What do we know as a starting point? There is relatively widespread consensus on a few basic facts, and I see no reason to disagree with the statement that Sherlock Holmes was born in 1854, while John Watson, slightly older, was born in 1852. Neither date is ever stated outright in the Canon; but Holmes is described without reservation as a “man of sixty” in 1914 in “His Last Bow” (LAST), and we know from A Study In Scarlet (STUD) that, in the very first words from Watson’s pen, “In the year 1878 I took my degree of Doctor of Medicine of the University of London,” which places the good doctor’s birth 26 years earlier given the typical duration of a British medical education of the day.
We know very few facts about either man’s childhood, family, or upbringing (although speculation is rampant). The next definitive information comes in a pair of “flashback” tales describing two of Holmes’s earliest cases, before he met Watson: “The Gloria Scott” (GLOR) and “The Musgrave Ritual” (MUSG), which appeared in print back-to-back in consecutive months in 1893. (It’s interesting to note, in passing, that departures from the usual narrative structure of the Writings seem to come in pairs; the only two cases told in the third person also appeared consecutively, in 1917 and 1921 respectively, as did the only two written by Holmes himself, both in 1926.) Thus, it is time to begin delving into the details of the cases.
In GLOR, Holmes relates to Watson “the very first [case] in which I was ever engaged,” which offered an epiphany for Holmes about how he might use his deductive powers. It involved young Trevor, “the only friend I made during the two years I was at college… the only man I knew.” In MUSG, Holmes tells his friend of the third case he handled in the difficult and meager days after he “first came up to London,” when “cases came in my way, principally through the introduction of old fellow-students, for during my last years at the university there was a good deal of talk there about myself and my methods.”
I feel compelled to agree with William Baring-Gould’s analysis here, in which he argued that Holmes was plainly describing two different college experiences. Without repeating all the details, suffice it to say that Baring-Gould looked over the academic calendars of both Oxford and Cambridge (the only two schools really worthy of consideration) for the years in question, compared them with the legal duck-hunting season at the time and certain other relevant details of the case, and concluded that GLOR occurred over the summer of 1874, after Holmes had spent two years at Oxford. Inspired by his role in those events, which in his own words in MUSG “turned my attention in the direction of the profession which has become my life’s work,” Holmes then presumably left Oxford and enrolled at Cambridge, where we may conclude he spent three years pursuing both his “methods” and his social contacts more assiduously, leaving in 1877 (presumably with a degree) and thereafter setting up his fledgling practice in London. MUSG involved Holmes’s former fellow student Reginald Musgrave, of whom “for four years I had seen nothing”; Musgrave could have graduated no sooner than 1875 if he had known Holmes at school, placing this story no sooner than 1879. The month of the story is “the sixth from the first,” and near to the solstice, per details of the case, thus July.
From there we proceed to the first case in which Dr. Watson was involved and the first actually published, STUD, in which we also learn how he met Holmes, introduced by their mutual acquaintance Stamford. Watson relates how he was injured in Afghanistan while serving as an Army surgeon at the Battle of Maiwand (which occurred in July 1880), and spent several months recuperating and traveling back to England, arriving on the troop ship Orontes (which other Sherlockian scholars have discovered docked at Portsmouth on November 25 of that year) and making his way to London “with my health irretrievably ruined, but with permission from a paternal government to spend the next nine months in attempting to improve it.” With only a modest injury pension to live on, he soon sought a way to share expenses, and was introduced to Holmes, who was under similar circumstances (his consulting detective practice being at this stage still far from lucrative). The case of the title (which also introduced readers to Inspector Lestrade of Scotland Yard) commenced “upon the 4th of March,” by which time Holmes and Watson were both well ensconced in their shared lodgings at No. 221B, Baker Street. So, our progress to this point:
GLOR • Jul-Sep, 1874
MUSG • Jul, 1879
STUD • Mar, 1881
There’s little controversy thus far (except for the fact that GLOR describes a prison transport ship wrecked no less than thirty years earlier as embarking in “1855,” but almost all commentators agree that this must be a misprint for 1845, as an 1885 setting for Holmes’s college years would obviously be impermissibly late). The real problems start to arise with the next case, The Sign of the Four (SIGN), and the introduction of Miss Mary Morstan.
