Experimentum crucis syndrome II

[mmcirvin]

RealClimate eviscerates a Wall Street Journal editorial about global warming. The commenters bring up something I've noticed before, the tendency for fringe critics to assume that an entire scientific field of study rests on one or a small number of historical results that they can then knock down (in this case it's the Mann et al. "hockey stick" paper arguing significant 20th-century warming after centuries of relative flatness).

They notice that global-warming denialists tend to do this, but it's actually far more general. Mathematical cranks do it; it's why creationists spend so much time talking about Haeckel and Piltdown Man; Velikovskians behave as if their opponents are all 19th-century uniformitarians like Charles Lyell; Einstein-refuters spend countless hours looking for irregularities in the Michelson-Morley experiment. When the field under attack has a towering historical progenitor, they sometimes work in a story about a deathbed recantation, as if that would mean anything (I recently read that alternative-medicine fan Bill Maher claims Pasteur recanted). The same patterns come up over and over.

Comments

[swinehund.livejournal.com]

It seems kind of like an attempted ad hominem attack, if you treat an entire scientific discipline as the individual. I wanted to suggest something further like that the dissenters themselves might think this is an appropriate strategy because they often tend base their beliefs on cults of personality (and thus their own beliefs might be undermined by discrediting the great ideological leader who originally espoused them), but that is not always the case.

[swinehund.livejournal.com]

Though the simplest answer is just that they have not been keeping up with the current advances in the field, and really do not know how far the original ideas have been developed and improved upon.

[mmcirvin.livejournal.com]

Often the pseudo-history of the field imagined by the attacker corresponds closely to what you'd get by just reading popular and introductory textbook accounts, especially if your mind wanders after the opening pages. In fact, I think scientists share some of the blame for being sloppy about their own history and repeating conventional accounts that give short shrift to the field beyond the historic crucial results.

[mmcirvin.livejournal.com]

...There's another thing going on. If a new scientific theory is going to replace or subsume an earlier successful theory, it has to explain why the previous theory was empirically successful in its domain of application. Cranks typically can't do this. Sheldon Glashow (who must get a lot of crackpot mail) once pointed out in a class lecture that in the theoretical realm, this is probably the most immediately noticeable difference between crank theories and non-crank theories.

So instead their tack is to insist that the previous theory was not empirically successful, because everyone was making the same, usually elementary error. The task then becomes explaining the old theory's political success, which comes down to some sort of basically ad hominem argument about venal motives or ideological pressure. You see this a lot in popular books promoting fringey theories that haven't cut it in the peer-reviewed literature; the book will spend a lot of pages talking about the sociopolitical origins of the blinkered dogmatism that is keeping people from adopting this wonderful idea.