“Who gets to be a suspect in the Sherlock Holmes canon? Suspicion as a conservative socio-political structure or as narrative brio.”

Suspicion is definitely at home in detective stories. Every literary text is, of course, a network of interpretative possibilities meant for the readers to unfold, but the reading pact of detective stories is explicitly that they should be rich in virtualities in order to engage the readers in what Umberto Eco called “suspicious reading” (“une lecture soupçonneuse” [Eco 104]). Detective fiction is a genre that encourages what Eco called “suspicion and interpretative waste” (“le soupçon et le gaspillage interprétatif” [104])—a genre where suspicion is indeed expected to proliferate and reach interpretative frenzy, before excess is eventually nailed down to certainty, before the multiple suspects are reduced to one single culprit. So, suspicion is a major generic expectation in detective stories—and the canonical adventures of Sherlock Holmes, 56 short stories and 4 novels written between 1886 and 1927, certainly explore the complexities of the links between suspicion and detection but also, I think, between suspicion and narration itself.

What I would like to show in this essay is that Conan Doyle’s treatment of suspicion is a case in point to exemplify the essential literary ambivalence of his text.1 On the one hand, Doyle certainly provides Victorian readers with their comforting list of usual suspects—unsurprisingly, foreigners, the lower-classes, and obviously women, who are all naturalized suspects, and the more intersectional the better (typically a lower-class, Peruvian, striking second-wife in The Sussex Vampire ranks extremely high on the list of suspects, and according to the common law of suspicion, she cannot be anything less than a vampire), but also all those who fail to embody fantasized British character and values—that would be the idle, the gamblers, the womanizers. The text then works as a reassuring disciplinary structure, as an instrument of control, where suspicion is a reflex, an “instinct,” as Watson puts it­—“my unspoken instincts, my vague suspicions” (Doyle, “Hound” 742)2—a mechanical method of designating those who are always already known as deviant: it becomes a mock trial of pre-conceived guilt.

Yet, at the same time as such prejudiced hermeneutic fever is unleashed, it is also exposed in three distinct ways, and the text proves much less conservative than its major movement seems to suggest. First, the suspects very rarely coincide with the actual culprits, which is quite predictable in novels that are supposed to wrong-foot expectations; but the thing is that the actual culprits heavily disturb the typology of the usual suspects, which thus discomforts the sense of social order resolution is supposed to bring. The worm is in the fruit, and those who should be beyond suspicion worryingly betray essentialized hierarchies.

Second and more originally, Sherlock Holmes does not suspect, suspicion is not part of his method—either he immediately knows (and those are the many cases that he serially solves, and that Watson relishes to listing in order to enhance the prestige of his champion), or he is presented with something radically new, something different that does not fall under the capital rule of repetition,3 a rare case that fulfils his expectations and drags him out of “the commonplace”: “My life is spent in one long effort to escape from the commonplaces of existence” (Doyle, “The Red-Headed League” 190)—and he then finds himself in front of a very horizontal map, with a great many possible, un-hierarchized ways that account for all the facts. In those cases when identification is not possible, he totally depersonalizes suspicion­—he does not suspect a “who;” instead he theorizes the many possible “hows” and “whats,” unequivocally marking suspicion as a misleading prejudiced emotion. He leaves this to others, while he indulges in multiple theoretical wanderings.

Finally, what is most remarkable in the text is that suspicion changes objects, and concerns the narration itself just as well as the plot. Once the compulsive course of suspicion is interrupted by Sherlock’s solution, the final outstanding stabilization proves to be highly suspicious. The imposition of a unitary meaning—what Denis Mellier called “the monosemic order, the order in which every sign is referred to, where every trace corresponds to an object that is situated in a reorganized universe where everything, in the end, is meaningful” (Mellier 65, my translation from “l’ordre monosémique, cet ordre où chaque signe est référé, où chaque trace correspond bien à un objet situé dans un univers réordonné, où tout fait finalement sens”)—that eventually disciplines the many lines of escape, seems to be highly arbitrary, and the status of the certainty that is reached highly unclear. The final reduction to “the truth” and the spectacular dismissal of suspicion from the plot displaces and reopens Doyle’s narrative regime of suspicion: suspicion actually becomes a synonym for narration itself.

