I suppose it's better to say "roots," plural. Nobody knows what lies at the roots of crime for _certain_, but there has been a lot of research into the topic. S

I suppose it's better to say "roots," plural. Nobody knows what lies at the roots of crime for _certain_, but there has been a lot of research into the topic. Several things seem to be strongly correlated with increased risk for criminal behavior, but we can't draw direct causality relationships quite yet. The Wikipedia article on it is informative, and viewpoint-neutral: https://en.wikipedia.org/wiki/Statistical_correlations_of_criminal_behaviour https://en.wikipedia.org/wiki/Statistical_correlations_of_cr... One of the more interesting possible roots of criminal behavior is lead exposure as a child. It's not entirely clear, of course (socioeconomic status is a confounding factor that is hard to eliminate), but there seems to at least be some strong correlation: https://en.wikipedia.org/wiki/Lead–crime_hypothesis https://en.wikipedia.org/wiki/Lead–crime_hypothesis My point isn't that we should be redirecting more resources to lead removal than to recalling a prosecutor (even though I _do_ think lead removal is a wise use of resources). My point is that deterrence and punishment is "cleaning up after the fact," rather than trying to address problems at their source. The question "what should society do with criminals" implies a belief that criminality _must_ exist in a society. Perhaps to some degree, it is true. But it seems as though redirecting our efforts to prevent criminality in the first place by addressing the underlying root causes would ultimately be a far better use of our resources, and far more effective.