IS POST HOC SOMETIMES EVIDENCE?–THE MIME EXPERIMENT. I tend to rant against the many examples of post hoc propter hoc arguments (after this, therefore because of this). The fallacy is similar to the fallacy of taking correlation to be causation. Nevertheless, the fact that event B occurred after event A is sometimes persuasive evidence.
The Economist (July 11) had an article about Bogota, Colombia’s attempts to use street theater to persuade citizens to be more polite. The Economist points out that this experiment follows up on actions by Antanas Mockus, a mathematician who served two terms as mayor between 1995 and 2003. Mockus replaced the traffic police with 420 mimes who took up traffic control activities.
The Economist says: “There is no real evidence that such tactics did much good, but they apparently did little harm.” A decline in traffic deaths would be a typical post hoc propter hoc bit of evidence. However, the drop in traffic deaths was quite large—from 1300 deaths a year to 600 deaths a year.
If there are not alternative explanations of a drop from 1300 to 600, I find myself thinking that the drop is evidence that the tactics had some success.