劳丹悲观归纳中的两个统计学错误

Two Statistical Fallacies in Laudan’s Pessimistic Induction

劳丹借助科学史对科学实在论的观点进行反驳,特别是通过一组曾经取得成功但无指称的理论列表来否认成功和近似真理之间的关联。为了使反驳有效,劳丹归纳出的列表必须具有统计学的意义,但同时又犯了统计学错误。刘易斯认为悲观归纳犯了假阳性错误,米兹拉西认为悲观归纳犯了样本错误。米兹拉西同样以科学史为依据提出了一种支持实在论观点的乐观归纳,乐观归纳能同时避免以上两种错误,比无奇迹论证更有力。

Laudan refutes the viewpoint of scientific realism with the help of the history of science.In particular,he denies the relationship between success and approximate truth by generalizing a plethora of theories that were both successful and nonreferential with respect to many of their central explanatory concepts.To make the argument valid,Laudan’s list must be statistically significant,but at the same time it makes two statistical mistakes.Lewis points out that there is a false positive fallacy in Pessimistic Induction,and Mizrahi points out that there is a sample fallacy in Pessimistic Induction.Based on the history of science,Mizrahi provides an optimistic induction which supports the viewpoint of realism.Optimistic induction can both avoid the two fallacies,and scientific realists would be better off relying on an optimistic inductive generalization,rather than the No-Miracles argument.