Engaging with the world

Eriksen's site

  • facebook
  • wordpress
  • yu
  • Academia
  • Nflag






The author as anthropologist

Some West Indian lessons about the relevance of fiction for anthropology

Thomas Hylland Eriksen

In Eduardo P. Archetti, ed.: Exploring the Written. Oslo: Scandinavian University Press 1994.




Fiction and anthropology

The relationship between fiction and anthropology has been discussed from several perspectives in recent years. Some of the contributors to these debates have argued the irrelevance of any absolute distinction between the two kinds of writings, consequently relegating anthropology from the ivory tower to the world of literature; others have tried to justify a clear distinction between the two, often seeking refuge in a Popperian philosophy of science. From the perspective of the creative writer or the literary critic, it might be interesting to identify the influence of anthropological thought on contemporary novels; an influence which is obviously visible in books by such writers as Ursula LeGuin (the daughter of A. L. Kroeber, herself trained as an anthropologist), Kurt Vonnegut (who studied under Redfield at Chicago), the inimitable Bruce Chatwin and the early Anthony Burgess, writing upon his return from several years of colonial service in Malaysia. From an anthropological perspective, a presumed inverse relationship has been investigated; it has, for example, been argued that Malinowski's style was profoundly influenced by Conrad (Clifford 1986).

During the last decade, moreover, considerable attention has been directed towards the distorting and disguising aspects of style present in purportedly "neutral and scholarly" writing (Geertz 1988; Clifford and Marcus 1986; Manganaro 1990), as well as the fact that social reality is often inadequately represented through anthropological language (cf. Bloch 1991).

The main perspective in this article is different from these, and it has been depicted as the bottom-right square of Fig. 1. Instead of regarding the anthropologist as an author involved in creative writing, and instead of comparing the genres in an intertextual way, I shall examine some aspects of the relationships of the texts to other descriptions of that social reality which they purport to represent, and I shall in line with this explore possible practical uses of fiction in ethnographic research. Criteria from anthropology are in other words applied to works of fiction not vice versa. In this way, some differences between the fictitious and the anthropological modes will eventually become evident. The three novels which have been chosen for scrutiny share a sociological preoccupation, and all of them were written by members of the society with which they deal. All of them belong to that realist or naturalist literary tradition where a main aim consists in providing an illuminating interpretation of a society or a social environment.


Let us for once assume, at the outset, that the distinction between fiction and anthropology is a simple one. Fictional accounts, then, present persons and events which have been invented by the writer. Anthropological texts try to present a few aspects of social reality as accurately as possible, taking account of the limitations entailed by fieldwork, "cultural translation" (or, if one prefers, cultural reduction) and attempts at linguistic representation of society.1 Lies and deliberate misrepresentations are banished from anthropological scholarship, which should additionally unlike fictional writing try to present empirical material systematically and comprehensively, and distinguish between description and analysis, so that the reader may draw his or her own theoretical conclusions from it. The rules of consistency and the criteria of evaluation are different within each genre of writing. According to this deliberately naïve view, which will be challenged later on, the main grey zones between anthropology and fiction would consist in travel writing and literary essays dealing with society.

Literature in Trinidadian society

Trinidad is a poly-ethnic island in the southern Caribbean with a population of slightly over a million, forming an independent republic with the lesser neighbouring island Tobago. Discovered by Columbus on his third journey in 1498 and subsequently settled by slaves, colons/adminstrators and indentured labourers from Africa, Europe and India, respectively, Trinidad does not represent that kind of society typical of our shared anthropological folklore; it is a burning hot society, so to speak. Known to the non-Caribbean world primarily for its limbo dancing, its calypso music and its annual Carnival, Trinidad is in many regards a well-integrated part of the modern world (cf. Miller 1993). To the anthropologist wishing to undertake research in such a society, several methodologically and epistemologically relevant aspects crucially distinguish it from tribal societies: (i) The society lacks any pre-colonial history. (ii) The level of literacy is officially nearly 100%. (iii) Capitalism and wagework have been universal since the foundation of the society. (iv) Events from the entire history of Trinidad have been recorded by contemporary chroniclers.

In addition, the recent ancestors of the entire population of Trinidad were uprooted and displaced in historical times. Nobody except a very small group of self-professed "Carib Amerindians" in the Arima area claims the island as their ancestral land. These facts, which are known or at least easily available to non-Trinidadians because they have been stored and distributed through texts, suggest that native representations about their own society and reflections about their own condition could be widespread. This is indeed the case. There are in fact many natives who have reflected theoretically on the foundation of their society and their own social identity, and many of them are likely to challenge the interpretations of a visiting anthropologist in a well-informed way.

Such "quasi-theoretical" statements from informants are sometimes warned against by anthropologists (Bourdieu 1977; Holy and Stuchlik 1983). These statements tend to be regarded as somewhat less authentic and less authoritative than so-called spontaneous utterances. As Bourdieu rightly warns (1977: 21), the study of society should not be confounded with the study of subjective representations of society. The fact is nonetheless that Trinidadians talk and write extensively about themselves and about their own society, even in the absence of anthropologists. This reflexivity is in itself a significant ethnographic fact to a researcher trying to understand the workings of local social classifications and ideology. Consciousness of self and others is reflexively monitored in this society, and it is to a considerable extent mediated through shared communicational interfaces such as the national educational system and the national mass media, including domestically published books. Trinidadian society is inherently sociological in character; its members continuously discuss its nature (cf. Giddens 1990, 1991, on modernity and reflexivity). During my own fieldwork in 1989, I met local writers, essayists and social scientists who had struggled for years trying to conceptualize the essence of their society. Some of them had published books as testimonies to this long-standing attempt. Can we, as foreigners briefly staying in their company, enrich their analysis with our view from afar? It is my view that we can, but any anthropologist working in this kind of society must inevitably be struck with awe and sometimes even a certain humility when facing these individuals and their production.

Does the presence or absence of indigenous fictional texts crucially distinguish societies from each other? Generally speaking, the answer is yes. The widespread social appropriation of texts generally suggests the presence of a certain kind of historical consciousness and a certain kind of complexity in social organization, as Gellner (1983), Anderson (1983), Goody (1986) and others have convincingly argued. More relevantly in this context, the social appropriation of fictional texts, notably novels, may enable the members of society to reflect critically about their own identity, and they also propose and articulate a particular model of the world. In Trinidad, an indigenous genre of popular music, the calypso, has been much more important than written fiction in the creation of a domestic civil society, but novels and novelists can also be important in several ways.2

The widespread presence of sociological reflexivity, that is to say an ability and a willingness to reflect on one's own society and social identity, will be taken for granted in what follows. Presently, we shall turn to the "data", which in this case amount to three novels written during Trinidad's post-war history and their relationship to society. Novels, which are simultaneously the production of a society and contributions to the self-definition and reification of that society, have the additional virtue of presenting some kind of ethnographic evidence although the status of such ethnographic material can be uncertain. In the present readings, the perspective, or Vorverständnis ("pre-understanding", Gadamer's term) is anthropological. In considering the ethnographic and anthropological value of the books, it will be evident that even highly reductionist readings of novels (such as anthropological readings) imply several levels of interpretation, and that it is important to keep these levels apart. I shall distinguish between three such levels of reading relevant in the anthropological appropriation of novels.

