Basil Rathbone and Nigel Bruce in Sherlock Holmes and the Secret Weapon, c. 1943.
I first discovered Sherlock Holmes as a boy of ten, on a dark and stormy night in November, huddled under blankets with a paperback of The Hound of the Baskervilles. Two days later I rode my bicycle to the public library and signed out a worn one-volume edition of the master detective’s complete adventures. Tremulous with excitement, I immediately sat down at a secluded desk near an old-fashioned radiator and turned to A Study in Scarlet, described on its title page as “a reprint from the reminiscences of John H. Watson, MD, late of the Army Medical Department.” Ever since, I have never stopped reading and rereading those reminiscences.
Even now, just murmuring the titles of Sir Arthur Conan Doyle’s Sherlock Holmes novels delivers a little jolt of anticipatory excitement: A Study in Scarlet (1887), The Sign of Four (1890), The Hound of the Baskervilles (1902), The Valley of Fear (1915). In them Conan Doyle—borrowing elements from nineteenth-century “sensation” fiction—incorporates ancient curses, dark doings at the outposts of empire, criminal masterminds, blackmail, and revenge. The novels are absolutely essential to a full understanding of Holmes and Watson.
Compared with the tightly composed short stories, however, the book-length cases can be structurally awkward. Three contain protracted “flashbacks” that are effectively separate stories in themselves. Conan Doyle resorts to these historical narratives—which can be quite gripping on their own—to explain the reasons behind the murders Holmes has just solved. The Hound of the Baskervilles doesn’t feature one of these extensive tailpieces, but its central chapters focus on Watson without Holmes, which is a little like Hamlet without the prince.
There are several possible reasons why Conan Doyle adopted such a cumbersome framework. During the 1880s the French detective-story writer Émile Gaboriau enjoyed immense popularity in England, and he employed just this odd structure for his books. (His Monsieur Lecoq, for instance, includes such a long explanatory flashback that one modern edition simply drops the second half of the novel entirely.) An additional reason may be, quite simply, Conan Doyle’s own passion for history. Even though Sherlock Holmes’ exploits made Conan Doyle rich and famous, he much preferred to write historical fiction. He relished re-creating an era in detail and would happily immerse himself in background research for months, piling up facts and data with which to give his narratives authenticity.
So even though A Study in Scarlet turned out to be the young writer’s first published novel, it was soon succeeded by far more ambitious—and much bulkier—works, such as Micah Clarke (1889), set in the seventeenth century, and The White Company (1891), set in the fourteenth. The latter, a chivalric romance, was Conan Doyle’s favorite among his novels, especially when read in tandem with its prequel Sir Nigel (1906). Many pages linger over the minutiae of period manners, costume, and weaponry. As Hesketh Pearson observed in a biography of Conan Doyle: “To the end of his life it never occurred to him that the accumulation of detail, however accurate or picturesque, does not vivify an age but nullifies it.”
That lifelong love for history, coupled to an uneasy fascination with American life, re-emerges in A Study in Scarlet and The Valley of Fear, where the second halves are, respectively, a western set in Utah with a titillating sexual theme (Mormon polygamy) and a laconic, hard-boiled account of violence and class warfare in the mining community of “Vermissa Valley, USA.” Some readers may recognize Conan Doyle’s principal literary sources for these dime-novel episodes: “The Story of the Destroying Angel,” part of New Arabian Nights: The Dynamiter (1885) by Robert Louis Stevenson and his wife Fanny Stevenson, and The Molly Maguires and the Detectives (1887) by Allan Pinkerton, who based his book on his own Pinkerton Detective Agency’s infiltration of a secret society of miners.
And yet, while Sherlock Holmes is the most famous fictional character of modern times, and the most filmed as well, he truly would be nothing without Watson, his chronicler, his Boswell, his friend. Christopher Morley—founder of that quintessential Sherlockian society, the Baker Street Irregulars—once called their collected adventures “a textbook of friendship.” The stories record how Holmes saved Watson from going to seed and how the good doctor gradually humanized a great thinking machine. As Vincent Starrett said of them in his poem “221B”: “Here dwell together still two men of note/Who never lived and so can never die.” Even now people still write to Baker Street pleading for their advice and help.
