Week 5: Problems of Induction

We will continue our discussion of the problems of induction next week, starting class on Tuesday by making sure we are clear about the argument for inductive skepticism, its importance, and some of the potential responses to it discussed in Foster (please bring that reading again). I expect to spend most of our time on Tuesday talking about the justification problem and potential solutions.

As I mentioned, it turns out that justifying our inductive practices is only half the battle. Even describing them seems challenging. This was roughly the problem we faced with H-Dism: the simple formulation of that method seemed vulnerable to technical objections. It turns out that there are general worries about our ability to describe our inductive practices. We’ll discuss two: Goodman’s “New Riddle of Induction” and Hempel’s Ravens Paradox. I expect to only get to the Ravens Paradox on Tuesday — we’ll save the “New Riddle” for Thursday. Note that while the Goodman reading is optional, it is great. That is not to say that the ideas are easy: but it’s definitely worth reading (if nothing else than as an example of lovely, simple prose). 

Reading for the Week: (Good thing I scheduled Thursday as a day for “breathing room”!)

• Goodman, "The New Riddle of Induction" [PDF]*

• Lipton, "Induction" [PDF]

• Godfrey-Smith, "The Ravens Problem" (§3.3 from his Theory and Reality)*

Questions: (respond to one; I’ll post more questions for Thursday that arise from our discussion.)

1. According to Lipton, what is “underdetermination”? How does it play a role in arguments for inductive skepticism?
2. Is the inductive skeptic trying to show that the inferences we often make in science are bad or that we need new methods for making inductive inferences? If not this, what is the inductive skeptic trying to do? Explain the intended conclusion of the inductive skeptic’s argument. How damaging is this conclusion.
3. Explain clearly what the difference is between the descriptive and justificatory problems of induction.
4. Lipton’s statement of the Ravens Problem for the Instantial Model is brief. But see if you can piece together the argument more specifically.