What, precisely, does Ockham's razor recommend that we believe, and (why) ought we follow it? In Ockham's Razors: A User's Manual, Elliott Sober attempts to answer this question, revealing a fascinating variety of meanings, approaches, and justifications for arguments that purport to be about parsimony. Along the way, readers are introduced to some of the recent work on the philosophy of statistics, especially as it is used in discussions of what constitutes good scientific reasoning.
Ockham's razor is usually interpreted to advise something like “all else being equal, the simpler explanation is to be preferred.” But what exactly is this recommending, and why ought we follow that recommendation? Is it supposed to be more likely to lead us to “truth” (because simple explanations are simply more likely to be true, all things considered, than complex ones)? Or is it supposed to make our hypotheses more testable (because simpler...
