Pierre Bayard reads The Hound of the Baskervilles, and comes to the same conclusion many other readers have: In this particular case, Sherlock Holmes was wrong.
This is a work of slightly tongue in cheek, very French literary criticism. As such, it's not for everyone. This isn't a fault in the book, but simply a matter of taste.
Bayard begins with an explanation of what he calls "detective criticism," and a recap of the events of The Hound of the Baskervilles. This is the preliminary to analyzing the wild improbabilities in Holmes' analysis, as well as an examination of the difficulties of Arthur Conan Doyle's most famous, popular, and, for his creator, least loved hero. There is extended and lively discussion of to what degree fictional characters have autonomy from their creators; if that strikes you as an annoyingly foolish idea, this isn't the book for you. If you've suspected the same yourself, on the other hand, it's great fun to engage with Bayard on the subject.
Some oddities result from the fact that Bayard not only writes in French, but read Hound in French translation, not in English. There are a few points where he infers something of Conan Doyle's intent from the use of a metaphor which was, in fact, introduced by the French translator and does not occur in Conan Doyle's original text. Also, and I hesitate to mention this, Bayard overlooks what I think is the most glaring and obvious error, an error which is set up in Holmes and Watson's first conversation with Dr. Mortimer.
Despite these and a few other minor flaws, this is an interesting and challenging book, if you enjoy this kind of literary analysis. Not everyone does, but I do, and I thoroughly enjoyed this example of it.j
I borrowed this book from the library.