I think we have a disagreement about the number of problems. I agree with you at to the 0th iteration but in the particular example of alcohol, the externalitie

I think we have a disagreement about the number of problems. I agree with you at to the 0th iteration but in the particular example of alcohol, the externalities are so large, the perturbation is important and we should make sure the model is “as simple as possible, but no simpler”. So your statement “If someone is being belligerent or otherwise disruptive in a public space, whether or not they're drinking alcohol should not be a factor” Please find a different example, or, if you believe in the goal of improving human society, please revise that statement. Belligerence and other disruptiveness involving alcohol should be tracked, because it can reveal the deeper problem that is present in some (honestly, many) cases of public belligerence and disruptiveness once a trend is established. Saving lives is important all on it’s own.