The Indian Census: Risley and the Racial Type

FROM THEIR FIRST ARRIVAL in India, the British collected political, economic, and ethnographic information about the diverse peoples among whom they found themselves. For much of the history of the British East India Company (1600 to 1858), such information-gathering was guided primarily by the need to conquer new territories and to raise land revenues for profit and for defense of British interests. However, the Indian Rebellion of 1857–58 brought about a sharp rethinking of British efforts to assemble data about the Indians they governed.²

Most Britons viewed the Revolt as an expression of Indian anger with perceived attacks upon their social and religious customs. To forestall the possibility of another rebellion, the British largely abandoned a liberal policy of social reform that was seen to have fed this anger and substituted for it a more conservative policy of authoritarian rule that focused on supporting traditional landed elites (Metcalf viii). In an attempt to know the peoples of traditional India better, British authorities transformed an official epistemology based on history into one founded on anthropology, an anthropology that took the category of caste as the enduring principle on which a predominantly Hindu society was founded. Thus arose what Nicolas Dirks calls “the ethnographic state” (43). The problem, however, with focusing on caste and other features of traditional society was that the idea of change within such a society was effectively repudiated. Thus, according to Dirks, “the relentless anthropologizing of India served the larger purpose of misrecognizing the social and historical possibilities for nationalist awakening, even as it worked to reify categories of social classification” (153). Under the regime of the ethnographic state, the British denied Indians, individually and collectively, a capacity for complex political agency.

In 1871 the effort to produce a definitive demography and ethnography of India resulted in the establishment of the first all-India census.³

As might be expected, caste held pride of place in both the demographic and ethnographic portions of the census. As a result, an enormous taxonomy of the Indian populace came into being, in which census-takers sought to place every subject of the diverse subcontinent upon a stable grid of classification by caste, by religion, or, in the case of those aboriginal peoples who fell outside the caste system, by tribe. An important aspect of this taxonomy was its hierarchical arrangement. Although census officials sought to avoid inter-caste antagonism, they tended to follow the advice of Brahman consultants in formulating their classificatory schemes. Thus, the British construction of a caste society bore an unmistakable Brahman stamp–one which reflected the ranked divisions of a Brahman reading of the ancient Sanskrit scriptures: the Brahmans at the top of Hindu society, followed by the Kshatriyas, then the Vaishyas, and at the bottom the Shudras (Dirks 150–51, 209–10).

In placing Hindus along a graduated series by caste, British ethnographers often froze their subjects in arrangements that failed to recognize the fluidity of social relations in India, relations that were affected not only by movements within the caste system over time but also by modifications of traditional political structures due to changes in caste and class. Thus, as Bernard Cohn notes, caste was lifted out of its living contexts and transformed into “a ‘thing,’ an entity, which was concrete and measurable [… and] had definable characteristics.” “Pigeon-holed” into caste slots, individual Hindus found their identities narrowed to a one-to-one correspondence between person and caste (“Notes” 15). Fixed in this way, Hindu subjectivity was reified in ways that emphasized the timelessness rather than the historicity of Indian actions.⁴

The early Indian census culminated in the Census of 1901 under the direction of the most powerful census commissioner of the late nineteenth century, Herbert H. Risley. In conjunction with this census, Risley was able to launch the first Ethnological Survey of India, also under his supervision. This dovetailing of census and survey represented the high-water mark of the ethnographic state of British India. Under Risley, the 1901 Census also represented the convergence of turn-of-the-century Anglo-Indian and European ethnographic thinking–that is, the merging of a Brahman interpretation of caste and a tradition of European physical anthropology based on the idea of the racial type. In The People of India (1908), in which he summarized his census findings, Risley acknowledged the influence of the Brahman concept of social precedence enshrined in the four varnas, or caste divisions (111). In the same monograph, he also made clear his allegiance to the theory and practice of physical anthropology, a movement that began within ethnography in the early nineteenth century and peaked at its end.⁵

Physical anthropology sought to measure, through anthropometric calculations (particularly of headforms), the differences among human beings as a way of inferring the distinctive racial types from which they sprang and to which they all, despite racial mixture, still belonged. Risley argued that the Indian populace comprised seven racial/physical types (32–47). Because of the strict laws of endogamy within both Hindu and Muslim communities, he claimed that these types existed in the population of the subcontinent in a purity that was rare in the world. He acknowledged that some mixture of types had occurred over time, but that, in the absence of continued crossing, the progeny of mixture tended to revert to their original types (27). The idea of reversion held a powerful place in the arguments of physical anthropologists; through its mechanism they were able to contend, even in the teeth of Darwin's theory of evolution, that the main racial types had remained separate and unchanging since the origins of humankind.⁶