Conan Doyle once wrote: “So elementary a form of fiction as the detective story hardly deserves the dignity of a preface.” But are the Holmes adventures so “elementary”? I’m writing this during the centennial year of The Valley of Fear. The Sherlock Holmes “canon” has received the kind of microscopic attention usually bestowed on the Bible, Shakespeare, and James Joyce. Every word has been studied, every angle explored, multiple interpretations expounded.
I myself have contributed to this plenty. In 2002 I was invested into the Baker Street Irregulars and given the canonical name “Langdale Pike,” after a journalist—a gossip columnist, really—who appears in arguably the worst Sherlock Holmes story, “The Three Gables.” I would certainly argue that these accounts of Holmes’ and Watson’s adventures are, as much as any Keatsian urn, a joy forever. Nevertheless, it was a very near thing that they were ever recorded at all.
Before he created the most illustrious residents of Baker Street—whom he nearly called J. Sherrinford Holmes and Ormond Sacker—Arthur Conan Doyle had already written a novel that was lost in the mail, and contributed excellent short fiction to various magazines. “The Captain of the Pole-Star” (1883), set in the Arctic, is one of the most haunting Victorian tales of the supernatural. But the young writer could hardly think of quitting his day job as a doctor in Southsea. A Study in Scarlet was turned down by one publisher after another, until it was finally accepted by Ward, Lock, and Co., who offered to buy the British copyright for a derisory twenty-five pounds. Out of desperation, Conan Doyle took the paltry sum, then still had to wait a year before his short novel came out in Beeton’s Christmas Annual for 1887. Today, that annual may be the most valuable magazine in the world. Only thirty-three copies are known to exist and many are tattered or incomplete. If a truly fine copy were to appear on the market today, it might bring a quarter of a million dollars or more.
The 1887 Beeton’s containing A Study in Scarlet sold moderately well, and the novel was later republished as a book, with rather crude illustrations by Conan Doyle’s artist father. And that was all. There was no great hoopla, no recognition of a new star in the nascent detective story firmament.
Yet from the first page, Conan Doyle’s storytelling mastery—the genial narrative voice, the fast-moving action—sweeps the reader along. In short order we learn that John H. Watson has been an army doctor, was grievously wounded at the battle of Maiwand in the Second Afghan War, and now, broken in health, has wearily returned to England. One day he encounters an old acquaintance who tells him about a chap looking for someone to share digs with in Baker Street.
Watson and Holmes meet at St. Bart’s hospital, where Holmes’ first recorded words are “I’ve found it!”—that is, the English for “Eureka,” exclaimed by Archimedes when he grasped the displacement of liquids as he sat in his bath. A Study in Scarlet also provides the first appearance of the original Baker Street Irregulars, the London street urchins who can go anywhere and overhear anyone, and who serve the detective as a city-wide surveillance system. Most important of all, Watson discovers his own new vocation: Near the story’s end, he tells Holmes, “You should publish an account of the case,” and then adds, “If you won’t, I will for you.” The detective shrugs. “You may do what you like, Doctor.”
What, though, led Conan Doyle to bring back Holmes, given that he probably never intended to write about him again after A Study in Scarlet? It is a remarkable story. In 1889 J.M. Stoddart of Lippincott’s Magazine, published out of Philadelphia, traveled to London in search of new material for his pages. He invited two rising young authors to dinner at the Langham Hotel—which now displays a plaque commemorating the evening—and signed up both for short novels. So Oscar Wilde wrote The Picture of Dorian Gray (1891) and Arthur Conan Doyle produced The Sign of the Four (in later British editions, The Sign of Four, 1890). The latter book’s portrait of the fussy and hypochondriacal Thaddeus Sholto is, in part, an affectionate caricature of Wilde.
