An excerpt from the chapter "Sherlock Holmes--The Series,"

in Martin Priestman's Detective Fiction and Literature (NY: St. Martin's, 1991), 75-85.

    Sir Arthur Conan Doyle's Sherlock Holmes is the supreme 'character' of nineteenth-century detective fiction. As such he has inspired two very different critical traditions: the biographical and the archetypal. The first consists of the lugubriously tongue-in-cheek following-up of minor inconsistencies about dates and names which accounts for the bulk of 'Holmesian' literature. The second, more critically respectable, avoids getting sucked into the labyrinth of particular cases by concentrating on the archetypal Holmes formula - drawing attention either to the famous 'methods' or to the romantic-scientific blend which accounts for Holmes's overall appeal as a 'personality'. (1)

    In the face of these two powerful traditions, I hope I am not simply being perverse in attempting first of all to introduce the
'canon' (a phrase I gratefully borrow from the Holmesians) without an initial close-up on the figure of the great detective himself. I shall eventually get round to him, but it seems to me that such a discussion will be more meaningful after some attempt to establish the social concerns and tendencies of the full range of stories he appears in. This synoptic approach has at least the benefit of nodding both to the minutiae of the Holmes industry and to the abstracting tendencies of archetypal criticism - but here it seems important to establish early that there are about ten basic formulae at work, and not just one. Some of these are best explored by looking closely at particular cases, but some emerge most clearly in the lump. Indeed, if we substitute 'story' for 'man', Holmes's own pronouncement in The Sign of Four provides what could almost be the motto for this approach:

    What follows only makes full sense in conjunction with the list, "The Contents and Outcomes of Holmes's Cases."  To include specific titles would have been cumbersome, but my general categorisation aims finally to illuminate every single story by indicating underlying themes which might otherwise be overlooked. While some cases inevitably straddle categories or fail to match up in every detail I have tried to cut as few comers as possible in the name of symmetry or neatness. As a final explanatory point: I have put 'weak' and/or late works on a par with more obviously important ones because, though some stories read like simple copies of others, the repetition of a formula still indicates its comparative importance for author and readers alike.

    There are sixty Holmes stories in all, including the four novels. Of these the clearest-cut sub-category is the 'secret society' story concerning a full-blooded secret organisation with countless members, an initiatory code, and the will and power to destroy its enemies. There are eleven such stories, in nine of which the main secret society featured is foreign. In six, which include the two novels A Study in Scarlet and The Valley of Fear, the secret society is American, and in three, European (though some of the American groups have European origins). The American stories run quite rigidly to type: a fugitive from the society either kills or (more usually) is killed by its agents. Despite Holmes's best efforts, the society usually evades him. The European groups - Russian anarchists, the Italian Mafia, a German spy network - are less predictable in their behaviour and final treatment. The British group which is central to only two stories (though it also crops up in The Valley of Fear), the Moriarty organisation, forms an apparently unique bridge between the American-style secret society and the more miscellaneous indigenous crimes described in other tales.

    A related group of stories concerns a conspiracy of silence between those in the know about a crime committed in the colonies or at sea, which leads to the blackmail or revenge-killing of some apparently respectable resident of the Home Counties. There are five stories (including the novel The Sign of Four) which conform to this pattern. In these stories the idea of a silent bond, whose actual or threatened betrayal is the subject of the story, constitutes a significant link with the 'secret society' stories. In both types the notion of foreign evil coming to roost on British soil is crucial.  Lumping the two types together, this broad theme of 'conspiracy' is the subject of sixteen of the sixty stories. Furthermore, it is the subiect of three of the four novels, and helps to account for their greater length, so that in raw bulk it occupies a good third of the total Holmes canon.

    The next large group of stories to be considered involves professional criminal gangs. There are nine of these. In two, which
partly overlap with the foreign conspiracy group, a young man is secretly hired to do a job for a ruthless gang of mixed nationality, who violently threaten his life before fleeing abroad. In another story a British gang takes a 'conspiracy'-style revenge on a former member. Apart from these cases, British gangs are not homicidal, only incidentally so, but are far more likely to be caught. In three closely related stories, a team of two (or in one case the survivor of a duo) removes an innocent party from the scene of a proposed robbery by sending him on a skilfully fabricated fool's errand.  Though appealingly clever, these duos contain firmly delineated 'outsider' elements: an effeminate aristocrat, two Jewish brothers, an ex-Chicago gangster. Other professional gangs are more miscellaneous, but all partly foreign: an Australian husband-and-wife team, an incompetent duo headed by an Italian count, a largish gang of thugs denoted mainly by its one black member. Except for this last one, professional gangs operating on British soil are small: two or three as against the massed hordes of the foreign-led secret societies.