The author as ethnographer: A Morning at the Office

Edgar Mittelholzer's A Morning at the Office (Mittelholzer 1979), first published in 1950, can be read as a micro-sociological analysis of social relations at an office in Port-of-Spain. The office has 14 employees who between them span virtually the entire scope of variation with respect to social classification in late colonial Trinidadian society. The classificatory dimensions of ethnicity, class, gender and locality are all covered through Mittelholzer's very varied cast, which even includes an anomaly, namely a homosexual coloured man.3

The simple idea behind the novel consists in describing what happens in the office between four minutes to seven and lunchtime, in order that the reader may observe how a particular pattern of social classification is confirmed and reproduced through the difficult and subtle art of social interaction. Like any good ethnographer, Mittelholzer tries to fuse the universal with the particular and thus accounts for individual idiosyncracies, as well as structural and cultural defining characteristics of the different situations. His cast introduces the secretary Miss Yen Tip, who "was a creole Chinese who could not speak Chinese"; there is Mr Jagabir, the East Indian accountant who unsuccessfully tries to feel at ease in the urbane creole environment of the office and continuously fears that his superiors will send him back to the cane fields; there is the creolized Indian girl Miss Bisnauth who is in love with a coloured artist and rejects the constraints of caste; there is the young black boy Horace whose Uncle Tom attitudes will no doubt help him to a successful career in independent Trinidad a decade later, and so on. Although my fieldwork took place four decades after Mittelholzer's, I have met all these characters.

Read as ethnography in the 1990s, the novel indicates that ethnic relations have changed, and one of the author's most impressive achievements is his depiction of the ambiguous and complex relationship between the colonial white upper class and the indigenous coloured middle class. Since many of them were beneficiaries of the "jobs for the boys" principle, the whites in Trinidad were often of more humble origins than the local coloureds. As the light-brown secretary Miss Henery muses on page 93, after having been humiliated by her boss:

A dirty lot of people. And who was Murrain at all! For all she knew, she had much better class than he. Most of these English people who came out to the colonies were of the dregs. But the instant they arrived they turned gods. Who knew if Murrain had not been dragged up in some London slum? His white skin was all that made him somebody in Trinidad. Her parents and grandparents were ladies and gentlemen

Today, the relationship between whites and coloureds is less important in Trinidadian social classification than it was then, although it remains ambiguous in a similar way. In this novel, further, a great deal of attention is granted to the fine distinctions within the coloured segment; the distinction between kinky hair and light brown on the one hand and straight hair and olive skin on the other is considered important. In contemporary Trinidad, it would seem inappropriate to grant such a distinction great social importance.

Mittelholzer's concern with rank and social classification is evident throughout the book. Through descriptions of bodily movements from gracious and elegant to clumsy and inept, through depictions of the characters' speech, from gross rural Trinidadian creole to Queen's English, and in his descriptions of the relations between the sexes, he also gives the reader abundant information about cultural differences between the rank categories. On this score, Mittelholzer could be challenged if his book is read as an ethnographic description, according to which premisses he might be criticized for portraying the local cultural variation in an exaggerated and biased manner.

Since Mittelholzer's book is a novel, convention dictates that it is not used as hard ethnographic evidence. However, A Morning at the Office is doubtless based on first-class ethnographic field material; it covers many fine nuances of inter-ethnic micro relations, and it is surprisingly comprehensive. It can teach us, for example, that small-islanders from the Lesser Antilles constituted an important category of significant Others for the Trinidadians blacks and coloureds at the time, but not for the Indians and whites. This remains true today.

If one compares its insights and virtues with sociological research carried out in Trinidad during the same period, such as Lloyd Braithwaite's well-known study Social Stratification in Trinidad (1975 [1953]), one is compelled to conclude that the novel defends its place as an important piece of Trinidadian ethnography. In fact, Braithwaite's arguments concerning ethnicity and rank resemble Mittelholzer's, and his evidence is frequently anecdotal and thus similar to that of the novelist. Braithwaite's study lacks some of the detail and introspective qualities of the novel, but contains more comprehensive and accurate descriptions about rank categories, historical circumstances and features of Trinidadian society. Braithwaite's explanations follow the basic Parsonian schema fashionable at the time. In sum, the novel and the sociological study are complementary, and they tend to support each other. Mittelholzer's ethnography is superb, and his examples are striking and rich in connotations this should not come as a surprise, since he has himself invented them. Like a sociological or anthropological treatise, a book like A Morning at the Office can be distorting as well as liberating as an addendum to one's own ethnography. It is littered with ethnic prejudices and attempts to persuade the reader about the validity of a particular model of Trinidadian society. Since its central assumptions are not made explicit and since the argument, as it were, is clothed in the poetic and suggestive language of literature, it can be seductive reading. Since scholars try to present their argument in a clear and unambiguous fashion, it may be easier to argue against a sociological study than a novel because it is easier to discern its central contentions.

There is a second level at which Mittelholzer's novel functions as ethnography. At this level, it can be read as an ethnographic source rather than an ethnographic description. As already suggested, the book is an inadvertent statement of the author's biases and ideological position in multi-ethnic colonial Trinidad. At this level, the author makes spontaneous, non-reflexive and frequently implicit statements about his cultural universe; in Holy and Stuchlik's (1983) terminology, he performs an act rather than uttering a statement. In order to appreciate this aspect of Mittelholzer's novel, one must know something about the author. One will need to know that he was an immigrant from British Guiana to Trinidad, that his social identity from boyhood was that of a lower-middle class coloured, whose main ambition since adolescence had been to live in England and write books for an English audience. Mittelholzer's own positioning in Trinidadian society can thus contribute to explaining his unusual sensitivity to ethnic processes. As a foreigner, he could adopt a fairly detached view, and as a coloured person from a poly-ethnic society similar to Trinidad, he belonged to an ambiguous ethnic category himself. In order to understand the significance of the author's social identity here, one must have additional knowledge of the societies in question. Only then can one discern, between the lines, how Mittelholzer produces through his novels a version of a world where good manners and proper language matter more than racial origins, and where Indian culture is ultimately a crude peasant culture which is justly marginalized in confrontation with the sophisticated, witty and gracious creole culture characteristic of the coloured bourgeoisie.4 At this level, the book cannot be evaluated as ethnography by a reader who is not already familiar with West Indian societies.