Risley's most important claim in The People of India was that caste had its beginnings in racial sentiment (5–6). When the fair-skinned, long-headed Aryans began their invasion of the Indian subcontinent in 2000 B.C.E., they kept themselves aloof from the dark-skinned, broad-headed Dravidians whom they conquered. From that two-fold distinction ensued the four-fold division into the four varnas (the Aryans occupying the Brahman, Kshatriya, and Vaishya caste orders and the Dravidians making up the Shudra order), which in turn gave rise to the present-day taxonomy of seven racial/physical types–an arrangement that, given the fundamental immutability of type posited by physical anthropology, would likely last for all time.

Although Risley claimed that he took his idea of social hierarchy from Hindu sources, he, like many of his fellow Anglo-Indian anthropologists, also embraced European notions of stratification. Unlike some of the early physical anthropologists, Risley avoided making crudely pejorative discriminations among racial types. For example, he stressed that differences in headform had no correlation with differences in intelligence or culture. Thus, Risley distanced himself from the idea of linkage–the idea, widespread among physical anthropologists, that measurable external characteristics were tied to less measurable internal qualities such as intellectual capacity or moral character (27).⁷

Yet, if Risley refused to make invidious racial distinctions among Indian types, he was not above making belittling racial discriminations between Indian and European types. Like many Anglo-Indian ethnographers after the Uprising, he saw Indians as so hopelessly divided among themselves as to be unable to conceive of themselves as a nation. Whereas Europe had long ago formed into a group of nation-states, India had no sense of national unity: “There is…no national type and no nation or even nationality in the ordinary sense of these words” (26). The impediment to nationhood was clear: “Caste in particular, an institution peculiar to India, seems at first sight to be absolutely incompatible with the idea of nationality…” (300). In his denial of effective political agency to Indians, a constant note in the discourse of Anglo-Indian ethnography after the Rebellion, Risley implied India's inferiority to Europe.

If Risley and others pursued ethnography in a spirit of disinterested scientific inquiry, they also offered their ethnographic knowledge to the governors of India in the interests of efficient administration. The census gave rise to a vast array of gazetteers, handbooks, manuals, and other texts that provided district and provincial administrators, army officers, plantation owners, and others with the information they felt necessary to know and to manage the peasants, sepoys, clerks, and coolies who fell under their control. The discourses of the census powerfully molded perspectives on India not only among Anglo-Indian officials but among late nineteenth-century British fiction writers, like Doyle, who wrote about India. We will see this shaping power especially in that writer's susceptibility to the language of racial type, a susceptibility that reveals, as it does in Risley, the foreclosure of political consciousness and activity in Indians.

The Sign of Four and the Racial Type

BY SETTING RACIAL, caste, and tribal types in authoritative taxonomies, the Indian census formed an epistemological template that left its stamp on imaginative literature about the Indian subcontinent. Rudyard Kipling's Kim is the most prominent example. From the opening tableau of Kim and his Hindu and Muslim friends wrestling atop Zam-Zammah to Colonel Creighton's leadership of the Ethnographic Survey to Kim's mapping of the land and peoples of India as a member of the Great Game, the novel is steeped in the language and structures of the ethnographic state metonymized by the census. Yet the census also left its mark on colonial fiction set in England. It may sound odd to speak of colonial fiction set in the British imperial center; however, Said has stressed the need to read colonial literature in broader, more global terms than we have customarily done, to perceive that imperial core and periphery are often complicit in the enterprise of colonial fiction (74–80). The Sign of Four enacts this complicity clearly. Although the story begins in India, it ends in London, with imperial center and perimeter being drawn together in mutual interchange.

Published in 1890 in Lippincott's Monthly, The Sign of Four was the second work by Doyle to feature the extraordinary talents of his master sleuth, Sherlock Holmes. In this novella, Holmes is hired by Mary Morstan, the daughter of Captain Morstan, to assist in the righting of a wrong perpetrated against her deceased father and herself. As revealed later by the thief Jonathan Small, the injustice against the Morstans involves a treasure hidden by four conspirators–three Indians and Small–in the Red Fort at Agra during the Indian Mutiny of 1857–58. Before the four can lay hold of the treasure, they are caught by the British authorities and jailed at the penal colony on South Andaman Island. While serving time, Small tells the story of the treasure to the penal colony officers Major Sholto and Captain Morstan, promising them shares of the treasure in exchange for helping the conspirators to escape. Sholto and Morstan agree, but Sholto reneges on the deal, stealing the treasure and returning to England, where he comes into sole possession of the treasure when Morstan dies. Sholto refuses to tell Morstan's daughter, the late Captain's heir, about her portion of the treasure. However, after the Major's death, his sons, Bartholomew and Thaddeus, tell Mary of the treasure. Before she can receive her share, Small, seeking revenge and his part of the heist, enters Bartholomew's house. Tonga, an Andaman Islander whom Small had befriended while in prison and engaged as a junior partner, kills Bartholomew with a poisoned dart from his blow-pipe. Small and Tonga abscond with the treasure, but Holmes, Watson, and the police overhaul them in a wild boat chase on the Thames during which Tonga is killed and Small is captured. Small reveals that he has dumped the Agra treasure in the Thames. Watson is engaged to Mary, and Holmes returns to his solitary cocaine habit.