In SIGN Holmes and Watson solve the enigmatic case of the Sholtos and the Agra Treasure for Miss Morstan, and Watson, smitten, proposes marriage. In the very next case published, “A Scandal in Bohemia” (SCAN) (which also introduces the infamous Irene Adler, who out-thought Holmes and was thereafter known to him as The Woman), Watson begins by reporting that “I had seen little of Holmes lately. My marriage had drifted us away from each other… [and] I had now returned to civil practice.” He clearly intends the reader to understand that the marriage in question is to Mary Morstan—and if any doubt remained, in a later tale, “The Stockbroker’s Clerk” (STOC), Holmes explicitly inquires of him, “I trust that Mrs. Watson has entirely recovered from all the little excitements connected with our adventure of the Sign of Four.” So that seems simple enough… right?
Except. SIGN is understood by most chronologists to occur in the year 1888: Mary reports that her father “disappeared upon the third of December, 1878—nearly ten years ago,” and that “About six years ago–to be exact, upon the fourth of May, 1882–an advertisement appeared in the Times asking for the address of Miss Mary Morstan.” It commences on “a September evening” (there’s an odd reference to a letter postmarked July 7, but Conan Doyle himself remarked in a letter to the publisher that this was a manuscript error, doubtless having had it called to his attention by the irate author). It would thus seem that the earliest John Watson and Mary Morstan could possibly have wed was the end of 1888… yet in SCAN, set after the marriage, Watson tells us in no uncertain terms that the case begins “on the twentieth of March, 1888.”
How to resolve this conundrum? It only grows more complicated when one considers “The Five Orange Pips” (FIVE), a case published only four months after SCAN, in which Watson begins by describing “the latter days of September” in “the year ’87” (corroborated by being “two years and eight months” from a murder in “January of 1885”), yet goes on to say that “My wife was on a visit to her aunt’s, and for a few days I was a dweller once more in my old quarters at Baker Street.” Add in the fact that we already know from SIGN that Mary had “no relative in England,” and we can understand why some commentators are inclined to conclude that Watson had another wife before Mary, one who apparently met her end in some untimely and unmentioned way. This is, indeed, the conclusion to which Baring-Gould came… and as a result of it he wound up rearranging a significant number of other cases, irrespective of Watson’s dates, to accommodate that marriage.
I reject this approach. Such an unseemly multiplication of wives inevitably winds up causing more problems than it solves in the process of arranging the cases, and moreover it violates our watchword: trust Watson. Notwithstanding the minor discrepancies just described, the good doctor never gives us the slightest suggestion that he had any earlier marriage… and those discrepancies are slender evidence indeed on which to hang an entire wife. They can in fact be explained away without drawing any such rash conclusion.
First of all, SIGN should properly be dated in 1887, not ’88. Mary’s descriptions of events “nearly” and “about” a given number of years ago are not precise. What is precise is the fact that in May of ’82, she received by anonymous post ” ‘a very large and lustrous pearl. … Since then every year upon the same date there has always appeared a similar box, containing a similar pearl, without any clue as to the sender…’ She opened a flat box as she spoke and showed me six of the finest pearls that I had ever seen.” Six pearls, received in six consecutive years, leaves us with a date prior to May of ’88 (which would bring a seventh)… and we already know that it’s September, and hence the previous year. (Note: in reflecting back on the Jefferson Hope case, Watson remarks “I even embodied it in a small brochure, with the somewhat fantastic title of ‘A Study in Scarlet’ “; this is no obstacle, as he would surely have written it by this date, even if it had not yet seen print.)