Mittelholzer's novel is not very well known in Trinidad, and it is certainly not widely read. Its direct impact on Trinidadian society can therefore be considered negligible, unlike that of the next novel which I will consider.

The ultimate ethnographic fiction: A House For Mr. Biswas

Trinidadian literacy does not entail that most adults read several novels every month; however, large segments of the political and cultural elites do read books, and for this reason it can be assumed that novels written in Trinidad may have profound social and cultural repercussions in Trinidadian society itself. Whereas few Trinidadian novels have been written with such aims in mind, they can sometimes have visible social consequences in so far as an important domestic audience is aware of them. This has been the case with V.S. Naipaul's A House For Mr Biswas, which was first published in Britain in 1961 (Naipaul 1984 [1961]).5 To Trinidadian intellectuals of Indian origins, this was to become the novel which depicted all of their fears and anxieties, their alienation in an "artificial" and fast-changing society, their thwarted ambitions and frustrations of being Indians in a non-Indian environment. Passionately debated at the time and still a highly controversial book, Biswas is simultaneously a rich and sensitive ethnography and an historical event in its own right.

A book which may yet win its author a Nobel prize, A House For Mr. Biswas is widely acknowledged as V.S. Naipaul's masterpiece. Its main plot is simple. It follows the life of Mohun Biswas, a Trinidadian of Indian origin from the Chaguanas area, from birth to death, focusing in particular on the possessions he successively acquires. Towards the end of the novel, he finally acquires his own house thus the book's title and subsequently dies.

A pivotal point in an anthropological reading of the novel could be the contradictions and paradoxes lived in by the "East Indians" or Indo-Trinidadians from the 1930s on. Their dual struggle, represented in Mr Biswas's environment, consisted in modernizing their way of life without losing their Indian identity. Having lost their original language and under strong pressure to modify their customs, many Indians turned strongly traditionalist, while simultaneously trying to catch up with creole society, particularly in economic matters. Mr Biswas, untypical since he rebelled openly against local Hindu tradition, impersonates the many setbacks and disappointments experienced by Indians trying to take on the challenges from modernization and urbanity during this period. Rejected or sneered at by other East Indians, they were also treated condescendingly by the urban creoles; white, coloured and black.

The first major anthropological monographs on Trinidadian East Indians were published at about the same time as Biswas. Both of them (Klass 1961; Niehoff and Niehoff 1960) focus on change and continuity in the Indo-Trinidadian community, and argue with varying strength in favour of the so-called retentionist hypothesis: a main argument in both monographs is that East Indians in the West Indies have retained the essence of their original culture despite having been uprooted since the late nineteenth century.

Naipaul's novel is highly relevant as a context for any reading of these monographs. It could certainly be cited as counter-evidence against a too one-sided defence of the retentionist notion, but it may also serve as a complementary source of information. Since both of the anthropological monographs in question are structural-functionalist in character and emphasise the structural setup of the communities, the novel may add texture, movement, atmosphere and real-life encounters to these descriptions. The novel also has definite ethnographic virtues which the monographs lack; for example, it depicts ruralurban links, aspects of ethnic relations and unique details of family life, which are important factors for an understanding of change and continuity of life in Indo-Trinidadian villages during the postwar years.6 We should also remember that Mr Biswas was written by an author who was in many ways more familiar with Trinidadian society than Morton Klass and the Niehoffs could possibly have been.

Let us first look at presentations of the retentionist hypothesis in the anthropological works. The Niehoffs conclude their study by enumerating the Indian institutions which have survived in the new setting and those which have been profoundly influenced by the surrounding Creole communities:

What can we say now in final summation of the core of Indian society as brought out by the Trinidad situation? There is a strong drive to retain cultural identity. There is a peasant's love of land as well as a drive toward wealth, both characteristics intimately connected with the drive toward continuity in family relations. In inter-relationships with other peoples the Indians are willing to borrow freely in technical improvements, in religious beliefs, and in many other areas of culture except in social institutions. Here again there is an interconnection between the drive toward wealth, the continuity of the family, and the reluctance to tamper with the social organization. (Niehoff and Niehoff 1960: 188)

Morton Klass was a more outspoken defender of the retentionist hypothesis, and this is interesting since he conducted fieldwork in the area where Mr Biswas grew up.

In basic structure, ... Amity is an "Indian" community and not a "West Indian" community. The similarity between Amity and what might perhaps be called a generalized North Indian community structure must certainly be apparent to students of the Indian socio-cultural system. Students of the West Indian scene cannot but be aware that Amity is not "West Indian" in almost any sense but the geographic. (Klass 1961: 239)

Naipaul presents a more complex picture. Early in Mr Biswas's career, he works as a sign painter in the Indo-Trinidadian countryside, and the signs he is commissioned to paint suggest that there is a not negligible influence from mainstream Trinidadian society on the Indians, as well as suggesting a tension between the Indian and non-Indian communities in the island.

He learned to draw bottles, and in preparation for Christmas drew one Santa Claus after another until he had reduced it to a simple design in red, pink, white and black. Work, when it came, came in a rush. In September most shopkeepers said that they wanted no Christmas-sign nonsense that year. By December they had changed their minds, and Mr Biswas worked late into the night doing Santa Clauses and holly and berries and snow-capped letters; the finished signs quickly blistered in the blazing sun. (Naipaul 1984 : 77)

Another recurring theme in Naipaul's book is the concern of the adult Indians for their children's education and career possibilities, and the reader also learns that they are far from indifferent to the annual Carnival a creole cultural institution if anything. As a child, Mr Biswas was sometimes asked to read for his uncle; not the Bhagavad Gita, but "a syndicated American column called That Body of Yours which dealt every day with a different danger to the human body". Naipaul's Indians eat Canadian tinned salmon, smoke Anchor cigarettes, read the Trinidad Guardian, take Dodd's kidney pills, celebrate Christmas, learn poems from Bell's Standard Elocutionist at heart, and quarrel bitterly about Christianity and the Aryan heterodoxy. In fact, the novel presents a mass of ethnography which suggests that the Indians of Chaguanas were quite heavily influenced by West Indian and American culture even before the Second World War.