    A further clear-cut group of five stories about criminal associations, if not about gangs, is the 'inside job' group. In three very
similar cases a man in a position of trust has a vital government document stolen by a member of his family for foreign spies. In two other stories the missing object is not a document but has similar overtones of grave national importance. In all five cases Holmes recovers the object of trust and, in all but one, hushes up the family scandal, though foreign spies and lower-class confederates may be arrested and hanged.

    Criminal groups, including conspiracies and inside jobs, then, account for thirty of the sixty Holmes stories: exactly half. There are fifteen stories dealing with individual criminals. Five, including The Hound of the Baskervilles, deal with an apparently prosperous gentleman, often of the squirearchy and/or a good colonial position, who murders, or in one case imprisons, a relative for money. This group is characterised by an innate savagery expressed in a reliance on vicious animals and/or tropical drugs, and accounts for most of the occasions in the canon when Holmes is threatened personally. (A single story, 'The Reigate Squires', resembles these in atmosphere but has equally strong links with the 'conspiracy' group.  Four stories deal with actual or attempted murder from motives of sexual passion or jealousy: these are confined to the lower middle class, and always involve some striking physical disfigurement as a symbol of the ugliness within. Three stories deal with a thieving servant, also something of a Don Juan, who loses his booty and sometimes his life, even before Holmes enters the scene. Two deal centrally with a man who preys professionally on women, either as blackmailer or seducer, and is destroyed by a past victim whom Holmes allows to go free.

    This leaves no less than fifteen cases where no technical crime has been committed, at least by the living. The most clear-cut
group consists of four 'returned lover' stories. An apparently happily married woman is confronted by an earlier lover, husband or, in one case, child. While her present husband's response ranges from the understanding to the homicidal, the wife's loyalty to her past commitment is applauded, and in two cases where the husband dies as a result of his own violence, it is only on Holmes's discretion that the lovers are ruled innocent. This category has some interesting resemblances to the 'conspiracy' category, in that the old affair which now comes home to roost took place in the 'freer' world of the colonies or America.

    Though otherwise diverse, most of the other non-criminal cases involve a cover-up of some kind. A king tries to hide his past from his fiance; a man woos his own stepdaughter in disguise; a beggar hides the source of his income from his wife; a servant protects a cheating student; a sportsman hides his marriage from his uncle; a leper is concealed by his family; a woman covers up for her twisted stepson; a professor conducts a secret rejuvenation experiment; a squire conceals his sister's natural death from his creditors. Once unravelled, Holmes almost always perpetuates the cover-up on being assured that the peccadilloes involved will cease. Eluding all my categories so far are a well-plotted whodunnit where the only technical crime is suicide, and a story about a giant sea-anemone.

    In summary, then, the Holmes canon divides into four almost equal groups: sixteen conspiracies, fourteen gangs, fifteen solo
villains and fifteen noncriminal cases. Of the forty-five punishable cases, only eighteen actually end in the arrest and legal punishment of the main offender (that is, ignoring henchmen). Of the remaining twenty-seven, eleven end in the culprit's 'onstage' death or maiming, which is generally seen as providential. In seven further cases the culprits escape, only to be providentially struck down later.  Finally, there are no less than eleven cases where Holmes deliberately lets the criminal go (admittedly this includes two technically  'non-criminal'  cases  where,  however,  the  technical judgement has been made by Holmes himself).

    Reviewing the canon for the class attitudes it reveals, we can chiefly generalise that the working class, barely visible in most
stories, proves strikingly incompetent when it does step out of line; the lower middle class suffers from dangerous tensions, but
chiefly in the sexual domain; the professional class, when it is what it seems and not providing a mask for tainted gentry like Professor Moriarty and Dr Roylott, tends to align itself with the detectives; the aristocracy, British and foreign, supplies a number of black-sheep cads and seducers who are rarely brought to book unless by their victims, just as the real magnates can depend on Holmes conceal their peccadilloes although they may have to endure a tongue-lashing or, worse, his silent contempt. By far the most dangerous class is the one to which both the savage squires and the ex-conspirators belong: the rural gentry. Fortunately, in obedience to the Gothic traditions at work in these particular stories they tend to be graphically and horribly hoist by their own petards.