The discourse of racial type exerts a decisive guiding force on much of the Holmes canon.⁸

In The Sign of Four, we can see this influence in the description of the Andaman Islanders that Holmes reads to Watson to prove his hunch about the ethnicity of Sholto's murderer. This description appears in a work that Holmes identifies as “the first volume of a gazetteer which is now being published” (127; ch. 8). The Indian gazetteers were vast, alphabetized summaries of geographical, historical, and ethnographic information about the subcontinent.⁹ The particular work that Holmes retrieves is probably meant to be the first volume of The Imperial Gazetteer of India, compiled under the direction of William W. Hunter and first published in 1881. The Indian census and Hunter's gazetteer were intimately linked. According to Bernard Cohn, “The census was the necessary prerequisite…for the Imperial Gazetteer” (“Census” 242). In his preface to the first edition, Hunter acknowledges that his gazetteer is based on the Census of 1871–72 (xi). The entry on the Andaman Islands as it appears in the novella reads: What first strikes the informed reader of this account is its blatant unreality. It is the rankest tissue of popular stereotype, crude sensationalism, and bare-faced fabrication–an astounding passage in a writer who sought to make the art of detection in fiction “an exact science” (Doyle, Memories 69).¹⁰ In constructing this entry, Doyle may have drawn upon that portion of Hunter's gazetteer that mentions early descriptions of the Andaman Islands as “cannibal islands” (283). However, in every other particular, Doyle diverges from Hunter's entry on the archipelago. Indeed, Holmes's gazetteer is contradicted in every aspect by the serious ethnography of the day represented by the writings of such physical anthropologists as Edward H. Man, William H. Flower, and Maurice V. Portman, all of whom studied the Andaman Islanders in the last decades of the nineteenth century. The Andamanese were not on average “rather below four feet,” as the gazetteer claims; according to Flower, the men averaged 4 feet 8 1/2 inches and the women 4 feet 6 1/2 inches in height (298). They were not “fierce, morose, and intractable,” but, in Man's words, “merry, talkative, petulant, inquisitive, and restless” (27). Where the gazetteer portrays them as physically repulsive, Portman writes, “In appearance…the Andamanese men and the young women are not unpleasing, some indeed are distinctly good looking…” (qtd. in Singh 59). Where the gazetteer mentions their stone-headed clubs and poison arrows, no late nineteenth-century ethnographer lists either. Tonga's use of a poisoned dart to kill Sholto is also flatly contradicted; indeed, the Andamanese had neither poison to tip their weapons nor blow-pipes.¹¹ The most egregious misconception of Holmes's gazetteer concerns the cannibalism of the Andamanese; every late nineteenth- and twentieth-century account asserts the utter groundlessness of this charge.¹²

Yet if the substance of the gazetteer entry is absurd, its underlying concepts match those that undergird the physical anthropology of Risley and others. Most of the central concepts affiliated with the idea of racial type–essence, fixity, distinctiveness, reversion, linkage, and hierarchy–are reflected here. The pervasive use of the copula in the account (“They are a fierce, morose, and intractable people”; “They are naturally hideous”) suggests the idea of essence and fixity. The extreme smallness of the Andamanese as described in the gazetteer points to their distinct difference from other races. Indeed, Risley defined the Andaman Islanders as an unusual, eighth physical type, the Negritos, that fell outside his main catalogue of Indian types (32). The idea of linkage appears in the correspondence between the “misshapen heads” of the Andamanese and their alleged savage behavior. Although other scientific and pseudo-scientific disciplines, such as physiognomy, pathognomy, phrenology, and criminal anthropology also studied such correspondences and doubtlessly informed Doyle's description of the Andaman Islanders,¹³

the physical anthropology of the census has a particular relevance here. The idea of hierarchy is also present in the gazetteer entry. In Hunter's Imperial Gazetteer, the Andamanese are described as “a very low type” (284). The racial science of the day regularly placed the Andamanese at the very bottom of the ladder of races, together with the Bushmen of Africa, the Digger Indians of America, and the Terra del Fuegians. The mention of these peoples in Holmes's gazetteer confirms the close tie between Doyle's narrative and physical anthropology. Only the idea of reversion does not appear in the gazetteer's description, for reasons I will discuss later.