Sometimes, Naipaul's book can in this way serve as a critical comment on the professional ethnographies. More often, however, Naipaul's rich ethnography is complementary to that presented in the monographs. For example, Klass devotes a major chapter to religion (pp. 137183), while a third of the Niehoffs's book concerns various aspects of religion (pp. 112180). Both monographs note the coexistence of Christianity, Islam and Hinduism among the Indians, and the Niehoffs, in particular, describe important rituals in great detail. Naipaul does not do this, but themes relating to religion do crop up regularly in Biswas; for example, when he remarks that "the doctor came, a Roman Catholic Indian, but much respected by the Tulsis for his manners and the extent of his property" (p. 297). However, his finest contribution to the analysis of religion, which is unparallelled in any of the anthropological monographs, concerns the relationship between Sanatanist Hinduism and Aryanism, that controversial reformist movement which caught on in Trinidad from the first decade of the twentieth century onwards.

As a young man, Mr Biswas is virtually a dependant of his wealthier affines, the Tulsis. A rebellious man, he develops sympathies for Aryanism, which is an unspeakable heresy for the conventionally-minded Tulsis. One of them tells Biswas that if he had his way, "I would cut the balls off all these Aryans" (p. 119), adding, for good measure, that the Aryans seem to "have made some creole converts. Brothers for you, Mohun!" The following paragraph describes the outcome of Mr Biswas's meeting with one of the great Aryan ideologists, Pankaj Rai.

After he had spoken Pankaj Rai distributed copies of his book, Reform the Only Way, and Mr Biswas asked for his to be autographed. Pankaj Rai did more. He wrote Mr Biswas' name as well, describing his as a "dear friend". Below this inscription Mr Biswas wrote: "Presented to Mohun Biswas by his dear friend Pankaj Rai, B A L L B."

He showed book and inscriptions to [his wife] Shama when he got back to Hanuman House.

"Go ahead," Shama said.

"Let me hear what you have against him. You people say you are high-caste. But you think Pankaj would call you that? Let me see. I wonder where Pankaj would place the Big Bull. Ha! With the cows. Make him a cowherd. No. That is a good job." He remembered his own cowherd days. "Better make him a leather-worker, skinning dead animals. Yes, that's it. The Big Bull is a member of the leather-worker caste. (...) Pankaj would say that your mother ain't a Hindu at all! I mean, look at the facts. Marrying off her favourite daughter in a registry office. Sending the two little barbers to a Roman Catholic college. As soon as Pankaj see your mother he would start making the sign of the cross. Roman Catholic, that's what she is!"

"Why don't you shut your mouth?" Shama tried to sound amused, but he could tell that she was getting angry. (Naipaul 1984: 16-17)

Through the first half of the book, Mr Biswas lives more or less as a dependant of his affines in the country; in the second half, he lives in the capital, Port-of-Spain. In moving his protagonist from country to town, Naipaul is able to describe urbanrural relationships in a way unavailable to the Niehoffs and Klass, since their studies were classic anthropological community studies. As some of the aforegoing quotations have indicated, he is also much more concerned with ethnicity and the cultural complexity of a society like Trinidad than the anthropologists were. Both the Niehoffs and Klass admit not having studied inter-ethnic relations systematically; the Niehoffs rely heavily on written sources and statistics in their statements about blacks, while Klass virtually ignores their presence in the more or less immediate neighbourhood of his Indians.7 For a researcher primarily interested in ethnicity and modernization, Naipaul's novel therefore brings a great deal of valuable ethnography which is unavailable elsewhere. Consider the following paragraph, which is one of many highly condensed ethnographic descriptions in Biswas. This tells us about the relationship between town and country, and between East Indian and black.

The other tenants were all Negroes. Mr Biswas had never lived close to people of this race before, and their proximity added to the strangeness, the adventure of being in the city. They differed from country Negroes in accent, dress and manner. Their food had strange meaty smells, and their lives appeared less organized. Women ruled men. Children were disregarded and fed, it seemed, at random; punishments were frequent and brutal, without any of the ritual that accompanied floggings at Hanuman House. Yet the children all had fine physiques, disfigured only by projecting navels, which were invariably uncovered: for the city children wore trousers and exposed their tops, unlike country children, who wore vests and exposed their bottoms. And unlike country children, who were timid, the city children were half beggars, half bullies. (Naipaul 1984: 311)

Reflexivity and levels of reading

Through the story of Mr Biswas, being an East Indian in the West Indies appears a comical and absurd enterprise. The butt of most of the jokes in the book, Mohun Biswas consistently fails to behave in a way acceptable to urban creole society, like Mr Jagabir of A Morning at the Office. Uncomfortably wedged between traditions, he is truly an uprooted and homeless person. And despite his tragicomical appearance, Mr Biswas has been an object of identification for many Indo-Trinidadians up to the present; he was among the first to give their frustrations and confusions a powerful and sensitive verbal form. When my Indo-Trinidadian acquaintance Pete exclaims, in a desperate tone of voice, "My nerves are raw!", he is actually quoting Naipaul. The novel thus has a part to play as an instance in the reflexive monitoring of social identity in Trinidad.8

Like Mittelholzer, Naipaul inadvertently presents the reader with ethnography not of his own contriving. The good-humoured satirical depiction of low-caste people, for example, is never completely absent from the novels written by the young Brahmin Vidiadhar Surajprasad Naipaul up to the early 1960s. In this regard, Biswas and other novels may be read as ethnographic source material, not as ethnographic evidence.

One should also note the explicitly autobiographical character of Naipaul's work, which is absent from the more detached and ethnographically oriented writer Mittelholzer. A House for Mr Biswas was in fact a novel about Naipaul's father, Seepersad Naipaul. In later books, Naipaul cunningly analyses his own society and thereby himself by way of fictional accounts. It is not easy to say in which way these interpretations shed light on West Indian societies in particular, or if their appeal simply lies in their diagnosis of uprooted and homeless individuals everywhere. Some of his novels, notably The Mimic Men (1967) and Guerillas (1975), can be read as theoretical statements about West Indian society. They can to some extent be judged and argued against on such premisses, although it is notoriously difficult to tell where the storytelling stops and the analysis begins. In so far as such a reading seems viable, the border between fiction and anthropology becomes fuzzy: the novel assumes some of the same ambitions as the anthropological analysis, and becomes in part comparable to it. Biswas does not belong to this category of novels. Any theoretical conclusions which may be drawn from a reading of this book would be those of the reader, not of the author. Unlike Klass's and Niehoffs's monographs it is unsystematic and contains no testable or tested hypotheses; it does not attempt to assess the representativity of the sample and does not purport to account for mechanisms of integration in the community.