The first point to be made clear is that there is a dominant pattern of suspicion in the Sherlock Holmes stories, and that such pattern is both a generic requirement, as already said4—a detective story needs to fuel suspicion in order to increase suspense and reader participation—and the sign of ideological bias or predisposition. Analysing the objects of suspicion makes it clear that suspicion is not doubt, that it is not a neutral rational structure aiming at finding the truth, but that it has got its own agenda, its pre-identified objects, in one word its politics—and the politics of suspicion in Doyle’s text are very conservative, totally in agreement with the comforting mission of the detective story, which is to purge society from those who threaten its dominant values and social order.

It is thus quite easy to draw a predictable list of the usual suspects, and to conclude that suspicion works like the fantasy of a disciplinary structure that establishes a coincidence between those who are deemed undesirable and threatening, and those who are guilty. The list of suspicious types is an obvious one to draw—and everyone, regular characters and criminals alike, with the notable exception of Sherlock himself, indulges in the process of suspicion.5 The official police, whose own job it is to suspect in order to defend the state, enforce the common political prejudices, and foreigners are automatically rather high on their lists of suspects as “The Naval Treaty” exemplifies:

The suspicions of the police then rested upon young Gorot, who, as you may remember, stayed over time in the office that night. His remaining behind and his French name were really the only two points which could suggest suspicion; but, as a matter of fact, I did not begin work until he had gone, and his people are of Huguenot extraction, but as English in sympathy and tradition as you and I are. (Doyle, “Naval” 455)

Watson’s commentary is crystal clear: suspects are those that are not “you and I,” those who are simply other—it is Gorot being “English in sympathy” that wipes out Watson’s doubts. Suspecting the foreigners thus becomes an explicit structure: in the “The Adventure of the Speckled Band,” Roma are looked into as suspects for no other reason than that they are Roma, and in “The Adventure of the Three Students,” “Daulat Ras, the Indian” who is “a quiet, inscrutable fellow; as most of those Indians are,” though he is “steady and methodical” (Doyle, “Students” 455) rouses suspicion much more than the very British, very Christian, gentle family boy, the son of Sir Jabez Gilchrist, adequately named to make suspicion desecration. In his own account, always an embodiment of Victorian good sense and morals, Watson makes it clear that suspicion is not personal, that it works through pre-defined criteria, through establishing categories the aim of which is to expel collectively those who do not belong—with the demonstrative “those Indians,” Watson grammatically relegates foreigners out of the trustworthy, coherent, domestic social body and fuels suspicion.

The typology of suspicion stigmatizes another category, that of the previous offenders, and the real culprits actually play “the once a felon, always a felon” trump card in order to divert suspicion from themselves and frame those who are over-determined as suspects. Sherlock Holmes describes the process in “The Adventure of the Blue Carbuncle,” insisting that one can rely upon the conservative predictable workings of suspicion: “It seems to me, Ryder, that there is the making of a very pretty villain in you. You knew that this man Horner, the plumber, had been concerned in some such matter before, and that suspicion would rest the more readily upon him. What did you do, then? You made some small job in my lady’s room—you and your confederate Cusack—and you managed that he should be the man sent for” (Doyle, “Blue” 255).

In terms of being deemed guilty, of inviting natural suspicion, women certainly make the most long-suffering, most obvious suspects. Totally in keeping with patriarchal ideology, the characters of the Sherlock Holmes stories explicitly brand women are as “natural” suspects, and the references are countless. According to the overtone of collective textual suspicion, and the underlying fears and fantasies it exposes, the very sacred Victorian value of domesticity is endangered as women are first potential husband killers—and indeed, those who suspect them most are the husbands themselves: “Are you aware that Mrs. Hilton Cubitt has herself lain under grave suspicion of the murder of her husband, and that it was only my presence here, and the knowledge which I happened to possess, which has saved her from the accusation?” (Doyle, “Dancing” 525).

