A few years ago, a poll found that 58 percent of British teenagers thought Sherlock Holmes was a real person (meanwhile, 47 percent thought that Richard the Lionheart was not). That may be just a sad statement on the education system, but that doesn’t mean those kids are alone. There’s actually a whole group of people who enjoy the theory that Sherlock Holmes – or at least sidekick John Watson -- was real.
The explanation is simple: Dr. Watson chronicled the work of the London detective Sherlock Holmes and their relationship. Arthur Conan Doyle? He was Watson’s literary agent and helped bring the stories to The Strand magazine and other outlets.
Of course, Doyle wasn’t just an agent. He claims to have based the Holmes character off of his former teacher, Dr. Joseph Bell, who was said to have similarly impressive deductive powers. The Holmes pieces, starting with 1887’s “A Study in Scarlet,” quickly became his most famous works, overshadowing anything else he wrote (a fact that frustrated Doyle and led to his decision to "kill" Holmes in "The Final Problem"). Eventually, Doyle wrote 56 short stories and four novels featuring Holmes across 40 years.
However, Doyle’s presence creates another problem for the Holmesians who believe in the canon. The author was never one to stick closely to his earlier works and said on many occasions that he wouldn’t let the canon stand in the way of a good story. That means dates don't line up and scholars have had trouble putting the cases in chronological order. Likewise, characters meet and re-meet, physical descriptions change and even personality traits disappear or shift as needed.
For example, in several stories Holmes refuses to take a reward, even claiming that “my profession is my reward.” But in other cases, such as “The Adventure of the Beryl Coronet,” he takes as much 4,000 pounds. This seeming inconsistency has been brushed off with a simple explanation: Holmes only accepts money from wealthy clients when he needs it.
The “great game” of studying the Holmes canon began with Ronald Knox, who sought to apply Holmes’ own methods on the canon in his essay “Studies in the Literature of Sherlock Holmes.” In it, he tackles the exact order and date of the canon mysteries, Holmes' ever-shifting routine of investigation and even the flaws in Watson's personal history. Knox even offers up an explanation of the inconsistencies in Watson's work:
"I believed that all the stories were written by Watson, but whereas the genuine cycle actually happened, the spurious adventures are the lucubrations of his own unaided invention. Surely we may reconstruct the facts thus."
Interestingly, the idea of believing in canon and acknowledging the author can be applied outside of the Holmes literature. For example, TVTropes.org explains how Doylists and Watsonians exist in TV fandom: a Doylist would understand that an actor had to be recast, while a Watsonian would infer that the character in question had gotten plastic surgery to change his or her appearance.