Although it was not the intention of the author, Biswas has been instrumental in the forging of a genuinely Indo-Trinidadian identity. It has contributed to raising a certain historical consciousness, and in its time, it gave expression and articulation to hitherto muted concerns. The last novel to be considered was written with such aims consciously in mind.

The model of the text: The Dragon Can't Dance

Unlike the earlier authors considered, Earl Lovelace cannot be accused of involuntarily conveying details about his private life and personal perceptions. Akin to some anthropological texts from the same period (or somewhat later), his novel is a thoroughly reflexive and self-conscious contribution to the definition of Trinidadian identity. He has none of that sociological naïveté represented in the two former novelists; he overtly addresses a Trinidadian audience and has consciously constructed the book as a contribution to an ongoing public dialogue about shared concerns in the society.

The Dragon Can't Dance (1979) introduces the residents of a neighbourhood in Laventille, a lower working class area in eastern Port-of-Spain. Many archetypes of Trinidadian folklore are present in Lovelace's book, some of them feared and stereotyped personages. There is the badjohn, a rough black fellow with strong macho ideals and dubious moral character; there is the unsuccessful calypsonian, a middle-aged songwriter with lofty ambitions; there is the unspeakably beautiful carnival princess (every neighbourhood has one); there is the romantic carnival maniac who spends every spare cent and every spare minute on his carnival preparations; there is the shy Indian who never feels at ease in the black neighbourhood, and so on. Lovelace tries to depict them as real persons and to account for their life-worlds; he wishes to replace the clichés with individuals and full-fledged cultural contexts. The badjohn Fisheye, for example, eventually emerges as a reasonable and generous man hiding his admirable integrity behind a fearsome mask.

Although he uses Trinidad English in a much less consistent way than his countryman Sam Selvon, Lovelace uses colloquialisms quite extensively outside of dialogues, clearly in a bid for authenticity and closeness to social reality. Listen to this description of the quarter, where Lovelace poetically mixes the language of the street with journalism and high prose:

This is the hill, Calvary Hill, where the sun set on starvation and rise on potholed roads, thrones for stray dogs that you could play banjo on their rib bones, holding garbage piled high like a cathedral spire, sparkling with flies buzzing like torpedoes; and if you want to pass from your yard to the road you have to be a high-jumper to jump over the gutter full up with dirty water, and hold your nose. Is noise whole day. Laughter is not laughter; it is a groan coming from the bosom of these houses no not houses, shacks that leap out of the red dirt and stone, thin like smoke, fragile like kite paper, balancing on their rickety pillars as broomsticks on the edge of a juggler's nose. (Lovelace 1979: 9)

Widely read in Trinidad, the book is generally recognized as an artistically successful novel. In addition, it is sociologically interesting in that it depicts aspects of respectability and reputation as properties of the class structure, and exemplifies that ambiguous normative structure which Peter Wilson (1978) has spoken of as "crab antics", showing how social mobility can be incompatible with socially embedded values. The text can in this way be read as an anthropological analysis. In the present reading, I shall concentrate on a different aspect of the book, stressing its place in Trinidadian public discourse.

Political themes are present throughout Lovelace's book, and, like a good anthropologist, he depicts politics as a culturally constituted activity and adopts a view from below; from the perspective of the powerless, that is. The humiliation and anger experienced by the proud macho members of steelbands forced to seek commercial sponsorship is a typical example of this. Seen from the context of the urban slum, it appears rational and morally sound that the bandsmen should break their contract with the sponsor. The significance of the annual carnival to the slum residents is occasionally highlighted; this also sheds light on their social condition . Towards the end of the book, a radical political movement led by Fisheye the badjohn carries out an ambitious, but ultimately unsuccessful, plot against the state. Lovelace shows how the many political disappointments, the resentments and the wild hopes of the slum residents act as an explanatory background for their desperate political action. After its appearance, this novel actually had the effect of directing popular sympathy towards the Laventille slum. Its author was also among the first to indicate that several of the semi-official national symbols of Trinidad the calypso, the steelband and the carnival were rooted in the urban lumpenproletariat. A conclusion immanent in this insight is twofold: first, it implies that the middle-class people who read novels and run the country should acknowledge the rough ways of the lower class if not as admirable, at least as acceptable. Second, it could remind the ruling class, or civil society, that a segment of their own people were in a desperate economic and social situation despite the oil boom, which peaked at the time of the novel's publication. All of these effects, which had to a greater or lesser extent been realized, were anticipated by the author.

A lecturer at the University of the West Indies and thus an academic, Lovelace struggles for his street credibility and has consciously searched for the appropriate language for describing the human condition in the backyards of Port-of-Spain. In this, his novelistic project differs from most anthropological ones, which rather try to use a shared theoretical terminology for comparative purposes. His moral vocation, on the other hand, is comparable to that of many anthropologists working in similar environments: he intends to show that the despised dregs of society are competent, intelligent and cultured people who deserve the respect of others.

Indeed, such was exactly the project of Michael Lieber as well, an American anthropologist working among the black working-class of Port-of-Spain in the late 1970s (Lieber 1981). Lieber's ethnography is actually very similar to Lovelace's.

Like Braithwaite, Klass and the Niehoffs before him, Lieber distinguishes himself from the novelist through contextualizing the ongoing flow of life into an analytical framework which render his findings comparable and which can make them fit into a general theory of capitalism and state societies. In his analysis, he stresses in particular the dependent character of Trinidadian economic life and the class structure created by an individualistic ethos and a capitalist economic system. But so does Lovelace! The main difference, a parallel reading of the two works suggests, seems to be that the novelist does not make his position explicit in the same way as a social scientist would have to. Instead, he lets the badjohn turn out as a political hero who intuitively and spontaneously rails against the corruption, injustice and cultural humiliation which has been such a profound concern to many Trinidadians since independence.

Lieber's monograph is not a typical work of anthropology. Indeed, he appropriates some of the techniques from imaginative writing in order to add colour, life and substance to his field of study. He presents detailed portraits of individuals and their activities, fitting them into a vaguely marxist analytical framework intermittently. As he himself explains:

Providing biographical and stylistic vignettes such as those above may seem an unusual mode of presenting ethnographically derived information. But I feel these scenes serve a purpose in conveying something of the concrete quality of flesh-and-blood lives. In attempting to illuminate patterns of social relations in a place such as Port-of-Spain, there is very little "system" or "structure" to speak of (....). We have here a loose and unstructured society (...). (Lieber 1981: 53)

The parallel to Lovelace's novel is even more apparent in some of Lieber's judgements, which are frequently closer to political, or even personal, statements than analytical ones. Both writers wish to describe the humiliation suffered by working-class black men due to their powerlessness. Lovelace undertakes this by inventing a political revolt directed against the sources for the humiliation, while Lieber explicitly defends them against accusations to the effect that they are losers without culture.