    For future reference, it will be worth looking a little more closely at the two largest categories: the 'secret society' and the related but distinct 'colonial conspiracy'. In the secret society stories three distinctive elements emerge. First, the society is almost supernaturally powerful; secondly, its agents (and/or sometimes its enemies) manifest their presence by a use of codes and secret signs which Holmes occupies the foreground of the story in deciphering; and, thirdly, Holmes's efforts are largely ineffectual, since the threatened deaths usually take place and the killers usually escape, their final punishment belonging either to God or to a revenge-cycle outside Holmes's control. In so far as there is one, the message seems to be that these alien activities are no concern of British justice, though they are of God: hence the blustery weather conditions frequent in these stories. Britain's task, as embodied by Holmes, is to monitor rather than antagonise these powerful forces: hence Holmes's bravura displays of decipherment, which can be read as surrogates for the protection he never in fact extends to the threatened victims. We shall see later how the Moriarty stories ring certain changes within these basic conventions.

    In the 'conspiracy' category, the past crime also sometimes involves a special code, but here Holmes's decipherment is effective at least in procuring rough justice. The ex-criminals run very much to type: rugged and battered, with names like Black Jack of Ballarat or Black Peter Cary. Such a figure has made an illicit pile thanks to the rough-and-ready conditions obtaining in the colonies or at sea, and is either being pursued by an equally rugged ex-partner or blackmailed by an oily witness of his crime. (It is an interesting testimony to the strength of these formulae that in an apparently very different case, the crime-passionel story 'The Cardboard Box' discussed in an earlier chapter, the almost arbitrary combination of a rugged sailor and a sinister witness - here 'Jim' Browner and his Iagoesque sister-in-law - brings in its wake the severed ears which constitute a particularly gruesome version of the conspiratorial 'secret code'.) The Home-Counties setting of many of these stories, plus the common presence of an innocent daughter, suggests an imagery of the crimes of Empire coming home to roost in the respectable heartland, which can be compared to that of The Moonstone. The crime itself is often ambivalently portrayed,
however: as an Indian co-conspirator remarks in The Sign.of Four, 'We only ask you to do that which your countrymen come to this land for. We ask you to be rich' (Doyle, Complete Adventures, p. 147). The fortunes seized by the freebooters seem to be held out to them as the inevitable reward of their enterprising spirit (an implication further confirmed if we include the 'returned lover' category as a subgroup of this one). Accordingly, though the crime eventually claims the lives of its perpetrators, a frank confession can be enough to seal Holmes's lips, while the booty stays in the mother-country, in the redeeming hands of a new generation. A striking case of such collusion is 'The Boscombe Valley Mystery', where Holmes allows an innocent man to remain under suspicion of parricide so that he and his rich bride may remain 'in ignorance of the black cloud' of her father's record of banditry and murder (P· 217).

    It would be possible to carry on, delineating in ever more precise detail the family traits of each subcategory. Of course there are detailed exceptions, but there is also very little doubt that the Holmes 'world' breaks apart into a strictly finite range of possibilities. I would suggest further than this finally leads to a view of the world as thoroughly 'knowable': whatever our surprise at this or that detail, it is more important that, over the whole range, certain expectations are confirmed. Keeping this general background in mind, I propose now to discuss in closer textual detail the 'methods' through which this very specific world is created.

    Whether deliberately or not, this passage from the first Holmes novel, A  Study in  Scarlet,  pays strong indirect homage to both the predecessors it claims to dismiss. The attack on Dupin directly echoes Dupin's own attack on his own precursor Vidocq, for not being superficial enough.  And the grumble that 'there are no crimes criminals in these days', which follows the attack on Lecoq, is verbatim translation.from Lecoq's own mentor, Tabaret. (2)  While Doyle's many debts to Poe and Gaboriau are not his main claim to interest, a brief examination of them provides a convenient introduction to his technique within individual stories: partly because they also help to pinpoint some significant differences.

    It is often pointed out how closely Holmes's character, circumstances and relations with the narrator and the police, are modelled on those of Dupin. (3)  But the same can also be said of such celebrated early cases as the second novel, The Sign of Four, and the first short story 'A Scandal in Bohemia'.

