|
Abstract: Several states have enacted statutes that mandate an inference of paternity when the “probability of paternity” is high enough. This article discusses three objections to these statutes: their choice of the statistic that triggers the presumption, their failure to specify the nature and effect of the presumption, and the need for the presumption in the first place. Because the “probability of paternity” is not necessarily the probability the tested man is the biological father, the article examines the distribution of this statistic in simulated groups of falsely and correctly accused men. It demonstrates that the “probability of paternity” usually tends to identify the biological fathers and therefore leads to correct inferences at approximately the desired rates. Nevertheless, the article concludes that the statutes are not needed since, if the jury is given the probability that the defendant is the father, it hardly needs a presumption to tell it what to conclude. An expert’s description of the genetic findings and an appropriate characterization of their probative value should provide the jury with all the information it needs to assess the scientific evidence.
|