Throughout The Sign of Four, Tonga rarely deviates from type. Indeed, the gazetteer entry anticipates almost all aspects of his characterization. Only once is Tonga humanized in a way that diverges from the tyranny of typological discourse.¹⁴

This happens when Small describes how he came upon Tonga in the jungle “sick to death” with fever (155; ch. 12). Having nursed the patient to health, Small finds the Islander “devoted” to him (155; ch. 12). Yet his tenderness here is cancelled by the rough contempt he heaps upon Tonga elsewhere. Right after he recollects Tonga's sickness, he calls the Islander “as venomous as a young snake” (155; ch. 12). When the narrative shifts to Holmes and Watson's point of view, we find no sympathy for Tonga. When they finally see him during the boat chase on the Thames, they perceive only the hideousness and savagery that the gazetteer sketch has led them to expect.

The conformity of Tonga's characterization to Holmes's gazetteer's entry is the clearest link between Doyle's narrative and contemporaneous British physical anthropology. Yet there is a further way in which Tonga's portrayal reflects the concepts of Anglo-Indian ethnography: its occlusion of political agency in the Indian subject. I noted above how the focus on caste tended to reify that concept; a similar reification occurred with respect to “tribals,” the non-Hindu, aboriginal peoples of the Indian subcontinent. Just as the census-takers lifted caste out of its complex political contexts and placed it in a hypostasized realm free of any relation to colonialism, so these same officials were also liable to sever descriptions of tribals, like that of the Andamanese, from their colonial contexts. Moreover, just as taxonomies of caste served to downplay the political motivations of Hindu subjects, so classifications of tribals paid little heed to the tangled political lives of Santals, Todas, or Andaman Islanders under the British Raj.

Anthropologists did not entirely ignore the colonial situation in their notes on the Andamanese. They viewed the Andaman Islanders' hostility to foreigners as an inevitable response to continual aggression from the outside: the raids of Malay pirates, the depredations of Chinese and Malay traders in search of bêche de mer (sea cucumbers) and edible birds'-nests, and the sometimes brutal incursions of the British themselves.¹⁵

However, the entire project of the British anthropologists who studied the Andamanese in the late nineteenth century was predicated, like Risley's investigation of caste, on the idea of analyzing the pure type of the race to which their human specimens belonged. In their quest for the origin of the Negrito racial type, these scientists sought to strip away the excrescences of colonial influence and reveal the Andamanese in all their pre-colonial pristineness.¹⁶ Likewise, in the case of Tonga, the fixation on type obviates all consideration of the ramifications of colonialism. Doyle's narrative ascribes the Islander's violence not to any legitimate resentment of British invasions of the archipelago but to his race's innate proclivity for monstrous aggression.

The Sign of Four and the Criminal Type

THE DENIAL of the political element in Tonga's character is underscored by his association with another type in The Sign of Four: the criminal. We see this denial particularly in the parallels between the portrayal of Tonga and that of the Mutiny rebels. In its depiction of the Rebellion of 1857–58, Doyle's novella follows the conventional British view of Indian political activity as tantamount to crime. According to David Arnold, “For the British, crime and political opposition were always intimately related” (234). There existed a “colonial tendency to reduce all challenges to power, authority and property to a single ‘law and order’ paradigm” (149). The British tendency to equate political unrest with criminal activity was especially pronounced in Victorian accounts, both literary and historical, of the Rebellion.¹⁷

In T. Rice Holmes's A History of the Indian Mutiny (1883), an interpretation of the Uprising “that had become fairly generally accepted by the end of the century” (Embree 75), the author notes that the Indian rebels “were not a law-abiding people” (45). In a conclusion that curiously nullifies the racial difference that many Uprising commentators introduced into their accounts, Holmes sees resemblances between the rebels and violent ruffians throughout English history: In a refrain that runs throughout his history, Holmes reduces the motivation of all of these agents to the lowest common denominator: “selfish instinct.”¹⁸

Doyle's novella reflects this collapsing of political into criminal activity. The rebels in The Sign of Four are represented not as opponents of a political order against which they might have a justifiable grievance but as disturbers of a social order that they regard with a motiveless malignity. According to Small, the malefactors are guilty of “torture and murder and outrage” (146; ch. 12). While he stands guard over the Red Fort, he hears their “yells and howls” as they reel “drunk with opium and with bang [marijuana]” (147; ch. 12). At moments, in a move typical of fictions of the Uprising, the narrative associates the rebels' criminality with a heathenish diabolism.¹⁹