Men’s homes, sweet homes—which Ruskin desired to be “a place apart, a walled garden, the shelter not only from all injury, but from all terror, doubt and division, a sacred place, a vestal temple” (“Sesame and Lilies” 122)—are turned into fantasized dangerous lairs of liars and adulteresses: “I said nothing in reply, but turned my face to the wall, sick at heart, with my mind filled with a thousand venomous doubts and suspicions. What was it that my wife was concealing from me?” (Doyle, “Yellow” 356). When it is let loose by passion and fantasy—“all my suspicions rose into a fierce bitter flame” (Doyle, “Yellow” 356)—male suspicion turns a protecting mother into a lethal succubus in “The Sussex Vampire,” sucking the blood of her new-born, inverting her life-giving and nourishing function into an unnatural crusade against males. Of course, this is a less dangerous version for the coherence of the sense of self to see danger in a second, younger, attractive wife, than to realize that the culprit is one’s first-born son, jealous of his little brother and attempting to kill him. Women are all witches, they are all guilty, they incur the flames of their husbands’ suspicion, in a word, they are natural suspects, as is made clear in “The Crooked Man”: “The lady, against whom naturally the strongest suspicion rested, was removed to her room, still in a state of insensibility” (Doyle, “Crooked” 414). A very telling instance of the urge to suspect women is found in “The Yellow Face,” which holds a unique status in the canon, as it is Sherlock’s only lapse: he, for once, jumps to the wrong conclusion, and it is tellingly a woman who is the collateral victim. He recognizes without the shadow of a doubt the typical pattern of lechery and adultery behind the complex case of a hidden black daughter, as the simple present of general truth indicates: “This woman’s first husband is in that cottage.” Despite Watson’s warning that “it is all surmise” (Doyle, “Yellow” 359). Sherlock sticks to suspicion, before Effie’s alternative version exposes the mechanical workings of suspicion against women, and the Victorian fear that domesticity might be endangered.

There is in the text though a form of soothing compensation for the alleged “natural guilt” of women, an element that restores male control over their households, and that is part of the domineering politics of suspicion: servants are deemed beyond suspicion, they are seen as utterly devoted to their masters. We can quote two cases in point, one from “The Adventure of the Three Students,” and another from “The Adventure of the Blue Carbuncle”: “The only duplicate which existed, so far as I knew, was that which belonged to my servant, Bannister—a man who has looked after my room for ten years, and whose honesty is absolutely above suspicion” (Doyle, “Students” 597) ; “I have three maid-servants who have been with me a number of years and whose absolute reliability is quite above suspicion” (Doyle, “Beryl” 304). Even women are made less dangerous when reduced to the status of servants and in both cases nothing less than “absolute” reliability will do to reinforce social and “absolute” male domination.

Class structures thus work as a more dependable instrument of power than gender structures and domination is more easily achieved over servants than over women. In terms of class indeed, those who are branded as suspects are rather those failed gentlemen who betray the ideal of the British character, and loose morals among the upper-class amount to social disloyalty: gamblers, idlers, and womanizers underscore the values of the protestant work ethic and are what we could call “unnatural” suspects, inassimilable inside traitors, as the case of young Miles McLaren exemplifies: “Miles McLaren. He is a brilliant fellow when he chooses to work—one of the brightest intellects of the university; but he is wayward, dissipated, and unprincipled. He was nearly expelled over a card scandal in his first year. He has been idling all this term, and he must look forward with dread to the examination” (Doyle, “Students” 601). To the client Mr. Soames, Miles McLaren is thus the “least unlikely” suspect, more irregular than perfect English family-boy Gilchrist of course, but even shadier that the Indian, showing that traitors to their class are even worse that foreigners—although there might as well be another hint in the fact that the suspect is Scottish as well, not English, wilder, always slightly suspect.6

As a final illustration, and to show that suspicion is decidedly cumulative or intersectional, we can quote the example of Laura Lyons, who is triply the ideal intersectional suspect in The Hound of the Baskervilles: a woman, and an attractive one as well, she is also a foreigner from Peru, lower-class, and it is hinted at that her morals are quite loose, and she has challenged patriarchy twice by disobeying her father, and by decoying an older richer man. She is such a multifactorial target that the actual culprit, Stapleton, actually bets on the predictable logics of the politics of suspicion, and frames her because he knows she has got all it takes to be designated as main suspect: “He said that […] I should certainly be suspected if the facts came out. He frightened me into remaining silent” (Doyle, “Hound” 753).

