Thanks to Mark Garrison for the link to this N of 1 article. Science is the search for the causal structure of the world, and the history of science shows clearly that, while randomized trials can be useful, they are not necessary to gain such causal knowledge.
