清华科史哲讲座第27讲
时间：2019年7月31日（周三）下午15:00–17:00
地点：科学史系系厅（蒙民伟科技大楼南楼212）
主讲人：Marius Backmann (London School of Economics)
讲题：Time and the Old Riddle
内容简介：
In my talk, I argue that whatever stance one might take on whether a justification of induction is at all possible, inductive inferences to future instances are more problematic than inferences to past of unobserved instances. This is particularly the case for any justification of induction that aims to show that induction is in any way truth-conducive, and relies on a set of rules for the sampling, such as the Law of Large Numbers-approach [LLN]. As we will see, though, the arguments presented here are not restricted to the LLN-view.
Inferences to the future are more problematic than inferences to the merely unobserved due to two issues – one general, and one that applies only to temporal ontologies with an ontologically open future. First, there is the problem that the sample from which to infer information about the population is necessarily not random, but spatiotemporally ordered and restricted, which does not meet the standard for how a sample should be composed in these views on induction. Second, there is the issue that if one proposes a temporal ontology with an open future, it is unclear what we take the population to be. If future facts are not equally real as the present (or past) facts, then the population from which we draw the sample is different from the one we want to infer any information about. No matter how large the sample, we not only have to argue that the sample resembles the population, but also that the population we drew the sample from resembles the final population, which then includes the instances that were not real at the time we took the sample. Related to this issue, temporal ontologies with an open future present a further challenge to the LLN and classical probabilistic accounts of justification: typically, these accounts formally require that the population must be finite. But in the case of presentism and the growing block view, there is no fact of the matter as to whether the population is finite or not.
演讲人简介：
Marius Backmann is currently working as an LSE-fellow at the London School of Economics.
His work focusses on ontologically motivated attempts to solve the problem of induction, philosophy of time, metaphysics of science, laws of nature, and free will.