Small describes the rebel army as “two hundred thousand black devils let loose” (145; ch. 12) and as “fanatics and fierce devil-worshippers of all sorts” (146; ch. 12). Although Tonga never takes part in the Mutiny, the rhetoric of the narrative connects the Andaman Islander with the criminal disorder and savage demonism of the rebels. The adjective “fierce” that Small uses to describe the rebels is employed three times in the gazetteer entry alone to characterize the Andaman Islanders. Small calls his Andamanese assistant “that little hell-hound” and “the little devil” (140; ch. 11). Just as the racial typologies discussed above were often inflected by considerations of religion, so the criminality observed during the Revolt at times received a heated religious emphasis.²⁰ And just as racial ideology tended to obviate the political as a factor of analysis, so religiously-charged references obscured any political aspect that might be relevant. To the extent that Tonga resembles the rebels, the possible political intention in his actions gives way to the criminal.

Tonga's actions in England confirm his criminal character. Indeed, Small's narrative shows the criminal consideration consistently trumping the political. The matter of the treasure begins as a political choice–one involving a rich Indian's decision to hide a part of his fortune as a hedge against financial loss during a time of political uncertainty; thereafter, however, it becomes a matter of private larceny.²¹

The shifting of the treasure to London works decisively to downplay any political significance in the theft. In A Study in Scarlet, Watson luridly describes London as “that great cesspool into which all the loungers and idlers of the empire are irresistibly drained” (15; ch. 1)–a scatological metaphor that comes to epitomize the center of the Empire in all those Holmes stories that feature an invasion of the imperial gothic from the imperial fringe.²² Once the invasion reaches the shores of the Thames, it becomes, regardless of its political import, a matter of domestic order–or, in the image of the cesspool, of metropolitan sanitation. If all dirt is matter out of place, the colonial criminals who trespass against English values and institutions are simply persons who do not belong in English society. In the final reference to Tonga, we find an image that blends the scatological and the foreign: “Somewhere in the dark ooze at the bottom of the Thames lie the bones of that strange visitor to our shores” (139; ch. 10). If the “hall-light” in Mrs. Forrester and Miss Morstan's “tranquil English home” defines the center of British normalcy (116; ch. 7), Tonga's grave in the mud of the Thames marks its antipodes–a reminder of a strange foreign criminality lying at the edge of the knownable and respectable English community.

Tonga's criminality is further underscored by his belonging to a criminalized tribe. As we have seen, the census included ethnographic information on tribals in addition to data on major Hindu and Muslim groups. In the same year (1871) that the first census was initiated, the first Criminal Tribes Act was passed in British India “to provide for the registration, surveillance and control of tribes, gangs or other classes of persons” who were guilty of “the systematic commission of crime” (Gupta 57). The Andamanese were not considered an official criminal tribe. However, the idea of the criminal tribe, by its very name, became associated in the popular imagination with the aboriginal tribes of India, including the Andamanese (Thakkar i). Even administrators familiar with the Andaman Islanders were not immune from exoticizing the indigenous populace as criminal. As Satadru Sen notes, “The official view of the tribal population tended to vacillate between an impression of formidable savagery (that required violent police action by the state), and childlike amorality (that had to be dealt with more gently, even protectively). In both cases, however, the Andamanese were effectively criminalized” (29). As with all categories in the ethnographic state, the classification of certain tribes as criminal served to generate “factualized statements” that looked good on paper but did not necessarily correspond to realities in the field. As in the case of attitudes toward caste, the views on criminal tribes tended to ignore those causes of behavior that might lie beyond the perceived mendacity of the criminal group. Criminality was seen as an inherent quality of certain tribes or castes (indeed, of all Indians in some observers' eyes) rather than a response to external circumstances, whether social, economic, or political. In its portrayal of Tonga, Doyle's novella reflects the effacement of these circumstances as possible factors in the actions of the Andaman Islander.

