About a book by sociologist Duncan Watts, I write,
Watts’ book can be regarded as an extended argument in favor of what I might term Epistemological Skepticism about Social Phenomena, or ESSP. Those of us with ESSP believe that we should be skeptical about how much we can know with certainty in the fields known as the social sciences. We may learn things that are true for a majority of cases under specific circumstances. But we are less likely to find perfectly reliable, broadly applicable laws comparable to those found by physicists.
I’ve long felt that “social science” should be renamed “social research”–and I’m in the field. This is not a slight. I think the word “science” applied to social research is inaccurate. Experimentation is either not possible, or possible but limited–external validity issues. This places an unfair burden on researchers when comparisons are made to natural and physical science. Because of our inability to experiment, it is difficult to find robust laws (if they even exist), and most social theories are fragile, if not outright wrong.
I think that’s well said. By my poor lights Jim Manzi was on to something in the opening chapters of Uncontrolled when he compares the knowledge of social science, or research, to that of the specialist Historian (of Iran, in Manzi’s hypothetical). Social research has long suffered from a positivistic impulse to achieve what physics is thought to have achieved: an independence or replacement of other ways of knowing about human nature and society (e.g. literature, metaphysics, historiography, and so on) rather than as a handmaiden or helpful critic to them.
Don’t physical sciences suffer from some of the same issues though? Newtonian physics works well for the relatively large objects we deal with everyday, but doesn’t work well at all at the subatomic level. Quantum mechanics explains subatomic motion, but is too complex to be useful in understanding “macro” objects.
I’m not a fan of modern macroeconomics, but wouldn’t stripping the “science” label from economics and social science in general mean we’d have to strip it from fields of research like meteorology, oceanology, geology, astronomy, and evolution? Those are fields where experimentation is impossible or limited, too. I’d say microeconomics is to the social sciences what physics is to those “hard” sciences: known universal laws derived from experimentation that constrain the theories of the more observational branches.
Rather than limiting the use of the word science, perhaps we should teach our children that there are two categories of science: experimental and observational.
Reminds me of the Ulam challenge to find something in the social sphere that is true and non trivial.