The London of The Sign of Four is a kind of Arabian Nights realm of wonders, Baghdad on the Thames. Graham Greene once confessed that of all the Holmes adventures only “that dark night in Pondicherry Lodge, Norwood, has never faded from my memory.” The novel, however, was written quickly: Conan Doyle signed the contract for The Sign of Four on August 30, 1889 and delivered the manuscript a month later; A Study in Scarlet was written almost as fast. Short stories were also written quickly, and Conan Doyle would seldom require more than a couple of days. Once a plot had been worked out, the actual writing could be done in one draft, with perhaps a few light corrections afterward. Even so, the texture of Conan Doyle’s writing was thick with many covert allusions. In The Sign of Four, he uses the phrase “flitted from the gloom into the light, and so back into the gloom once more,” drawing the image from the Venerable Bede, who in a famous passage of his Ecclesiastical History of England (circa 731) compares the meaningless life of a pagan to a sparrow that flies from a stormy winter’s night into a warm mead hall and then, all too quickly, passes back into the surrounding darkness.
While The Sign of Four is intimately connected to the turmoil and bloodshed of the Indian Mutiny of 1857, its main action takes place in 1888. The 1880s were themselves a violent decade in English history, one that included a series of Fenian dynamite bombings, an attempt to assassinate Queen Victoria (by a man who hated the number four), and, most infamously, the 1888 Whitechapel “Jack the Ripper” murders. At one point Holmes analyzes a letter just as the police analyzed those of the serial killer. This was also the period when scientists began to chart and tabulate human behavior, when Alphonse Bertillon measured skulls (anthropometry) for indications of latent criminality, and Cesare Lombroso argued that criminals were an atavistic throwback to the primitive savage. Such classification and typology are regularly alluded to throughout the story. Watson clearly believes in these phrenological pseudo-sciences, while Holmes will have none of it.
Not least, Holmes’ second published case proffers the first iteration of what would become his best-known, and most repeated, aphorism: “How often have I said to you that when you have eliminated the impossible, whatever remains, however improbable, must be the truth.”
You might have thought that the second appearance of Sherlock Holmes and Dr. Watson would have been greeted with huzzahs from a grateful public. No such luck. The Sign of Four proved only modestly successful. Conan Doyle was still pinning his publishing hopes on The White Company and other “serious” books. But having few patients for his new London medical practice and needing money, in 1891 he decided to submit short stories to a recently established magazine called The Strand. The first was titled “A Scandal in Bohemia” and opened with the tantalizing sentence: “To Sherlock Holmes she is always the woman.” In the resourceful and daring Irene Adler, the sleuth of Baker Street truly meets his match: she turns out to be far more than just “the daintiest thing under a bonnet.” In the following months, “The Red-Headed League,” “The Five Orange Pips,” and “The Speckled Band” would finally make Conan Doyle famous and Sherlock Holmes immortal.
As the grateful editor of The Strand proclaimed, he had found “the greatest natural-born storyteller of the age.” But fairly soon Conan Doyle began to tire of these trivial entertainments; they kept him from “better things.” Only the writer’s formidable mother persuaded him to continue writing about Holmes and Watson for a while longer. In “The Greek Interpreter” he even doubled the narrative’s star power by introducing Sherlock’s lazy, corpulent, and smarter older brother Mycroft, whose specialty is “omniscience” and who sometimes “is the British government.” And then, finally, inevitably, he introduced Holmes’ most dangerous and implacable foe, the criminal genius Professor James Moriarty. When—despite the entreaties of friends, family, and editors—Conan Doyle irrevocably determined to kill off Sherlock Holmes in “The Final Problem” (1893), he arranged for the detective to confront the Napoleon of Crime on the treacherous paths high above Switzerland’s Reichenbach Falls. The two enemies grappled and, to all appearances, tumbled into the gulf below. Having elected to sacrifice his life to preserve the world from evil, Holmes—once merely an inhuman calculating machine and bohemian aesthete—had now become, in Watson’s words, “the best and the wisest man whom I have ever known.”
Thereafter, among other projects, Conan Doyle took to chronicling the glorious and comic exploits of the Napoleonic soldier Etienne Gerard (in their way, they are as good as the Holmes stories), while The Strand ran a series of mysteries solved by Arthur Morrison’s Martin Hewitt, the first of the so-called rivals of Sherlock Holmes. But the world wanted the one, the only. While on a golfing holiday Conan Doyle learned from a younger friend named Bertram Fletcher Robinson about the supernatural folklore of Dartmoor, including occasional sightings of a spectral dog of death. Together the two writers began to sketch out “a real creeper.” Before long, Conan Doyle concluded that this was a case for Sherlock Holmes.