To sum up, in the first movement of the text, the Sherlock Holmes stories unleash suspicion, and in the process, Doyle rather closely defines a politics of suspicion: the police, the clients, the witnesses, Watson himself amply suspect, but they suspect along very predictable narrow social, political, national, and gender lines. One of Watson’s trains of thoughts illustrates the incoherent gap between the tenuousness of the proof and the automatic jumping to categorised conclusions: “It had struck me that it was possible that some love intrigue was on foot. That would have accounted for his stealthy movements and also for the uneasiness of his wife. The man is a striking-looking fellow, very well equipped to steal the heart of a country girl, so that this theory seemed to have something to support it” (Doyle, “Hound” 717). The contradiction is obvious between, on the one hand, numerous lexical and grammatical markers of doubt (possible, some, would, seemed, something) and, on the other, ready-made phrases and stereotypes (“some love intrigue was on foot”, “to steal the heart of a country girl”)—there is no ground for such certainty apart from its pre-construction, and suspicion is thus exposed for what it is: a disciplinary tool to brand the pre-constructed undesirable, “hunting down deviance” (my translation: “une traque de la déviance” [Thibaudeau 2]): “suspicion in fact targets what is already known by the subject” (my translation: “le soupçon porte en définitive sur ce qui est déjà su par le sujet qui le forme” [2]).

***

Yet the common politics of suspicion are opposed by both ideological and narrative resistance, and the Sherlock Holmes stories undermine their own major show of suspicion in three dissenting ways. First, suspicion is marked as a fantasy, and the actual culprits, far off its pre-defined targets, quite systematically undercut its conservative laws. Second, Sherlock himself never suspects, or rather rephrases suspicion as an oxymoron, and thus dismisses it as a method. Third, the arbitrariness of the final reduction to “the truth” transfers suspicion onto the narrative regime itself: as a method to arrest suspects, suspicion proves highly prejudiced and erroneous, but as a device to disturb realist narration and foster textual rewriting, it is quite an effective textual virus.

First then and to make a quick important point, Doyle draws a much less Manichean picture of guilt than the list of suspects seems to imply, and the culprits are often hunted out from among highly respected English gentlemen, pillars of the most respected national institutions, forcing the Victorians to consider deviance as part of the self, from within the social body. In “The Adventure of the Three Students,” typically, the prognoses of suspicion are totally inverted, and the guilty prove to be a very “unnatural” party—an alliance between those beyond reproach, Gilchrist and the faithful servant. Women are also thoroughly exonerated in the text, despite Holmes’s legendary misogyny: the cohort of would-be suspects get inverted as so many victims of male mistreatment. In The Hound of the Baskervilles, the female suspects are all cleared, and the villain is a member of the English aristocracy passing as an amateur botanist—though Pierre Bayard, in L’Affaire du chien des Baskervilles, typically returns suspicion where it belongs, and proposes an alternative where Beryl Stapleton as well could be the culprit, framing her husband. In “A Scandal in Bohemia,” the culprit is no other than the king of Bohemia himself, and there are many stories where outwardly extremely respectable gentlemen, fathers, often scientists, prove to be guilty, and their crimes are motivated by the baser instincts, greed, envy, lust more often than not—for example the presumed innocent fiancé, Hosmer Angel, whose name is an efficient decoy, is after Mary Sutherland’s money in “A Case of Identity.” In “The Copper Beeches,” another woman tagged by her name as a presumed Amazon, Violet Hunter, falls prey to a landowning gentleman, Mr Rucastle (property granting him a birthright to respectability), only for Sherlock to discover that Rucastle sequestrated his own daughter to get her money. The same model is reproduced in “The Adventure of the Speckled Band,” and in “The Adventure of the Crooked Man,” where Nancy Barclay is the prime suspect in the death of her husband, Colonel James Barclay, before it turns out to be a case of betrayal among the military—another institution supposed to defend the country. Eventually, Moriarty and his right hand, Colonel Sebastian Moran, both belong to venerable British institutions, the university and the army, a characteristic which finishes to dismiss the policing cartography of suspicion: those who are deemed “abject” (Kristeva 1) actually belong in the heart of the society of gentlemen, and the sense of Victorian identity is questioned, as otherness is flushed out from the self, and not relegated to disreputable margins.

