In this essay, I address an objection to inference to the best explanation due to Bas C. van Fraassen, according to which explanatory virtues cannot confirm a theory, since they make the theory more informative and thus less likely to be true given the probability axioms. I try to show that van Fraassen’s argument, once made precise, is deductively invalid, and that even an ampliative version of the argument (i) implies, absurdly, that no theory is confirmed by its fit with empirical data; (ii) fails to account for confirmatory closure under deduction; and (iii) falsely presupposes that a theory and its sub-theories can be competing explanations.

中文翻译：

---

解释性美德和信仰理由

在本文中，我提出了 Bas C. van Fraassen 对最佳解释推论的反对意见，根据该观点，解释性优点不能证实理论，因为它们使理论提供更多信息，因此在给定概率公理的情况下，它们不太可能为真。 。我试图证明，范·弗拉森的论证一旦变得精确，在演绎上就是无效的，而且即使论证（i）的放大版本也荒谬地暗示，没有任何理论能够通过其与经验数据的拟合得到证实；(ii) 未能根据扣除说明确认性结清；(iii) 错误地假设一个理论及其子理论可以是相互竞争的解释。