Certainly interesting--wish the article mentioned the Pearlian techniques. My issue is with the overstated claims, e.g. "Statisticians have always thought it

Certainly interesting--wish the article mentioned the Pearlian techniques. My issue is with the overstated claims, e.g. "Statisticians have always thought it impossible to tell cause and effect apart using observational data. Not any more." Maybe including "bivariate" in the paper's title could have helped.