Second, with regard to FIVE, we should guard against unwarranted assumptions. If Watson’s wife has no relatives in England, yet is visiting an aunt (some early printings said “mother,” but this was quite properly corrected in light of the fact that Mary’s mother was deceased), we may logically conclude that she has traveled outside of England to do so, most likely to the Continent. This is hardly an implausible thing for her to be doing in (later) September of ’87, since (A) she has significant news to share, both of her father’s fate and of her betrothal, and (B) she has just come into a considerable sum of money (although the Agra treasure was lost in the Thames in SIGN, we have every reason to believe that Mr. Thaddeus Sholto considered her entitled to at least some portion of his family’s ill-gotten fortune). If we slightly revise Watson’s problematic sentence to read that his “wife-to-be” was traveling, and that “for a few weeks [not days]” he was “a dweller still [rather than once more] in my old quarters at Baker Street,” and attribute the misleading wording to a minor editorial error, the entire problem disappears. This is surely a more elegant solution than inventing an entirely separate wife and rearranging cases around her.
The pieces are beginning to fall into place. At this point we can look at the entire set of pre-Hiatus cases published from 1891-’93, later collected in the volumes The Adventures of Sherlock Holmes and The Memoirs of Sherlock Holmes, and see that we have a structure into which we can fit them. In fact, several besides those already considered do include specific dates, and there is no compelling reason to doubt any of them. In chronological order, these are:
- “The Resident Patient” (RESI)—introduced (prior to an editorial alteration in the book version borrowed verbatim from a separate case) with the words “it must have been towards the end of the first year during which Holmes and I shared chambers in Baker Street. It was boisterous October weather…”
- “The Speckled Band” (SPEC)—”early in April in the year ’83”
- “The Reigate Squires” (REIG)—”in the spring of ’87… On referring to my notes I see that it was upon the fourteenth of April”
- “The Noble Bachelor” (NOBL)—”a few weeks before my own marriage, during the days when I was still sharing rooms with Holmes in Baker Street”; the client, Lord St. Simon, was “Born in 1846. He’s forty-one years of age”; and among the clues is a hotel bill dated “Oct. 4th,” just before “Wednesday last;” all in this “four-year-old drama” as of its writing (prior to publication in April ’92)
- “The Stockbroker’s Clerk” (STOC)—”Shortly after my marriage I had bought a connection in the Paddington district… For three months after taking over the practice I was kept very closely at work and saw little of my friend Sherlock Holmes… [until] one morning in June” (thus, three months after SCAN)
- “The Naval Treaty” (NAVA)—”The July which immediately succeeded my marriage”
- “The Crooked Man” (CROO)—”One summer night, a few months after my marriage”
- “The Man with the Twisted Lip” (TWIS)—”it was in June, ’89″
- “The Engineer’s Thumb” (ENGR)—”in the summer of ’89, not long after my marriage”
- “The Red-Headed League” (REDH)—in which the sign that prompts the case reads “THE RED-HEADED LEAGUE IS DISSOLVED. October 9, 1890” (one line does describe a newspaper as dated “April 27, 1890. Just two months ago,” but this seems anomalous and should probably read “six”)
- “The Final Problem” (FINA)—”the early spring of 1891,” in which “I saw [Holmes] walk into my consulting-room upon the evening of April 24th,” and in which Holmes’s apparent death is described as occurring “on the afternoon of the fourth” of May
Many of these cases mention earlier ones in passing, further cementing this order (without raising any contradictions). Thus, our schema so far:
— [Pre-Watson cases]
GLOR • Jul-Sep, 1874
MUSG • Jul, 1879
— [The partnership begins]
STUD • Mar, 1881 (Watson meets Holmes; first appearance of Lestrade)
RESI • Oct, 1881
SPEC • Apr, 1883
REIG • Apr, 1887
SIGN • Sep, 1887 (Watson meets Mary Morstan)
FIVE • late Sep, 1887
NOBL • Oct, 1887
— [Watson marries]
SCAN • Mar, 1888 (Holmes encounters Irene Adler)
STOC • Jun, 1888
NAVA • Jul, 1888
CROO • Summer, 1888
TWIS • Jun, 1889
ENGR • Jul, 1889
REDH • Oct, 1890
FINA • Apr-May, 1891 (Holmes confronts Prof. Moriarty at Reichenbach Falls, Switzerland; believed dead)
— [The Great Hiatus begins]
We are left, however, with several cases of less-definite date that also fit in amongst those in this list. This is where the project really begins to get interesting. For these… the next installment!