Nineteenth-century British thinking on criminal tribes was strongly influenced by the idea of the “criminal type” as developed by the Italian criminologist Cesare Lombroso (Dirks 142). In Criminal Man, first published in 1876 and reprinted in condensed form in 1911, he defined this type as “an atavistic being who reproduces in his person the ferocious instincts of primitive humanity and the inferior animals” (Introduction xxiv–xxv). Through anthropometric measurement of the cranium and notation of facial and bodily anomalies, Lombroso claimed that he could identify the criminal type in the general populace. The resemblances between Lombroso's methods and those of the physical anthropologists are obvious–so, too, are the similarities in their conclusions: the idea that the type is a central human feature that determines human behavior, particularly in moments of atavism. Recognizing the affinities between his criminology and physical anthropology, Lombroso called his new science “Criminal Anthropology, which may be defined as the Natural History of the Criminal, because it embraces his organic and psychic constitution and social life, just as anthropology does in the case of normal human beings and the different races” (Lombroso-Ferrero 5). Like the physical anthropologists, Lombroso was guilty, especially in his early writings, of ignoring possible political motivations in his subjects. In his later work–for example, in Crime: Its Causes and Remedies (1906)–he sought to broaden his analysis of criminality by considering the social and environmental factors that, in addition to innate propensities, influenced delinquent behavior. However, in the eyes of his readers, he continued to be identified strongly with his early work on the criminal type.

Lombroso influenced Doyle's representations of the criminal in his fiction. We see this most clearly in The Hound of the Baskervilles. There Dr. Mortimer delights in measuring crania: meeting Holmes, he covets a cast of the detective's “dolichocephalic…skull” (672; ch. 1). The doctor's monograph on “Some Freaks of Atavism” establishes the discourse of criminal reversion that runs crookedly through the narrative (671; ch. 1). Dr. Mortimer expresses his admiration for Alphonse Bertillon, another criminologist who used anthropometry to classify criminal types (672; ch. 1), but his research is also deeply indebted to Lombroso. The concept of the criminal type is also clearly at work in The Sign of Four. The initial measurement of the killer's footprint gestures toward the anthropometry of Lombroso. The total physical description of the Andaman Islander also mirrors the Italian criminologist's attention to facial and cranial anomalies. Tonga's embodiment of the criminal type is plainest in the scene in which Holmes and Watson chase Small and Tonga on the Thames. As the police launch draws near the criminals' boat,

The “misshapen head” and “thick lips…writhed back from the teeth” are physiognomic indices of Lombroso's criminal type. The parody of Darwinian evolution in this passage– “It straightened itself into a little black man”–shows a creature who barely escapes his bestial origins. Lombroso's atavism is, in a sense, a misnomer in Tonga's case, for he hardly unbends enough to allow a relapse into his original type. The lips that “chattered at” Holmes and Watson are emblematic, again, of the text's tendency to render any meaningful intentionality in Tonga mute. In this moment, he is the closest to speech in the narrative, yet, like a monkey, his lips only chatter at their potential auditors. In his simianization, this “subaltern,” in Gayatri Spivak's words, “cannot speak” (308). Just as the ethnographic and criminological discourses on type emptied the Indian subject of any significant political potential, so Doyle's narrative silences Tonga's voice and, with it, any power of the character to convey a sense of political motive.

The discourses of the racial and criminal type meet in Tonga. But what of the other characters in the novella including Small, Watson, and Holmes? Are they, too, subject to the reductive and reifying force of the type? The answer to this question will impinge directly on the query with which we began this essay: whether this narrative serves to confirm or disconfirm Miller's subversion hypothesis.

The Sign of Four and the Subversion Hypothesis

CRITICAL OPINION has divided over whether Doyle's detective fiction supports or subverts the social and epistemological orders that shape its narratives. Ronald Thomas provides a neat survey of this critical divide:

The subtlety and even-handedness of Thomas's approach is attractive. However, he resists spelling out here how the rival oppositional and collaborative tendencies in detective narratives relate to each other. Moreover, he refuses to declare which tendency is privileged in the end.

Up to this point in my reading of The Sign of Four, I have focused mainly on Tonga. Given the conceptual and rhetorical homologies that link the discourses of type and the structure of Tonga's characterization, it would seem that the novella collaborates with the scientific and criminological taxonomies of the time. Yet when we move away from an analysis of Tonga's character to an examination of characters in relation to one another, we find a different story. Jaya Mehta astutely observes the unexpected and disturbing parallels between Small, on the one hand, and Watson and Holmes, on the other (636–37). The ways in which Tonga and Holmes mirror each other unsettles the differences between criminal and detective even more disturbingly. The adjective “fierce” that characterizes Tonga attaches itself to Holmes, too. As the detective urges on the men in the engine-room of the police launch, a “fierce glow from below beat upon his eager, aquiline face” (138; ch. 10); the fierce glow might just as well be beating within him. Tonga performs his “war-dance” when Small exhibits him in their travels (156; ch. 12); likewise, Holmes speaks of “go[ing] on the war-path” in pursuit of criminals (133; ch. 9). Holmes even finds himself an unexpected Doppelgänger of the rebels whose disorder he, as a protector of public decorum, would naturally deplore. While they are drunk on opium and bhang, Holmes enjoys the high of cocaine. Watson observes of Holmes, in a remark that he repeats in one form or another throughout their adventures together: “So swift, silent, and furtive were his movements, like those of a trained bloodhound picking out a scent, that I could not but think what a terrible criminal he would have made had he turned his energy and sagacity against the law instead of exerting them in its defence” (112; ch. 6). The combination in this image of the theriomorphic and the criminal recalls the atavism that figures so prominently in the discourse of criminal anthropology. In “The Adventure of the Creeping Man,” Holmes declares, in words that might apply to himself, “The highest type of man may revert to the animal if he leaves the straight road of destiny” (1082). In the potential collapse of the highest into the lowest, of Holmes into Tonga or the rebels, the narrative of The Sign of Four seems at moments to threaten the scientific idea of type that shapes it and to upset the social and political structures that its protagonist is supposed to defend. In these moments, the subversion hypothesis gains validity.

