Paul Horwich is Professor of Philosophy at New York University. He is best known for his contributions to the philosophy of science, including a probabilistic account of scientific methodology. His most recent publications include Wittgenstein's Metaphilosophy (2012) and Truth - Meaning - Reality (2010).

Paul Horwich's influential work offers resolutions to central issues in the philosophy of science through a probabilistic approach to scientific reasoning and methodology. Presented in a fresh twenty-first-century series livery, with a specially commissioned new preface, it has been revived for a new generation of readers in philosophy of science.

Preface to this edition Colin Howson; Part I. Methodology: 1. Introduction; 2. Aspects of the scientific method; 3. A taste of Bayesianism; Part II. Probability: 4. The primitive theory; 5. Subjectivism; 6. The rationalist interpretation; 7. The logical interpretation; 8. The evidential state; 9. The empirical interpretation; Part III. Confirmation: 10. Explications; 11. The paradox; 12. A Bayesian pseudo-solution to the problem of induction; 13. Projection; Part IV. Induction: Section 1. The Nature of Inductive Inference; Section 2. Conditions of Rationality: 14. Demonstrable reliability; 15. The demonstrable reliability of c+; 16. Immodesty; 17. Audacity; Section 3. The Justification of Induction: 18. The impossibility of a noncircular rationale; 19. Inductive demonstration of reliability; 20. Semantic justification; Part V. Prediction: 21. Surprise; 22. Severe tests; 23. Ad hoc hypothesis; 24. Prediction versus accommodation; Part VI. Evidence: 25. The evidential value of varied data; 26. The value of further data; Part VII. Realism: 27. Popper; 28. Realism versus instrumentalism; 29. Putnam; 30. Glymour; 31. Conclusion; Bibliography; Index.