Readers may have noticed that I seemed to have been taking sides throughout this book, sneering at the Trinidadian bourgeoisie and its presumptions. I have. The single inescapable fact about the Caribbean is oppression. And it is absolutely clear who have been the oppressors and who have been the oppressed. (...) [B]lack people have responded creatively to their immensely difficult circumstances, articulating perspectives and designing plans to give meaning to their situation and to enhance the flexibility of their lives. (Lieber 1981: 116)

This is exactly the message Lovelace tries to convey as well, and the two books both argue their point convincingly; Lieber through ethnographic cases illuminating a general argument about cultural and economic dependence and counterreactions against it; Lovelace rather through showing what working-class lives in Port-of-Spain look like. In this particular case, the distinction between fiction and anthropology seems a very fine one; the novelist has ambitions in the direction of sociological explanation, and the anthropologist has chosen a highly impressionistic form of presentation in order to retain some of the "flavour" of Port-of-Spain street life. One main difference is that Lovelace's book is much more widely read than Lieber's and thus has had a more substantial direct impact on society. Another difference concerns the degree of explicitness concerning causal relationships. However, supposing that Lieber's marxist-derived explanations were to be discarded, his ethnographic descriptions of Port-of-Spain street life would not necessarily convey more information about Trinidadian society than Lovelace's book. Does this mean that the most relevant difference between fiction and anthropology consists in the explanatory power of the latter? I shall now turn to a discussion of this and related issues, and will try out some general assumptions about the relationship between fiction and anthropology, seen from the perspective of the anthropologistethnographer.

Some implications

The three major novels of Trinidad which have been discussed each exemplify one main way in which fiction can be useful as a source of insight into a society where one is also carrying out fieldwork.

A Morning at the Office can be read as a series of ethnographic statements about ethnic relations at the micro-level in the year 1950, when it was completed. As a micro-sociological enterprise, it is complementary to Braithwaite's (1975 [1953]) study of institutional ethnicity in Trinidad, carried out during the same period. Mittelholzer's persons are probably no more fictional than the protagonists of many anthropological monographs, although the events they take part in were clearly invented. A novel like Mittelholzer's is only credible in so far as it conveys actual features of Trinidadian society. Need we trust it? Of course not, but in its sensitivity to the implicit, to subtle power mechanisms and to processes of over- and undercommunication of ethnicity, the novel enables the anthropologist to embark on fieldwork with a richer pre-understanding (Vorverständnis) than he would otherwise have, and provides him with hypotheses. My own observations indicated both continuity and change in the codification of ethnicity and ethnic relations since around 1950, and Mittelholzer's book is consistent with other sources from that period.

Unlike A Morning at the Office, A House For Mr Biswas has itself become something of an icon in Trinidadian society and has contributed to shaping ideology and reflexivity in that society. It can be seen simultaneously as an ethnographic description of the East Indian community in the first half of this century (and can in this way be a source of historical and ethnographic insight), and as a description of Trinidadian society which has reflexively fed back into the society with which it deals. Although many have never read the book, many have, and it continues to influence the way many individuals think about themselves and their society.

The Dragon Can't Dance combines some of the virtues of the two other books and adds others. Set in the recent past, the book supplies the reader with hypotheses and prepares him or her for the experience of urban Trinidad, as well as bringing ethnographic details from social fields where the anthropologist may for various reasons not be able to take part. In addition, the book is a self-conscious attempt to contribute to defining what and who is an authentic Trinidadian a contribution to nation-building and to the definition of national identity, and an explicit critique of the failure of the then ruling PNM party to help the poor. While A Morning at the Office is mainly an ethnographic statement and A House for Mr Biswas is part ethnography and part an aspect of Trinidadian society, The Dragon Can"t Dance is a contribution to public discourse in that country, as well as having some of the qualities of a sociological analysis.

What use can we then make of such novels? They cannot be used as plain ethnography, since they do not profess to represent the truth and since their relationship to social reality is ultimately uncertain. Besides, if they are to be exploited as ethnographic sources (and not as evidence), the reader must be familiar with the society at the outset of the reading. They cannot, therefore, replace the ethnographic footwork either. It therefore seems a paradox that some of the best anthropological writings extant on Trinidad are works of fiction (cf. Melhuus, infra, for a Mexican parallel). In order to assess their validity, a reader must have first-hand experience of the society. Objectivist ethnography is presumably meaningful without such prior knowledge.

It may have been noted that the aesthetic and artistic qualities of the novels have scarcely been considered in this essay. At certain levels of reading, such qualities are irrelevant, and there are good reasons for not including aesthetic evaluations of texts in assessments of their ethnographic qualities. Now Shakespeare is thought of as being no more typically English than Ibsen is typically Norwegian, and the main point about Cervantes is not that he was Spanish. Likewise, Naipaul is unlike Mittelholzer and Lovelace not considered a typical West Indian author. His topic is no less than the human condition, while the others paint vivid scenes of urban Trinidad and leave the issues there at the particular, the local. It could be said, therefore, that the ethnographic value of a novel is independent of its aesthetic qualities. A poor novel may be just as interesting for its ethnographic raw material as a work of genius.

On the other hand, novels which embody hermeneutic critiques of their authors' societies contain statements which may be comparable to anthropological statements. As these readings of three Trinidadian novels show, there are three levels of reading which are immediately relevant to ethnographic endeavour.

First, novels may serve as ethnographic sources and may to this effect rank with informants' statements. At this level, the author whether he is a Mittelholzer or a Naipaul more or less unwittingly reveals aspects of his society. As Bakhtin and many others have reminded us, the author is a prisoner of his own time. The author, known through the novel, is here seen as the production of a society.

Second, novels may be read as ethnographic descriptions; that is, the information conveyed may be taken more or less at its face value, as a kind of ethnographic documentation.

Third, some novels may profitably be read as theoretical anthropology. These books embody a cultural analysis and reflexive critique of the author's society. The most outstanding West Indian example known to me is Naipaul's The Mimic Men (Naipaul 1967), which is a devastating and controversial diagnosis of the inhibiting doxic structures presumably guiding the uprooted Caribbean peoples in their lives. The author's perspective here can sensibly be dealt with in a theoretical way, but it cannot be argued against in so far as the text is a novel.

