Inferring and Explaining

29 Dick was unfaithful—the leering, the excuse about being sick, the car outside the bar, and the missed phone calls. We had purported evi- dence about the copied take-home exam—the word-for-word identical submissions, Charlie’s chronic absences, and his failures as a writer. Finally, we had the evidence about the expand- ing universe—the red-shifed light fromdistant galaxies. In each of these cases, the suggested theory explains signifcant parts of our evi- dence. Charlie being a cheater doesn’t explain his bad writing, but it sure helps us understand how the two exams ended up being the same. Dick’s cheating (in a very diferent way) would explain why he was at the bar when he said he was sick. And an expanding universe explains the Doppler shif we observe in the light from galaxies. Tis suggests a generalization. Suppose we treat the theory being defended in an inductive argument as an explanation of the data (at least some of the data) contained in the evidence. We get the following very symmetrical picture of an argument: arguments Inference to the best explanation assumes this gen- eral picture of inductive arguments. Te rela- tionship of support or following from becomes one of good explanation. Evidence for a theory is strong, or good or sound, if and only if, the the- ory best explains the relevant data that is being ofered as evidence. Tis defnition of good evi- dence gives us a very useful device for testing the quality of purported evidence. In the next chapter, I intend to lay out a kind of practical test for answering questions about the strength of arguments, about the quality of evidence. Does Connie have good evidence that her boyfriend was smooching Mary Jane during his absence at the record hop? Is there a strong argument that Charlie copied the take-home exam? Do astronomers really know that the phys- ical universe is expanding? We can only begin to answer these questions when we are in abso- lute agreement about what the argument is in the frst place. A Couple of Arguments from Sherlock Holmes Let’s look at a couple of examples of evidence that lead to some conclusions for Sherlock Holmes. Here are the missing links of the very simple chain: 1. You had chalk between your lef fnger and thumb when you returned from the club last night. 2. You