This paper explains how the "no miracles" argument escapes the objection of committing the base-rate fallacy, and it shows how the argument must be refined in order to rebut various other objections The “no miracle" argument (NMA) is generally considered the “ultimate” argument for scientific realism (Musgrave 1988). However, its roughest formulations (e.g.: “The success of science would be a miracle unless scientific theories were true”; “the only non-miraculous explanation of the success of science is scientific realism”) are as vague as open to many criticisms. Therefore, realists have already introduced several refinements to fend the objections which in fact have been raised. Some antirealists even concluded that this attempt to immunize it from objections is an endless and hopeless process, but I maintain that this is not the case, since there is at least one effective and undefeated form of the NMA. To show this, I begin by briefly discussing four refinements that have already been introduced (§ 2). Next, I introduce the “base-rate fallacy” objection, which until now has not received sufficient attention from realists (§ 3). Before answering it, I need to present yet another objection and the further refinement it calls for (§ 4). In § 5 I argue that the truth-conduciveness of the scientific method warrants a nonzero prior probability to hypotheses, and in § 6 I explain that for this reason the NMA avoids the base-rate fallacy. Finally, in § 7, I briefly reply to some further objections.
M. Alai (Wed,) studied this question.