Novels also form part of reflexive socio-cultural reality and to this effect are part and parcel of that society within which they were written. Lovelace's novel, in particular, had the explicit aim of being read by many Trinidadians in order that it might contribute to the definition of Trinidadian national identity and politics. In considering this aspect of fiction, we enter the sociology of literature, where texts are seen as the products of society and where the relevant readings of these texts will be those of the members of that society, not our own. Anthropological studies of societies where reading is widespread should not ignore the direct cultural and social effects of texts or of other forms of "cultural consumption", for that matter.

Anthropology and fiction

The respective relationships between the text and social reality postulated by anthropology and fiction differ, although one cannot offhand say that one of the genres represents society better than the other. An anthropologically relevant difference is that one cannot argue with a novel in the same way as one can argue with an anthropological work. The novelist can always retort that he or she has made everything up, and his or her analysis is rarely unambiguous and fully explicated.

We may also wish to claim that anthropology is fundamentally different from fiction because of its comparative dimension. However, fiction is also comparative. Aldous Huxley (1931) once wrote of an English author who went to the West Indies in order to collect material for a novel about people in Mayfair. In good fiction, as in good anthropology, there is an underlying assumption of something universally human. On the other hand, anthropology is unique in its specification of dimensions for comparison and its standards for ethnographic descriptions. Are such dimensions and standards straitjackets? If one thinks so, one might turn to fiction for consolation. In any case, it could be argued that fiction and anthropology tend to represent the social world in radically different ways.

Direct introspection is deemed unacceptable in social anthropology. If one ventures to consider the inner states of persons, one must always refer to acts or statements as evidence. Within the psychological realism represented in the three novels, introspection is an important literary technique.

The styles of writing differ for similar reasons. Although it may be argued that anthropologists, like novelists, often write in a persuading and seductive style, it is our aim to avoid it. In this, our profession differs strikingly from that of novelists. Comparing Braithwaite's sociological study with Mittelholzer's novel, one might well discover that they are similar in the use of evidence, but whereas Mittelholzer is concerned with introspection and the portrayal of particular characters, Braithwaite tries to substantiate his often impressionistic claims by referring to statistics and historical facts. Anthropology and fiction represent different, although sometimes overlapping and frequently complementary reductions of social reality.

Works of fiction are not anthropological writings and vice versa, although both genres can be read as though they represented the other genre. For example, there are anthropologists who, no matter how sloppy their theorizing may be, are praised for the literary quality of their writing. There are also anthropologists who combine the systematic and descriptive approach of anthropology with the poetic and evocative approach of fiction; good examples would be some of Michel Leiris's and Roger Caillois's books, or indeed parts of Michael Lieber's Street Scenes. In a stimulating discussion of a fictive ethnography, Needham (1985) has argued that there seem to be no sound criteria for distinguishing the false from the true provided the only available source is a professed ethnographic text; but criteria of comparability, consistency, accuracy and comprehensiveness could all be invoked in distinguishing between the approaches even if both are applied in the same work. On the other hand, it cannot be taken for granted that the ethnographic text is necessarily the more faithful and perceptive depiction of relationships within a society. Few ethnographers would be likely to claim that their description of Dublin is ethnographically more accurate than Joyce's descriptions.

The mainstream novelist and the anthropologist have in common their ambition to transform the world of sensations and thoughts into one of words. Kurt Vonnegut's description of the typical novelist as someone who feels that the world is a chaotic place, but who is nevertheless determined to impose order onto a heap of white sheets (Vonnegut 1982), could perhaps be extended to include the typical anthropologist. Fiction and anthropology, as modes of enquiry, therefore differ perhaps not so much in posing different questions, but in their approaches. It is scarcely true that every anthropologist is a failed novelist, as Leach once alleged, but each genre has qualities lacking in the other, although they sometimes overlap. What fiction gains from its vividness, freedom to experiment and evocative techniques, it loses in its lack of accuracy, empirical comprehensiveness and attempt to establish interesting comparative dimensions. Strathern (1987) argues that feminism and anthropology are neighbours in a "mocking relationship" and that they are ultimately irrelevant to each other because of their different normative positions. In the case of fiction and anthropology, the relationship, admittedly more distant in most cases, ought rather to be one of mutual challenge. They are relevant to each other, but they can never be the same thing. The criteria for evaluation and the internal rules of the genres differ, and yet as I hope to have shown they remain relevant to each other, particularly if we are aware that they are not "the same thing".

Feyerabend (1987) once suggested that the best translation of a novel by Dostoyevsky would be one by Dickens. In line with this extremely relativistic view on translation, it might be claimed that the poetry and sensible qualities of a way of life are lost through cultural translation. Maybe the conscious use of fictional texts may help build a bridge between the richness of experience and the sterility of the academic anthropological text, and maybe such texts may be useful bridgeheads in translation, but precisely since they are poetical in nature, they may be seductive and misleading through persuading their reader of the validity of a certain set of prejudices. Besides, novels should never serve as examples to be followed by writers acting in their capacity as anthropologists. Fiction certainly merits to be read as though it were something else than ethnography or inadequate attempts at anthropological theorizing, although I have argued, in this essay, that novels may have ethnographic qualities in addition to everything else. Conversely, anthropologists should probably resist the temptation to indulge in the rich and evocative language of creative writing. If ethnography were to adopt the literary ambitions of fiction, it would be but a short step from the genre of travel writing (cf. Louch 1966). In that case, nothing would ultimately be gained from the appropriation of fictional texts dealing with the societies we study; instead, anthropology would cease being an academic discipline, and its practitioners would inadvertently confirm Leach's darkest suspicions.


