Expanding the scope of the "detective criticism" he began in 2000's "Who Killed Roger Ackroyd? The Mystery Behind the Agatha Christie Mystery," French literature professor and psychoanalyst Pierre Bayard turns his attention to another canonical text of the genre, Arthur Conan Doyle's "The Hound of the Baskervilles."
In "Sherlock Holmes Was Wrong" (Bloomsbury: 196 pp., $20), Bayard analyzes "the way the facts are presented, accepting no testimony without reservation and systematically calling into question everything" in and outside of the text. To take on both Conan Doyle -- whose Sherlock Holmes adventures are almost sacred texts to many mystery fans -- and arguably his sleuth's most famous case seems doomed for failure. Undaunted, Bayard argues that Holmes was far from infallible and made mistakes in "Hound" and several other cases that "cast doubt on the official verdict" and allowed the real perpetrator to escape justice.
With wit and careful analysis, Bayard makes a convincing case for not only Dr. Watson's unreliability as narrator and eyewitness to the events in the novel but also the trickier assertion that Conan Doyle, who had actually killed off Holmes in an earlier story and was forced by popular demand to bring him back, became distracted by the fantastical setting and dog he had created and ignored the real killer operating in the margins.