Peter Lipton, “Inference to the Best Explanation”

Sources

• Peter Lipton, “Inference to the Best Explanation” (opens in a new tab)

  Peter Lipton, “Inference to the Best Explanation,” 2nd ed. (Routledge, 2004)

Objections & responses

Lipton’s “loveliest explanation” invites exactly the bias this claim denies. People find explanations lovely when they already like them. Religious believers will find religious explanations lovely.

Loveliness is Lipton’s technical term for explanatory depth and unification, not aesthetic appeal, and he argues at length that it tracks likeliness. The bias worry is real but symmetric: naturalists find naturalistic explanations lovely. The corrective is public criteria, which is why the arguments on this site try to state what each hypothesis explains and what it leaves out, so that you can check the scoring rather than take the verdict on trust.

Lipton himself noted that inference to the best explanation can give you the best of a bad lot. Even if the resurrection is the best explanation of your facts, that doesn’t make it probable.

The bad lot objection comes from van Fraassen, and Lipton answered it. The worry applies to every use of abduction, including in science, and the standard response is to ask whether the field of hypotheses has been searched thoroughly. In the resurrection case, two millennia of skeptical effort have produced a small, well-known set of alternatives (hallucination, swoon, theft, wrong tomb, legend, visionary origin), which is about as thorough a search as any hypothesis gets. If the lot is bad, it is not for want of looking.

Appears under 1 claim in this case

• Inference to the best explanation is a recognized form of reasoning used across science, history, law, and medicine.