Another central element in Doyle’s denunciation of the politics of suspicion is that Sherlock Holmes dismisses suspicion as part of his method, or rather provides a very oxymoronic definition of the word: “One hardly likes to throw suspicion where there are no proofs” (Doyle, “Students” 600). Such a sentence presents a strong paradox indeed, nearly an antiphrasis, as the Oxford English Dictionary precisely indicates that suspicion intervenes in the space left open to speculation by the lack of knowledge: “a feeling that somebody has done something wrong, illegal or dishonest, even though you have no proof”—yet this illogicality somehow reads like a definition of Sherlock’s method. For him, suspicion is always more than the common definition of suspicion, it is characterized by its own excess and so it is to a great extent an empty category, as he suspects only when he has proof. In “A Case of Identity,” he says, “It was obvious from the first […] my suspicions were all confirmed” (Doyle, “Identity” 201); in “The Five Orange Pips,” he says, “my suspicion became a certainty” (Doyle, “Pips” 229); in “The Cardboard Box,” he answers Watson’s question, “‘I presume that this Jim Browner, the steward of a Liverpool boat, is the man whom you suspect?’” by making suspicion a synonym to certainty, “Oh! it is more than a suspicion”(Doyle, “Cardboard” 895).

Two elements disqualify suspicion in the eyes of Sherlock, and the first is that one should never theorize in advance of the facts. As early as “A Scandal in Bohemia,” Sherlock states that crucial rule in this method: “It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts” (Doyle, “Scandal” 163), and this becomes a regularly rephrased leitmotiv. In “The Adventure of the Speckled Band,” he insists on “how dangerous it always is to reason from insufficient data,” while in The Hound of the Baskervilles, he urges Watson: “I wish you simply to report facts in the fullest possible manner to me, and you can leave me to do the theorizing” (Doyle, “Hound” 698). Thus, Sherlock never suspects—he scans the plurality of possibilities, all of which are equally possible for the very reason that they cover all the facts.

The second element is that suspicion, like love, is defined as an “emotional thing,” that dangerously jams the functioning of reason. Sherlock Holmes who defines himself as a rational creature—“But love is an emotional thing, and whatever is emotional is opposed to that true cold reason which I place above all things” (Doyle, “The Sign of the Four” 157)—thus depersonalises suspicion. He never suspects a person—he suspects the most probable course of events that fits the facts best, he suspects a configuration, as is made clear for example in “The Adventure of the Speckled Band”: “The discovery that this was a dummy, and that the bed was clamped to the floor, instantly gave rise to the suspicion that the rope was there as a bridge for something passing through the hole and coming to the bed” (Doyle, “Speckled” 273). The direct object of suspicion is never a subject, and this is of course a major departure from the rest of the characters, and crucially from Watson, who basically suspects all those he does not like, and automatically associates “feelings and suspicions”: “One page is missing, but otherwise they are exactly as written and show my feelings and suspicions of the moment” (Doyle, “Hound” 712). Suspicion is dangerous to Sherlock precisely because it is a blinding emotion—“My mind was instantly filled with suspicion,” says Munroe in “The Yellow Face” (Yellow “Yellow” 357)—a vague and fickle feeling, a passion, and Watson gives him ample justification: “Sometimes I suspect Barrymore of being a domestic tyrant. I have always felt that there was something singular and questionable in this man’s character” (Doyle, “Hound” 715). But then he switches to another easy suspect, moved by instinct and returning to the idea of the “natural suspect”: “All my unspoken instincts, my vague suspicions, suddenly took shape and centred upon the naturalist” (Doyle, “Hound” 742). As opposed to those shaky foundations of suspicion on vague feelings and likings, Sherlock’s own association displaces the definition—suspicion is a theory, it holds, it can be proved: “I will not bias your mind by suggesting theories or suspicions, Watson” (Doyle, “Hound” 698).

