I agree that the aesthetic interpretation, “simplicity for simplicity’s sake,” is what most often motivates adoption of Occam’s Razor. But what justifies it is the falsifiability criterion. As highlighted by Popper, we should prefer simpler explanations because the empirical content is greater; simpler explanations are better testable.

But this justifiable application of the Razor is frequently forgotten, stripping it of its utility and leaving us with empty promises.

(With much more space, a similar portrait can be drawn of Bayes: Great tool, stripped of context, loses its utility and becomes bad philosophy.)

Thanks for this article. We need more contrarian thinking.