归纳的非概率式辩护及其难题

Non-Probabilistic Justification of Induction and Its Difficulties

归纳的辩护可以划分为概率式辩护和非概率式辩护两种不同的形态。概率式辩护由于得到数理统计理论支撑而得以成立,然而概率式辩护并不足以承担整个为归纳辩护的职责。作为补充,归纳的非概率式辩护并不成功,不论是皮尔士所说的"归纳具有自我修正性"还是刘易斯的"先验分析式"辩护最终都没有说明归纳推理的有效性。

There are two theoretical models for justification of induction,that is,probabilistic justification and non-probabilistic justification.According to our analysis,probabilistic justification is justified on the basis of mathematical statistics,but it doesn't determine sufficiently the validity of justification of induction.On the other hand,as a supplement of probabilistic justification,non-probabilistic justifications,such as Peirce 's so called 'self-correcting nature of induction' and Lewis's 'a Priori analytical justification of induction',do not serve as the genuine supplement of probabilistic justification.