How not to avoid the problems of old evidence and normalizability for fine-tuning
摘要
Standardly formulated fine-tuning arguments for multiverse or design hypotheses take the fact that our universe is life-permitting as evidence and the fine-tuning requirements as background information. In so doing, they face both a problem of old evidence and a problem of normalizability. Alternative formulations purport to evade these problems by taking the fine-tuning requirements as evidence and the fact that the universe is life-permitting as background information. Leveraging an objection to fine-tuning arguments originally put forward by Cory Juhl, I argue, by contrast, that proponents of these alternative formulations cannot justify the likelihood comparisons they make without confronting these problems.