About | HeinOnline Law Journal Library | HeinOnline Law Journal Library | HeinOnline

LOG IN
LOG IN


13 Cardozo L. Rev. 1025 (1991 - 1992)
Influence Diagrams, Bayesian Imperialism, and the Collins Case: An Appeal to Reason

handle is hein.journals/cdozo13 and id is 1047 raw text is: INFLUENCE DIAGRAMS, BAYESIAN IMPERIALISM, AND THE COLLINS CASE: AN APPEAL TO REASON Ward Edwards* INTRODUCTION I have recently been rereading the legal literature stimulated by the famous Collins case.' My motives were innocent enough. I am deeply impressed by the usefulness of a new technology based on graphical structures called influence diagrams and Bayes nets; they make Bayesian reasoning about complex masses of conditionally non- independent evidence a fairly practical task. I have been learning how to use Bayes nets (influence diagrams that do not contain value or decision nodes) to structure cases, and assessing the resulting gains and pains. Collins is a good example because it is simple in structure, it can be made lean in size, and it has been pivotal in historical importance. The preceding paragraph contains several undefined technical terms; I define them later.2 Since a purpose of this paper is to intro- duce both Bayesian thinking and influence diagram technology to lawyers who may not be familiar with the language of either, its den- sity of new technical terms may be unpleasantly high. I try to define most of them in the body of the text at the first important use of each. In addition, the Appendix, containing a Glossary, may help. It con- tains definitions of forty-five words and phrases used in the paper. The definitions have a strongly Bayesian flavor; if you are comfortable with all of them, you understand Bayesian imperialism extremely well. * The Collins case is historically important mainly because it stim- ulated Michael Finkelstein and William Fairley to write a somewhat * Director of the Social Science Research Institute and Professor of Psychology and of Industrial and Systems Engineering, University of Southern California; Ph.D., Experimental Psychology, Harvard College, 1952. I am grateful for careful readings and insightful but sometimes unheeded comments to Bruce Abramson, Terry Anderson, Daniel Geiger, Gerald B. Greenwald, David E. Heckerman, Eric J. Horvitz, Eileen Kuhns, Jonathan J. Koehler, Tod Levitt, David Schum, Ross Shachter, and Peter Tillers. I People v. Collins, 68 Cal. 2d 319, 438 P.2d 33, 66 Cal. Rptr. 497 (1968). See discussion infra Part I. 2 See infra Part II(B); see also Glossary, infra, at 1070. 1025