[quixoticaxolotl]

I share eqdw's distaste for knee-jerk symptom addressal without adequate understanding of the cause.

I am somewhat disappointed by the other responses to eqdw here because they each immediately proffer or refer to competing interpretations of the cause without also referencing empirical tests of those interpretations. They miss eqdw's excellent point that there's always a hand-waved explanation waiting in the wings. Interpretations (which the other comments are) are not true explanations: they add dimensions to how we think, but not certainty in our conclusions. The historical method is not a valid substitute for the scientific method, as E. H. Carr long ago demonstrated with "What Is History?"

The point I'm making is that there's got to be a higher bar for reasoning about social issues than a preferred interpretation, and that is likely going to be experiment. I'm aware sociological issues are so-called wicked problems, are hard to replicate solutions consistently with, and are difficult to have error bars over in their contribution to the end result. I don't think that means it is not possible to do experiments well enough to make reliable inferences - simulated games and follow-up interviews with participants are an effective research technique in this area, for example.