Models of Discovery: And Other Topics in the Methods of Science (Boston Studies in the Philosophy of Science)

We respect Herbert A. Simon as an established leader of empirical and logical analysis in the human sciences while we happily think of him as also the loner; of course he works with many colleagues but none can match him. He has been writing fruitfully and steadily for four decades in many fields, among them psychology, logic, decision theory, economics, computer science, management, production engineering, information and control theory, operations research, confirmation theory, and we must have omitted several. With all of them, he is at once the technical scientist and the philosophical critic and analyst. When writing of decisions and actions, he is at the interface of philosophy of science, decision theory, philosophy of the specific social sciences, and inventory theory (itself, for him, at the interface of economic theory, production engineering and information theory). When writing on causality, he is at the interface of methodology, metaphysics, logic and philosophy of physics, systems theory, and so on. Not that the interdisciplinary is his orthodoxy; we are delighted that he has chosen to include in this book both his early and little-appreciated treatment of straightforward philosophy of physics - the axioms of Newtonian mechanics, and also his fine papers on pure confirmation theory. Front Matter....Pages i-xix Front Matter....Pages 1-6 Symmetric Tests of the Hypothesis that the Mean of One Normal Population Exceeds That of Another....Pages 7-13 Statistical Tests as a Basis for ‘Yes—No’ Choices....Pages 14-19 Prediction and Hindsight as Confirmatory Evidence....Pages 20-24 On Judging the Plausibility of Theories....Pages 25-45 Front Matter....Pages 47-52 Causal Ordering and Identifiability....Pages 53-80 On the Definition of the Causal Relation....Pages 81-92 Spurious Correlation: A Causal Interpretation....Pages 93-106 Cause and Counterfactual....Pages 107-131 Front Matter....Pages 133-136 The Logic of Rational Decision....Pages 137-153 The Logic of Heuristic Decision Making....Pages 154-175 Front Matter....Pages 177-181 Theory of Automata: Discussion....Pages 182-182 Aggregation of Variables in Dynamic Systems....Pages 183-213 The Theory of Problem Solving....Pages 214-244 The Organization of Complex Systems....Pages 245-261 Front Matter....Pages 263-267 Thinking by Computers....Pages 268-285 Scientific Discovery and the Psychology of Problem Solving....Pages 286-303 The Structure of Ill-Structured Problems....Pages 304-325 Does Scientific Discovery Have a Logic?....Pages 326-337 Discussion: The Meno Paradox....Pages 338-341 Front Matter....Pages 343-348 The Axioms of Newtonian Mechanics....Pages 349-369 Front Matter....Pages 343-348 Discussion: The Axiomatization of Classical Mechanics....Pages 370-375 Definable Terms and Primitives in Axiom Systems....Pages 376-386 A Note on Almost-Everywhere Definability....Pages 387-387 The Axiomatization of Physical Theories....Pages 388-402 Ramsey Eliminability and the Testability of Scientific Theories....Pages 403-421 Identifiability and the Status of Theoretical Terms....Pages 422-440 Back Matter....Pages 441-464 In these essays, the author attacks the central problems in the philosophy of science from the viewpoint of a working scientist who relates the philosophical questions to methodological issues that have arisen in his own research. The essays deal with six main topics: the validation of hypotheses, the logic of causal ordering and its relation to the definability and identifiability of theoretical terms, the logic of imperatives and decisions, the structure of complex systems, the theory of scientific discovery, and the axiomization of scientific theories. He treats these topics in a unified way by employing the analytic tools of model theory and by putting the discovery of theories, rather than their testing, at the center of the stage. Questions surrounding the axiomization of theories are examined in the light of real examples (Newtonian mechanics and Ohm's law) rather than simplified, hypothetical ones. Formalisms are not avoided, but they are always used as a means and not as ends in themselves.-- From publisher's description