This, however, is too partial a formulation of the way in which ideological and counter-ideological forces function in The Sign of Four. I have argued above that Tonga is not really susceptible to atavism because he has hardly evolved from his primordial form. Holmes does not revert either. However, unlike Tonga, who has little distance to fall, Holmes resists reversion because he is above falling. At the opposite end of the human scale from Tonga, he exhibits an atavism that is only apparent. Unlike the criminals who have no control over their atavistic lapses, Holmes chooses his transformation into a “bloodhound”–the hunting or tracking hound being the animal into which he is usually seen to change. He is a “trained bloodhound,” one who, through the long and arduous discipline of his art, can assume the form of the bestial hunter at will [emphasis added]. In this, he is singular among all the characters in Doyle's detective fiction, including Watson. A look at the discourse of singularity in the novella will show why this is so.²³

At the start of his search for Tonga, Holmes looks for some “traces of his individuality” (112; ch. 6). However, the detective finds the individual criminal and solves the specific crime paradoxically by linking the unique human being to a type–in this case, that of the Andaman Islander. Holmes reminds Watson of what Windwood Reade, the author of The Martyrdom of Man, says about the individual and the aggregate: “while the individual man is an insoluble puzzle, in the aggregate he becomes a mathematical certainty….Individuals vary, but percentages remain constant. So says the statistician” (137; ch. 10). In what may be the secret of his attraction to readers, Holmes is that individual who remains an insoluble mystery to his fellow human beings, beyond the reach of even those means he uses to solve crimes. His almost supernatural powers of detection make him a figure so extraordinary that he eludes percentages and, therefore, escapes the statistician who would assign him to an aggregate type. In his evasion of type, he is permitted to choose his identity provisionally, to perform the role or roles that will best assist his art. Watson calls the sequence of events in which he and Holmes dispatch Tonga and bring Small to justice “our little drama” (157; ch. 12). But it is really only Holmes's drama since he is the only one who exhibits a performative sense of self on the stage defined by the case. At the conclusion of The Sign of Four, Holmes quotes Goethe: “Schade dass die Natur nur einen Mensch aus dir schuf, Denn zum würdigen Mann war und zum Schelmen der Stoff” [“Nature, alas, made only one being out of you although there was material for a good man and a rogue”] (158; ch. 12).²⁴

The placement of the Goethe quotation at the end of the story underscores the importance of identity in the narrative. The irony of the aphorism is that, since Holmes is a sport of nature, he need not content himself with “one being” but may play at having two or more–even that of a rogue.

The handling of identity in the narrative befits the fin de siècle period in which it was written. During the dinner at which the editor of Lippincott's Monthly commissioned the publication of The Sign of Four, he also commissioned Oscar Wilde's The Picture of Dorian Gray to be published in the same magazine's pages (Stashower 103–06). Doyle may have modeled Holmes in his novella on Wilde, a writer and man he admired. Holmes is, like Wilde, a witty aesthete, a lover of the outré, and a seeker after fresh sensations that sometimes shock the bourgeoisie. In some respects, too, he embodies the views of Dorian Gray and Lord Henry. Dorian “used to wonder at the shallow psychology of those who conceive the Ego in man as a thing simple, permanent, reliable, and of one essence” (111; ch. 11). Lord Henry recommends the artifice of the actor: “Is insincerity such a terrible thing? I think not. It is merely a method by which we can multiply our personalities” (111; ch. 11). It would be hard to find a more devastating critique from the late Victorian period of the idea of the type, or a warmer recommendation to diversify the self.

