Abstract This note offers a response to Bakshi’s critical discussion of Schurz’s claim that meta-induction provides a non-circular justification of object-induction (induction at the object-level of events). Bakshi raises two issues: 1. The problem of identifying object-inductive methods, and 2. the problem of determining the pool of simultaneously accessible prediction methods. In this reply, a solution to the identification problem is offered, based on existing work. Concerning the problem of pool-determination, induction-independent criteria are proposed that justify disregarding Goodman-type methods.
Gerhard Schurz (Mon,) studied this question.