that determination, which means there must necessarily be a time lag between scientific understanding and public policy changes. But because this doesn't fit some agendas, the distinction between "the models suggest X" and "science suggests X" gets lost.

I don't mean to suggest that there's anything dishonest about changing the models to explain new data; that's a perfectly reasonable scientific practice. The problem is changing the models to explain past data and then pretending that the models represent established science, as opposed to mere hypotheses.