Almost all those who find it difficult to credit Sir Arthur Conan Doyle's thesis regarding the spirit world express surprise that a man of "such high intellectual attainments" should draw conclusions that seem to them unwarranted, and from premises they look upon as insecure.

Whatever reason there may be for the criticism of Dr. Doyle's argument, however, there seems no occasion for surprise that his logic is not always perfect. The author of Sherlock Holmes may or may not be right about the spirit world, but it is hard to understand why he should be generally rated as a master of scientific deduction because he invented a fictional character purporting to be such.

As the author of the "Father Brown" detective tales recently pointed out with a good deal of cogency, the methods of deduction employed by the famous Holmes are more effective within the covers of a work of fiction than in actual life. The real detective, however shrewd a ratiocinator, rarely meets with a situation where the source of a mystery can be reached by a straight chain of syllogisms. In detecting the perpetrator of a crime, the real detective usually examines all the suspects and finds the guilty party by a winnowing process of elimination, or by the even more haphazard plan of choosing the motive that seems to fit best with the circumstances and then examining suspects with reference to it. When he is successful, which is perhaps as often as not, it is generally by reason of the number of sleuths employed and the volume of evidence collected. Induction, in other words, has as much to do with it as deduction. Detectives fall in cases like that of High-tower, or of William Taylor, not because they are incapable of reasoning processes similar to those of Holmes, but because, as Chesterton says, clues never point one way, after the convenient manner of the clues Dr. Doyle plotted for his hero.

In actual life Dr. Doyle, has not succeeded in unravelling any of the great mysteries that baffle the police. Yet so powerful is the illusion exerted by his fiction that the average man persists in believing him a master of inference. By a similar fallacy, or perhaps for the fun of the thing, he has been consulted by the London police regarding a number of cases, but never with very noteworthy results. Even in the realm of fiction he has not shone analytically as Poe did when he prophesied the conclusion of Dickens's "Barnaby Rudge" after reading the first chapters in serial. Dr. Doyle's opinion, for example, regarding the conclusion of "The Mystery of Edwin Drood," the novel which the death of Charles Dickens interrupted, and his solution of the fictional mystery are rather less tenable than those of many other commentators hardly known to fame.

To the mysteries of the life beyond death, science will doubtless reveal many clues in time, but one must be distinctly credulous to accept Sir Conan Doyle's criteria in the meantime because he invented Sherlock Holmes. San Francisco Journal