Abrahams, Roger D. (1983) The Man-of-Words in the West Indies. Baltimore: Johns Hopkins University Press.
Anderson, Benedict (1983) Imagined Communities. London: Verso.
Bailey, F.G. (1991) The Prevalence of Deceit. Ithaca: Cornell University Press.
Bloch, Maurice (1991) "Language, Anthropology and Cognitive Science". Man, 26: 18398.
Bourdieu, Pierre (1977) Outline of a Theory of Practice. Cambridge: Cambridge University Press.
Braithwaite, Lloyd (1975 [1953]) Social Stratification in Trinidad. Mona, Jamaica: Institute of Social and Economic Research (ISER).
Clifford, James (1986) "On Ethnographic Self-Fashioning: Conrad and Malinowski", in Reconstructing Individualism, T.C. Heller et al., pp. 14062. Stanford: Stanford University Press.
Clifford, James and George Marcus, eds. (1986) Writing Culture. The Poetics and Politics of Ethnograpy. Berkeley: University of California Press.
Ellmann, Richard (1983 [1959]) James Joyce. Oxford: Oxford University Press.
Feyerabend, Paul K. (1987) Farewell to Reason. London: Verso.
Geertz, Clifford (1988) Works and Lives. The Anthropologist as Author. Cambridge: Polity.
Gellner, Ernest (1983) Nations and Nationalism. Oxford: Blackwell.
Giddens, Anthony (1990) The Consequences of Modernity. Cambridge: Polity.
Giddens, Anthony (1991) Modernity and Self-Identity. Cambridge: Polity.
Goody, Jack (1986) The Interface Between the Oral and the Written. Cambridge: Cambridge University Press.
Herskovits, Melville and Frances Herskovits (1947) Trinidad Village. New York: Alfred A. Knopf.
Holy, Ladislav and Michael Stuchlik (1983) Actions, Norms and Representations. Cambridge: Cambridge University Press.
Huxley, Aldous (1931) Music at Night. London: Chatto & Windus.
Klass, Morton (1961) East Indians in Trinidad. New York: Columbia University Press.
Lieber, Michael (1981) Street Scenes: Afro-American Culture in Urban Trinidad. Cambridge, MA: Schenkman.
Louch, A.R. (1966) Explanation and Human Action. Oxford: Blackwell.
Lovelace, Earl (1979) The Dragon Can't Dance. London: Longman.
Manganaro, Marc, ed. (1990) Modernist Anthropology. From Fieldwork to Text. Princeton: Princeton University Press.
Miller, Daniel (1993) Modernity &shyp; And Ethnographic Approach. Oxford: Berg.
Mittelholzer, Edgar (1979 [1950]) A Morning At The Office. London: Heinemann.
Naipaul, V.S. (1961) A House For Mr Biswas. London: André Deutsch.
Naipaul, V.S. (1967) The Mimic Men. London: André Deutsch.
Naipaul, V.S. (1975) Guerillas. London: André Deutsch.
Needham, Rodney (1985) Exemplars. Berkeley: University of California Press.
Niehoff, Arthur and Juanita Niehoff (1960) East Indians in the West Indies. Milwaukee: Olsen Publishing Co.
Smith, M.G. (1984) Culture, Race and Class in the Commonwealth Caribbean. Mona, Jamaica: Institute of Social and Economic Research (ISER).
Stewart, John O. (1989) Drinkers, Drummers and Decent Folk. Ethnographic Narratives of Village Trinidad. New York: SUNY Press.
Strathern, Marilyn (1987) "An Awkward Relationship: The Case of Feminism and Anthropology". SIGNS, 12(2): 27692.
Vonnegut, Kurt (1982) Palm Sunday. London: Granada.
Wilson, Peter J. (1978) Crab Antics (2nd edition). New Haven: Yale University Press.


An earlier version of this article was presented at the conference "The Multiplicity of Writing and Social Anthropology", Department and Museum of Anthropology, Oslo, 17-19 October, 1991. I would like to thank the participants at the conference, and particularly Eduardo Archetti, for useful critical comments.

1. Joyce nevertheless claimed that he was unable to make anything up (Ellmann 1983). And surely, there are good reasons for seriously doubting the "ontological" validity of anthropological texts. These problems do not require a detailed treatment presently; see Bailey (1991) for a provocative and stimulating discussion of these issues in contemporary anthropology.

2. Archetti (infra) argues that "second-rate authors" could become prolific and popular composers of tangos in Argentina. While this may also be true of Trinidadian calypsonians, few of them would regard themselves as second-rate in any reasonable sense of the word. The calypso, the dominant literary form of the island, has had a strong and continuous impact on Trinidadian civil society since the turn of the century.

3. The ethnic labels used in the text correspond to labels used locally. The most important ones are white, black (or Afro-Trinidadian, or African), coloured (or brown, or "red"), East Indian and Chinese. The term "East Indian" is used throughout the Caribbean to distinguish people of Indian origin from Amerindians. The word "Creole" usually means, in Trinidad, "Trinidadian but not of East Indian origin".

4. It might be added that Lloyd Braithwaite belongs to the same social stratum as Edgar Mittelholzer did: the coloured middle class. In one of his last publications, the late M.G. Smith (1984) dismissed Braithwaite, in a surprisingly rash way, as an apologist for the then (early 1950s) emerging "mulatto hegemony". In fact, he turns Professor Braithwaite into ethnographic source material, using causal explanations to account for his opponent's theoretical position.

5. All page references to Biswas refer to the 1984 Penguin edition.

6. Biswas admittedly deals largely with an earlier period, ending with Mohun Biswas's death in the early 1950s. This is not important for the present argument.

7. Towards the end of his monograph, Klass brings an interesting comparison between his village, "Amity", and a black village in Toco, north-eastern Trinidad, studied by Herskovits (1947), but this comparison is not intended to shed light on processes of cultural change; on the contrary, it presupposes the relative isolation of the communities from each other.

8. Another, related aspect of Naipaul's role consists in his reputation as a world-famous Trinidadian. His literary world fame is many times magnified at home. An assessment of the effects of this factor is not necessarily conditional on reading his work, just as it may not be necessary to study the Gita in order to study Hindus. The fact of Naipaul's reputation may nevertheless be sociologically interesting in its own right.


Dealing with a society not encompassed by the standard anthropological clichés about modern vs. traditional societies, this paper highlights methodological and epistemological problems faced by an anthropologist carrying out fieldwork in a society where creative writing is only one of many aspects of sociological reflexivity widespread among its inhabitants.

The setting is Trinidad, which is a society lacking a preliterate history. It shares with our own northern European societies a high level of literacy and widespread popular reflection, as well as discourse, about society. A telling anecdote is this. As the anthropologist once approached two young men at a public bar, telling them that he was curious to find out about Trinidadian culture and society, one of the men retorted, "So what's your project, brother? You wanna do an M.G. Smith on us?" (M.G. Smith is a Jamaican anthropologist writing on "the plural societies of the West Indies".)

The anthropologist studying knowledge and ideology in Trinidad must therefore account for the quasi-theoretical constructs of the agents themselves. In a similar way, he must come to terms with the texts created by Trinidadians for Trinidadians: In which sense do they form part of society, and how should they be read as ethnographic documentation?

Using as examples a novel by Edgar Mittelholzer from 1950, one by V.S. Naipaul from 1961, and a contemporary novel by Earl Lovelace, the paper critically discusses the possible uses of fictional accounts as ethnographic sources in anthropological research. Fictional texts are compared to observed interaction and to other written accounts (such as extant anthropological texts). The paper concludes that fiction may serve as ethnography in its own right and as a source of hypotheses, and that it should also be considered a part of reflexive socio-cultural reality. Finally, some criteria for distinguishing fiction from anthropology are suggested.