The Hound of the Baskervilles (1902) deals with ancient terrors in a desolate landscape of moor and bog, where—to borrow phrases from that devoted Sherlockian T.S. Eliot—the Grimpen Mire affords no secure foothold and the visitor is menaced by monsters and deadly enchantment. It is a tale, above all, about an aristocratic family haunted by a monstrous beast that brings terror and violent death. Conan Doyle dedicated the book to Robinson, who claimed he’d written parts of it and who sometimes called himself its joint author. No one will ever know for sure the extent of his probably minimal involvement. Still, only one thing really mattered: Holmes was back! Unfortunately, his creator hadn’t actually resurrected the great detective. Instead Conan Doyle subtitled his book “Another Adventure of Sherlock Holmes” and set this chilling case sometime before the fatal encounter with Moriarty. In that story, Watson had memorialized his friend in words originally applied to Socrates in Plato’s Phaedo. Here, as The Hound gave readers back their familiar hero, one could again hear echoes of Platonic dialogue in some of the detective’s exchanges with his old friend:
“Is it then stretching our inference too far to say the presentation was on the occasion of the change?”
“It certainly seems probable.”
The Hound of the Baskervilles could be categorized as a tragic regional novel, like Emily Brontë’s Wuthering Heights or Thomas Hardy’s Return of the Native: the desolate landscape—“the barren waste, the chilling wind, and the darkling sky”—dominates the merely human characters. By contrast, The Valley of Fear (1915) presents itself as a modern novel about business, not just because its second half deals with clashes between management and workers, but also because it focuses throughout on contracts, pledges and oaths, on marital promises and other binding commitments.
This novel, like the others, is set before the events chronicled in “The Final Problem.” Of course, by now contemporary readers had known for over a decade that Holmes was never really dead. Yielding to popular demand and substantial financial incentives, Conan Doyle had finally explained, in “The Adventure of the Empty House” (1903), how the great detective actually survived his meeting with Moriarty and what he had earlier called, with moving understatement, “the final discussion of those questions that lie between us.” After his “great hiatus” (spent partly in Tibet under the name Sigerson), Holmes eventually returned to solve numerous subsequent mysteries before retiring to keep bees on the Sussex Downs.
The Valley of Fear is the most metafictional of the novels. It is a book filled with other books and carefully manufactured texts, where truth is stage-managed and nothing is natural. Police inspectors allude to Watson’s published works and coyly remark, “when the time comes we’ll all hope for a place in your book.” One important character refers to Watson as “the historian of this bunch” and hands him a substantial manuscript with the air of a writing student soliciting the attention of a published author.
In Conan Doyle’s work, violence—usually in conjunction with ill-gotten wealth—nearly always originates outside England’s green and peaceful land, typically in gangsterish America. Consequently Holmes, like the medieval knights his creator idolized, often works to preserve the civilized values of the British Empire against barbarity and disorder. That said, The Valley of Fear is only partly about the danger of lawlessness. Throughout the book, Conan Doyle sets up, and quickly skates over, a series of moral ambiguities. Under what circumstances, if any, can a man stand by and allow bloodshed to occur without acting? Does Sherlock Holmes himself cause the death of two men? Has Holmes’ deductive genius actually been subverted by Moriarty to assist the professor’s nefarious purposes?
Valley of Fear’s epilogue, set amid the seeming security of Baker Street, adds further twists to this most intricate, most Heisenbergian, and final Sherlock Holmes novel. After all, nearly every element in the narrative is susceptible to multiple interpretations, so much so that reality itself comes to seem labile, any truth undecidable. Even more than in the other novel-length adventures, Conan Doyle actually avoids neat and tidy closure. Mysteries remain. The modern world has arrived.
From SHERLOCK HOLMES: The Novels by Arthur Conan Doyle, to be published on November 24th by Penguin Classics, an imprint of Penguin Publishing Group, a division of Penguin Random House LLC. Introduction copyright © 2015 by Michael Dirda.