OK, but lots of people would disagree with you, e.g., in the context of supervised injection sites. Suppose it were unambiguously clear from data that (1) soci

OK, but lots of people would disagree with you, e.g., in the context of supervised injection sites. Suppose it were unambiguously clear from data that (1) society as a whole and each individual were better off when marijuana was legal and tightly regulated; and (2) individuals would be better off not using marijuana. Would you then vilify marijuana providers? (I generally would not, although I would hold them to a stricter moral standard than I would for sellers of products that are generally beneficial to the user.) What if we replaced assumption (2) with assumption (2')?: 80% of individuals who use marijuana are better off (say, because they use it only socially and the negative motivational effects turn out to be small), and 20% are worse off (because they abuse it).