2021/01/01 by J. Dmitri Gallow · 3 citations
Arts and Humanities · Computer Science · #Philosophy and History of Science #Bayesian Modeling and Causal Inference #Logic, Reasoning, and Knowledge
paper · doi:10.1215/00318108-8699682
This article provides a theory of causation in the causal modeling framework. In contrast to most of its predecessors, this theory is model-invariant in the following sense: if the theory says that C caused (didn’t cause) E in a causal model, M, then it will continue to say that that C caused (didn’t cause) E once one has removed an inessential variable from M. The article suggests that, if this theory is true, then one should understand a cause as something which transmits deviant or noninertial behavior to its effect.