No abstract

This book is about preferences, principally as they figure in economics. It also explores their uses in everyday language and action, how they are understood in psychology and how they figure in philosophical reflection on action and morality. The book clarifies and for the most part defends the way in which economists invoke preferences to explain, predict and assess behavior and outcomes. Hausman argues, however, that the predictions and explanations economists offer rely on theories of preference formation that are in need of further development, and he criticizes attempts to define welfare in terms of preferences and to define preferences in terms of choices or self-interest. The analysis clarifies the relations between rational choice theory and philosophical accounts of human action. The book also assembles the materials out of which models of preference formation and modification can be constructed, and it comments on how reason and emotion shape preferences.

This essay explains what the Causal Markov Condition says and defends the condition from the many criticisms that have been launched against it. Although we are skeptical about some of the applications of the Causal Markov Condition, we argue that it is implicit in the view that causes can be used to manipulate their effects and that it cannot be surrendered without surrendering this view of causation. 1 What is the Causal Markov Condition? 2 Screening-off and the Causal Markov Condition 3 Factorizability 4 Manipulability and causation 5 Level invariance and modularity in linear equation systems 6 Modularity, linearity and mechanisms 7 Modularity, mechanisms and an argument for the Causal Markov Condition 8 Strong independence and the Causal Markov Condition 9 Cartwright's objection 10 Indeterminism and the Causal Markov Condition 11 Conclusion Whether or not one holds that causation is a deterministic relation, there appear to be connections between causation and probabilities. Causes are correlated with their effects. Effects of a common cause are unconditionally correlated with one another, but they are independent conditional on their common cause. Causal intermediaries screen off their effects from their causes. Events that are not related as cause and effect or as effects of common causes are uncorrelated. These claims are rough and, as just formulated, indefensible. But they point to an important connection between causation and probabilities. This paper is concerned with a promising formulation of this connection, which Peter Spirtes, Clark Glymour and Richard Scheines (hereafter SGS) call 'the

No abstract

This book, by one of the pre-eminent philosophers of science writing today, offers the most comprehensive account available of causal asymmetries. Causation is asymmetrical in many different ways. Causes precede effects; explanations cite causes not effects. Agents use causes to manipulate their effects; they don't use effects to manipulate their causes. Effects of a common cause are correlated; causes of a common effect are not. This book explains why a relationship that is asymmetrical in one of these regards is asymmetrical in the others. Hausman discovers surprising hidden connections between theories of causation and traces them all to an asymmetry of independence. This is a major book for philosophers of science that will also prove insightful to economists and statisticians.