    The murder of Bartholomew Sholto in The Sign of Four clearly echoes the double killing in "The Murders in the Rue Morgue'. Each takes place in a locked upper room inaccessible to anyone of normal physique; each turns out to have been committed by a superhumanly agile but subhumanly savage being, whose keeper, a rough seafarer, has climbed to the window after him but been powerless to prevent the murder. In each case the detective finds evidence on the body of the killer's primitive nature, and further evidence contradicting the police theory that the window was not used. And in each there is a moment of special horror when the narrator almost, but not quite, identifies the perpetrator: 'I felt a creeping of the flesh. . . . "A madman," I said, "has done this deed"' (Poe, p. 334); '"Holmes," I said, in a whisper, "a child has done this horrid thing"' (Doyle, Complete Adventures, p. 112).

    In general, however, Doyle tones his model down and blurs its outlines. Where part of the horror of Poe's murder is its total
arbitrariness, as shown in the heaps of money left untouched near the bodies, Doyle's is just the unplanned offshoot of a robbery; where Poe's baffling locked window is caused by an unprecedented combination of the ape's non-human behaviour and blind chance, Doyle's pigmy Tonga is subhuman enough to instill a general sense of fear but human enough to tidy up after himself, pulling up ropes and shutting windows like any other cat-burglar; where Poe's ape and sailor leave a genuinely sealed-looking room, Doyle's pair leave an open skylight and a superabundance of clues, from coils of rope to wooden-leg-prints, with which to solve the minimal problems as to their means of access.

    'A Scandal in Bohemia', concerning a foreign king's attempt to recover a compromising photograph from his ex-mistress, closely reproduces the basic situation of 'The Purloined Letter'. Though the sexes of royal personage and potential blackmailer are reversed, many other details, down to the bungled initial searches and the rigged disturbance in the street, remain the same. But the philosophical underpinning of Poe's story, the interchangeability of concealment and exposure, disappears completely, to be replaced only by some rather rickety assumptions about female psychology.

    Though the two cases mentioned are by no means Doyle's best bits of plotting, the contrast with Poe perhaps brings out how
much less interested Doyle is in philosophical puzzles. Action comes before elegance, and Poe's probings into the nature of reality are replaced by the 'certainties' that all women reveal where they hide their treasures in a crisis, and that all savages are savage. In these two cases there seems to be a direct link, whether or not intentional, between the specific changes Doyle has made to Poe, and some particularly blatant kinds of social prejudice. The substitution of the defenceless Irene Adler for the powerful Minister D- aligns Holmes with the king's sexist bullying for most of the tale (despite some belated signs of compunction); and the substitution of the pigmy Tonga for Poe's ape provides a kind of  'I told you so' support for The Sign of Four's generally gloomy view of such benighted human groups as Indians ('black devils') and dockworkers ('dirty-looking rascals') (pp. 145, 137).

    As far as Doyle's borrowings from Gaboriau are concerned, there are numerous passages where Holmes imitates the behaviour of Tabaret or Lecoq very closely: one such is in 'The Boscombe Valley Mystery', where his almost physical transformation into 'a dog who is picking up a scent' (p. 211) follows a similar description of  Lecoq almost point by point (Gaboriau, Lecoq, p. 37); another is in A Study in Scarlet,  where his 'little cries suggestive of encouragement .and of hope' (Doyle, Complete Adventures, p. 31), are identical to Tabaret's little cries of triumph or self-encouragement', as he makes virtually the same deductions about the criminal's height from the spacing of his footprints (Gaboriau, Lerouge, p. 26). The
other unacknowledged debt of the novel which dismisses Gaboriau with such contempt is its use of the 'split-narrative' novel form: I shall return to this later in the context of The Valley of Fear, where Doyle at last manages to exploit it effectively. But even when spending 'twenty-four hours', rather than 'six months', in the short stories in which Holmes really came into his own, Doyle continued to borrow plot-ideas freely from Gaboriau's novels.

    One case in which he does this very well is 'The Beryl Coronet', an economical and thought-provoking reworking of the Lecoq novel, Le Dossier no. 113. The banker Holder, Holmes's client, accuses his own son of stealing part of a coronet entrusted to him as security for a loan by a royal personage. Holder's beloved niece, who lives with him and his son, tries to throw blame on her maid's lover, but finally turns out to be the real thief together with her own lover, the dissolute Sir George Burnwell. The son had just wrested most of the coronet from Burnwell when 'caught' by his father, but kept silent to shield his cousin. Holmes recovers the rest of the coronet, though to avoid a national scandal the villain goes unpunished.

