Inferring and Explaining

140 for themselves and that rational people should be able to agree on what the facts tell us. I wish that were true, but I doubt that it is. It’s not just in the law or political theory that the way a story is told is relevant to whether the story convinces its audience. Tis is ofen the case in science, scholarly disagreements, and family arguments over Tanksgiving dinner. InferrIng and exPlaInIng Inference to the Best Narrative As we sawwithMary Ann andWanda, and with Derrick Bell andGeneva, we ofen ofer evidence for theories or positions that don’t explain any of the data provided in the evidence. We ofer the prediction that global warming will con- tinue to increase. We defend it with evidence from basic physics and chemistry, the histori- cal record, and the testimony of respected sci- entists. But the prediction doesn’t explain any of this. Derrick Bell ofered a constitutional analysis of Brown v. Board of Education , but this analysis doesn’t explain the history of race in this country nor what happened, and what did not happen, in the years since Brown was decided. Mary Ann and Wanda ofer a moral justifcation for murdering Earl, but this justi- fcation doesn’t explain the abuse, the divorce, or the assault. It would be nice to have a general tool for evaluating evidence in these kinds of arguments. So how might we do this? I have already suggested that we can capture much of the structure of the reasoning that seems to unite Connie’s diagnosis, and Mary Anne and Wan- da’s proposed course of action, by treating both their arguments as narratives (i.e., stories) that attempt to “make sense” of the relevant facts. We have seen that inference to the best explana- tion rests on a comparative procedure where we evaluate the success of competing explanatory stories. How are we to accomplish this? You will remember that at the beginning of chapter 5, we considered Gilbert Harman’s answer to this question: In making this inference one infers, from the fact that a certain hypothesis would explain the evi- dence, to the truth of that hypothesis. In general, there will be several hypotheses which might explain the evidence, so one must be able to reject all such alternative hypotheses before one is warranted in making the inference. Tus, one infers, from the premise that a given hypothesis would provide a “better” explanation for the evidence than would any other hypothesis, to the conclusion that the given hypothesis is true. 21 I believe we can extend Harman’s method to encompass not just straightforwardly explana- tory stories such as Connie’s but also norma- tive stories such as Mary Ann and Wanda’s and Geneva Crenshaw’s. We arrive at a kind of com- parative reasoning I am calling inference to the best narrative . Notice how nicely the concept of story replaces hypothesis, andmaking sense can be substituted for explanation. In making this inference, one infers from the fact that a certain narrative would make sense of the reasons to the truth of the story . In general, there will be several narratives that might make sense of the reasons , so one must be able to reject all such alternative narratives before one is warranted in making the inference. Thus