On a blog I’ve just started reading ardently, Venkat Rao explains why he doesn’t like the scientific method:
I don’t like or use the term scientific method. Instead, I prefer the phrase scientific sensibility. The idea of a “scientific method” suggests that a certain subtle approach to engaging the world can be reduced to a codified behavior. It confuses a model of justification for a model of discovery. It attempts to locate the reliability of a certain subjective approach to discovery in a specific technique…
…The scientific method is a sensibility crammed into the mold of a system. It is a an attempt to externalize something subtle and internal into something legible and external. The only reason to do this is to scale it into an industrial mode of knowledge production, which can be powered by participants who actually lack the sensibility entirely. Such knowledge production has been characteristic of the bulk of twentieth century science (in terms of number of practitioners, not in terms of value). Hence the Hollywood stereotype of the scientist as a methodological bureaucrat; someone who worships at the altar of a specific method. Sadly, Hollywood gets it right. The typical scientist is a caricature of a human.
Though I’m nothing if not an opponent of vague terms like TSM, we need to be careful here. I’ve had countless discussions on this, and Venkat makes TSM seem more rigid and emotionless than its adherents intend it to be. More often than not, TSM is simply a synonym for the notion that problems should be studied rigorously and with care. It’s the recognition that there are better and worse ways to study certain questions even if we allow that there is no step-by-step blueprint. I can agree with Feyeraband’s methodological anarchy, disdain Hollywood’s mechanized portrayal of scientists, while also believing that the phrase TSM is useful.
Personally I would be happy to get rid of the term. But we must wrestle with how scientists actually think of and use TSM rather than its public caricature (which I feel most scientists disagree with already). We’re trying to have a semantic argument without engaging in the boring, messy work of semantics. I suspect that if we did so, there’d be much common ground between the scientific method and the scientific sensibility.