    In Gaboriau's Le Dossier no. 113, on which this story is clearly based, an aristocratic villain deposits money in a bank and then arranges to have it stolen by the banker's wife, so that he can put pressure on the banker to give him his daughter's hand. The wife has been persuaded by the villain's accomplice masquerading as her long-lost illegitimate son. Until the end, the chief suspect of the theft is a bank clerk called Prosper, which also happens to be the unusual name of the suspected lover of the maid in Doyle's story. The noble rank and gambling habits of Gaboriau's villain and the 'angelic' appearance of his sidekick are clearly combined by Doyle in the single figure of Bumwell, while the desired daughter and guilty wife merge into the desired and guilty niece.

    In both stories the choice of a banker's household is particularly appropriate because it is precisely his position of public trust that blinds him to the realities of his family's private feelings, and the presence of the all-important safe or bureau in the very home which epitomises his 'security' permits the robbery to strike at the heart of the bourgeois power he represents. But in both stories also, the threat from within is activated by a very specific kind of threat from without: the unregenerate aristocrat postulated by a long melodramatic tradition as the chief enemy of bourgeois order.

    Doyle's concentration of cast whereby the banker's mother and daughter are merged in the more ambivalently placed figure of his niece, adds a source of pressure absent in Gaboriau. In Gaboriau we are placed first in the emotional positions of the young lovers (the daughter and Prosper) and then in that of the banker's wife; in Doyle our point of view is confined more straightforwardly to that of the banker himself. But Holmes's inquiries soon clearly establish that the doting Holder keeps his niece a virtual prisoner, while apparently unconscious of the sexual feelings clearly involved on his part: when Holmes points out that her apparent readiness to stay at home is 'unusual in a young girl', Holder is unaware of the irony of his insistence that 'she is not so very young. She is four and twenty' (Doyle, Complete Adventures, p. 307). As long as she calls him 'dad' and passes 'her hand over his head with a sweet womanly caress', he is sure she will willingly marry his son and perform all the other female roles he expects of her, from doting daughter to motherly angel. The reasons for her final elopement with the deplorable Bumwell are very clearly established in terms of the quasi-incestuous pressures to which, like other young women in the early Holmes stories, she has been subjected.

    Though Doyle forfeits the brilliant irony whereby Gaboriau's noble bank-client pulverises bourgeois notions of security by
stealing his own money, he creates a slyer circularity of his own which obliquely queries the acquiescence of late-Victorian society in the  escapades  of its  future  monarch.  The august  customer  who has started to pawn the crown jewels to pay his gambling debts recalls the future Edward VII, 'Bertie', as clearly as does the shamefacedly-masked 'Bohemian' monarch of the first short story (especially as illustrated by Sidney Paget). It is also relevant that the banker's name is Holder, since his main task is to 'uphold' the good name of a disreputable upper class at whatever cost to his own family and feelings, by 'holding' its property in pawn.  Holder's enthusiastic readiness to sacrifice his own son on the altar of the prince's extravagances exactly echoes the blindness which has already 'sacrificed' his beloved niece to another noble gambler, Burnwell. If Holder denotes the obsequious wait-and-see conservatism of the middle classes in the last years of Victoria, the reworking of Gaboriau's depositer-thief into a mixture of 'Bertie' and the most memorable robber of bourgeois tragedy is even cheekier.  'Sir George Burnwell' transcribes into suitably hellfire terms the fate of Lillo's anti-hero George Barnwell, hanged for robbing (like the niece in this story) a notably thrifty uncle: (4) the knightthood and the final lack of punishment perhaps just indicate how dangerously respectable this kind of figure has become.


(1) For a magisterial introduction to the 'Holmesian' world, see William S. Baring-Gould's two-volume The Annotated Sherlock Holmes (London, 1968), which rearranges the canon in 'biographical' sequence and then scrutinises it for inconsistencies and cross-references, adducing a barrage of 'scholarship' in the process.  One thing this does inform us about very clearly is a certain--indeed a Sherlockian--confidence that such researches will finally yield some indivisible objective truth.  For the 'formulaic' approach, see most of the more seriously critical studies.

(2)  See Gaboriau, Monsieur Lecoq, p. 345:  'The race of great criminals is dying out.'

(3)  See, for instance only, Dorothy Sayers, 'The Omnibus of Crime,' in Robin W. Winks, ed., Detective Fiction: A Collection of Critical Essays (Englewood Cliffs, NJ, 1980), pp. 69-71.

(4)  George Lillo's 'bourgeois tragedy' George Barnwell, or, The London Merchant (1731), in John Hampden, ed., Eighteenth-Century Plays (London, 1928), was still well remembered in 1861, being a favourite of Mr. Wopsle's in Great Expectations.