***

Such revised definition—“One hardly likes to throw suspicion where there are no proofs” (Doyle, “Students” 600)—is what we could call an analytic flawless version of suspicion, where suspicion defines itself more as an epistemological method than a hermeneutical device, emptied of all feelings and doubts. It enables Sherlock to convince the readers that the solution he comes to is “the truth”: the realist text fulfils its controlling mission with Sherlock’s narration coinciding with the facts, filling in the blanks, faithfully reproducing reality. He ratifies the fantasy of a text that is competent to comprehend reality, to frame it in linear and causal order. Sherlock’s literary mission thus comforts and duplicates the social mission of the text: literary order reinforces social order as society is purged of one of its enemies. The check of analytic suspicion thus ends the mad course of random emotional suspicion—and it does it in a brutal way as criticism on detective stories has established.7

Yet, in the Sherlock Holmes stories, such final stabilization does not really hold, as the end of the emblematic The Hound of the Baskervilles proves. Watson explicitly diverts suspicion from the plot to the narration, and gives it quite another object. It is not Stapleton who is the suspect anymore, but Sherlock’s final text of resolution. Indeed, Watson questions Sherlock’s solution, he does uncover holes and improbabilities in his account of the truth, and confronts his champion: “If Stapleton came into the succession, how could he explain the fact that he, the heir, had been living unannounced under another name so close to the property? How could he claim it without causing suspicion and inquiry?” (Doyle, “Hound” 766). But Sherlock’s answer is much more disconcerting than Watson’s question, as he does not deny that his account of the facts is going to open up a trail of narrative suspicion: “It is a formidable difficulty, and I fear that you ask too much when you expect me to solve it” (Doyle, “Hound” 766). The thing is that Sherlock does not seem to care much, and his flippant outing forces the reader to reconsider the apparent comforting sense of narrative closure, and to realize that the single objective account might well be a possible version of the truth.

A sense of arbitrariness thus contaminates the solution—the more so since, as is quite often the case in these stories, all the reader has is Sherlock’s word, Sherlock’s authority over his audience. In Sherlock’s version, the suspect has conveniently disappeared, there is no body to evidence his death, the only eye witness could just as well be playing her own game and framing her husband—it is the alternative narrative Pierre Bayard chooses to develop in L’Affaire du chien des Baskervilles, also proving there could still be many more substitutes. Sherlock’s version is an explanatory model, a narrative structure that covers all the facts and makes sense of them—and he proposes it as the truth. However, Sherlock himself, in a very famous quote, has made it clear that there are always multiple possible ways to link a number of stable facts: “I have devised seven separate explanations, each of which would cover the facts as far as we know them” (“The Adventure of the Copper Beeches” 323). The final reduction, the final abrupt textual cut does not hold, precisely because the reader has been explicitly told that other explanations would cover the facts, that the version he is given is an aesthetic choice, a stylistic decision—the most spectacular one, the one that makes the best narration.

Narrative suspicion is thus all in that conditional: “other explanations would cover the facts” (my emphasis). The nature of the suspicion lies in the recognition that there are always more narrations than there are facts, that Sherlock’s truth is a stylistic figure, that realism is a fantasy, and Zola’s glasshouse a beautiful deluded mirage: “I wanted a simple composition, a neat language, something like a glasshouse allowing ideas to be seen inside, human documents delivered in their stern nudity” (Zola 92, my translation: “Je voulais […] une composition simple, une langue nette, quelque chose comme une maison de verre laissant voir les idées à l’intérieur […], les documents humains donnés dans leur nudité sévère”).

The major movement of the text, which presents language as a transparent medium to access documentary truth and systematize reality through the coherent structures of a single perfectly mimetic text, is thus opposed to a dissident minor movement, where language is re-defined as a desiring machine, producing texts the nature of which is to be endlessly rewritten. It allows us to redefine suspicion as a compulsion to narrate ad libidum, “seven separate explanations” that narrate away the facts. We can return to the definition Eco gave of suspicion within the plot of a detective story as “interpretative excess” (“gaspillage interprétatif”) and rephrase it for narration itself. Within Doyle’s narration, suspicion would then be the mechanism of textual excess—the celebration of an endless obsessional desire to produce text, to exceed the frustrating stabilized unicity of the ending, and to have it both ways—both a comforting heavily determined finale, and the sense that it is all a show of fiction, a celebration of the ability of fiction to exceed the facts. Suspicion then becomes a concept to analyse narrative excess, or as Barthes put it, narrative brio: “The brio of the text would be its desire to climax: the way in which it exceeds what is asked, goes beyond babbling and tries to overflow, to counter the firm grasp of adjectives” (Barthes 1501, my translation: “Le brio du texte, ce serait sa volonté de jouissance: là même où il excède la demande, dépasse le babil et par quoi il essaie de déborder, de forcer la mainmise des adjectifs”).