Anything goes -- the soft sciences abandon any pretense of method

A piece in the Sunday New York Times, called "Overcoming 'Physics Envy'", by Kevin Clarke and David Primo, political scientists at Rochester, addresses the question of whether the social sciences are in fact science.   

Economists, political scientists and sociologists have long suffered from an academic inferiority complex: physics envy. They often feel that their disciplines should be on a par with the “real” sciences and self-consciously model their work on them, using language (“theory,” “experiment,” “law”) evocative of physics and chemistry.

What makes a field 'science'?  It turns out that following the scientific method makes it science -- proposing a model, devising a hypothesis and testing it.  Called hypothetico-deductivism, it's the approach that the National Science Foundation has touted in a big way in the last decade.

But Clarke and Primo no longer accept that this is the way to do social science.  Economists, political scientists and psychologists no longer have to pretend to be physicists to do their work.  They've matured into their own way of being. And anyway, the hard scientists don't even really follow their own method, so why should the soft ones?  So, now it's ok to think up models and put them out there, unsupported by any empirical data.  And, conversely, it's ok to have data without a theory (which sounds suspiciously close to the current genetic model of non-hypothesis driven science).  So, say Clarke and Primo, "social scientists would be better off doing what they do best: thinking deeply about what prompts human beings to behave the way they do."

This is fascinating.  The point of having an agreed upon method or approach to understanding a particular corner of the world is to standardize the assessment of cause and effect.  What's the point of the social sciences?  To explain why people behave as they do?  To predict future behaviors?  While it's perfectly clear that economics, political science and psychology don't adequately fulfill either of those goals as currently practiced, it's also true that abandoning any pretense of a method won't do it either. Is this basically an acknowledgement that these fields haven't accomplished what they've set out to accomplish?  And can't?  

We frequently make the point that the hard science of genetics should be held responsible for its claims and for how grant money is spent.  The same of course should be true for the soft sciences.  When hundreds of millions of tax dollars are happily spent every year by people who claim they'll give us something for it, shouldn't they be required to show that they have?  A publication count doesn't do it.  The idea that we should now be paying people to "think deeply" -- and that should be based on what criteria, political viewpoint, or foundation of knowledge, all of which will make a difference? -- is frightening. At least when the philosopher of science, Paul Feyerabend, said "Anything goes" with respect to the scientific method and what science can offer, this was based on an honest assessment of the possibilities.

Evolutionary biology has the flexibility of social science in that because evolutionary processes, like those in society, are statistical reflections of distributions of behaviors and traits, almost anything can be fit into its overall theory.  That is the power of the Darwinian method, and its weakness.  Likewise, the statistical nature of social processes does not lead to very precise predictions, and anything once observed can be 'explained' by free-wheeling retro-hypotheses.

Nobody can seriously claim that our society is now in good social or psychological health because of research in the social sciences, whatever their successes.  It can't get any better if the future is "thinking deeply".