Page:Burton Stevenson--The marathon mystery.djvu/333

Rh But now that the curtain’s down and we’re alone together with plenty of time to talk, I’d like to understand”

“And you shall—down to the minutest detail. Let’s see—this is the smoker, isn’t it? Well, suppose we light up—I can think more clearly when I’m smoking.”

“All right; fire away,” I said, as soon as the cigars were going.

“Well,” began Godfrey; “as I pointed out to you this morning, for good and sufficient reasons, I started out in this investigation with the assumption of Tremaine’s guilt.”

“Of course,” I observed, “you know it is the duty of every jury to start out with exactly the contrary assumption.”

“Certainly I know that; but a detective has to work with some definite end in view, or he never gets anywhere. In other words, a detective, after carefully studying the details of any crime, must form a theory concerning it, and must work along that theory. As soon as he discovers any fact that fails to fit with his theory, he must modify it or form another; and he must keep on doing this until he finds the theory which agrees with all the facts—not all but one or two, but with every one. A good many detectives fall into the mistake of being satisfied with the theory which fits most of the facts—a serious error, for the right theory must, of course, inevitably, fit them all. That’s the scientific method and the only safe one. When a detective hits upon a theory which fits all the known facts, he’s got as much right to assume it’s true as an