Where does this leave Miller's subversion hypothesis? Holmes clearly transgresses against the idea of the aggregate that constitutes the type. While Watson goes off at the end of the narrative eventually to wed Mary Morstan and settle into a conventional marriage, Holmes returns alone to his room and his cocaine, “an insoluble puzzle” of a man. The division of rewards here–Watson gets the girl and Holmes the aesthetic and professional satisfaction of a job well done–underscores the aberration that Holmes constitutes within the standard marriage-plot novel.²⁵

Holmes's divigation from the norm established by that genre emphasizes his resistance to conventional notions of type and the disciplinarity it maps. But Holmes is the only character who can manage this feat. The other characters remain true to type and, as such, the concept of the type retains its validity. No matter how much Dorian Gray might scoff, the type “as a thing simple, permanent, reliable, and of one essence” retains its credibility in the epistemology of the novella. The chameleon hybridity that characterizes Holmes cannot disturb the norms of identity-formation that define the other characters. In his discussion of the detective novel, Miller writes of its “final localization of culpability within a single individual” (36), a localization that reassures readers that culpability does not include them. A similar localization extends to the sleuth himself, whose panoptical gaze (potentially all-seeing but itself unseen) restricts itself to the limited area of the criminal.²⁶ I would argue that Holmes is localized in yet another way. Just as too comprehensive a detective gaze might frighten middle-class readers, so also too freely distributed a hybrid concept of self might endanger those bourgeois values and institutions that are founded on essentialist notions of identity. Thus, Holmes's liminality is confined within him alone.

Miller argues that localization serves, in the end, to engender the illusion that persons beyond the detective's gaze are free. However, according to him, surveillance and discipline are in fact total. The technologies of social discipline create the fleeting appearance of freedom only to make the reality of pervasive control more palatable. A similar totalization works in The Sign of Four. The social milieu of the detective novel gives rise in Holmes to a professional of unique talent, the only man who can save the system from those who wish to exploit, scuttle, or invade it. He may seem beyond the system; indeed, it has been argued that Holmes is the outsider who makes the population inside the social system secure (Clausen 114). However, he is not really an outsider but an insider who merely gives the appearance of being alien. In the self-contained and highly-contrived world of Doyle's detective fiction, he is that point of discontinuity within the continuity of the social network that ensures its perpetuation. His extraordinary talent for shifting identities–his apparent reversions to the bestial, his doubling of criminal traits, and his freewheeling use of disguises–is purely instrumental, made to serve the larger social good. To put the matter in terms of this argument, he embodies a freedom from type to preserve the stability of type at home and abroad. He conducts himself as an exception to those “venerable set[s] of factualized statements” about the world in order to prop up taxonomic orders in England and in the Empire. As he acts to confirm the legitimacy of British taxonomies of knowledge about India–in the censuses, the gazetteers, and criminal tribes legislation–so he implicitly supports the British Raj, too. Through his support of these taxonomies, he also helps to ensure that even a rudimentary sense of political agency is moot for the likes of Tonga and his fellow Indians.

It may seem contradictory to speak, on the one hand, of power relations whose effects are totalizing and to maintain, on the other (as I have done implicitly throughout this essay), the potential of Indians to enact political change. However, the contradiction resolves itself, I would argue, when we clarify what is meant by political change. If we mean by it liberation from power relations, then we speak, as Foucault asserts, of an impossibility. However, if we mean by it transformations within power relations–subversions, insurgencies, even revolutions–then political change is possible, for the flows of power are never one-way but are dynamic and reversible. Power is dispersed in ways that may serve either to shore up or resist hierarchies of race, gender, sexuality, or class.²⁷

Even the dominance of Holmes, the guardian of social order, is threatened at times. In “The Scandal of Bohemia,” he is outwitted by Irene Adler. And in “The Yellow Face,” he is so humiliated by his failure to detect a young black woman passing as white that the name of the town in the case, Norbury, serves as a reminder of his comeuppance. But these are minor and fleeting lapses; moreover, they compose no significant pattern of inversion of Holmes's powers. At the same time, they are complemented by no sustained pattern of dissidence from his authority. Professor Moriarty is too perfect a mirror-image, too tightly tied to Holmes as his alter ego, and therefore too phantasmal an evil to pose as a credible threat to the social system the sleuth upholds. Though Doyle's detective fictions are obviously both ideological and counter-ideological in spirit, they are finally deeply conservative fantasies of disciplinary control rather than disquieting fables about the subversion of national and imperial norms. Doyle could not conceive of an alterior racial or criminal type as a worthy agent of political change. That, in 1947, latter-day Indian rebels would create their own fantasy of nationhood and make a new nation would have come as a shock to the creator of Sherlock Holmes.