15th European Conference on Artificial Intelligence
|July 21-26 2002 Lyon France|
This paper develops a qualitative, logical, theory of induction. It begins Hempel's attempt to produce a ``purely syntactical'' theory of confirmation and the demise of this attempt as a result of Goodman's paradox. Ideas from the informal, pragmatic, solutions to this paradox proposed by Goodman and Quine are then adopted, adapted and extended in order to produce a formal, pragmatic, theory of induction. According to this theory, induction takes place in an evolving context of inference; which includes an evolving system of kinds and, typically, a background theory. The theory is illustrated by giving a formal solution to Goodman's paradox, and a further difficulty raised by Davidson is discussed.
Keywords: Inductive Logic, Philosophical Foundations, Induction
Citation: John Bell: A Pragmatic Theory of Induction. In F. van Harmelen (ed.): ECAI2002, Proceedings of the 15th European Conference on Artificial Intelligence, IOS Press, Amsterdam, 2002, pp.564-568.