[In 1993 I published a volume of translations from the work of one of the great literary critics/social scientists of the Twentieth Century, the Brazilian writer Antonio Candido, little known outside of Brazil (Candido 1993). This essay is the introduction I wrote for that volume, which accounts for the references to essays that were contained in it.]

Introduction To Antonio Candido on Literature and Society

The opposition of "art and society" is one of the great commonplaces of social theory. (See Zollberg 1990 and Hennion 1993 for good summaries of the state of this debate.) On the one hand, works of art (often, perhaps usually, of literature, but encompassing music, visual art, drama, dance, and the rest) with all their internal complexities, mysteries, and aesthetic qualities. On the other, the society in which these works came into existence and in which people read and respond to them. Only extreme formalists expect to understand works of art fully without referring to the organized social context they exist in. Only extreme sociologizers expect to understand those works completely by analyzing the conditions of their existence.

But: how to combine a sociological sense of the complex worlds in which artists create works and audiences respond to them with a full aesthetic appreciation, of the kind ordinarily provided by critics, of the works themselves?

Sociologists have considered these problems since the beginning of the discipline. The great names of the field—from Durkheim, Weber, and others of the founders to Pierre Bourdieu—have all weighed in with theories and research. Literary theorists from Taine, through the Frankfurt School, to the wide variety of critical theorists writing today have contributed their solutions. Many such discussions deal only with large—scale relations between the two realms of art and society considered in the abstract, not attempting the analysis of specific works of art. The ensuing heated, complex, arcane discussions have seldom produced much light: writers are usually more interested in proving that "their variable" is the important one than in understanding the works-in-context, or they get so involved in the detailed analysis of specific works that they lose sight of the general connections.

Antonio Candido (pronounced KHAN-dji-doo), for many years one of Brazil’s leading men of letters, has, in  books and essays published over the last fifty years, proposed effective ways of dealing with this classical problem  and has provided exemplars of those methods at work. He combines detailed literary analysis with profoundly sociological insights. His writing has a quality, a profundity, a grace, and a penetration which, combined with the originality of his thought, ought long ago to have led to his recognition in the world outside Brazil. But, since Antonio Candido writes in Portuguese, and his work has never appeared in English or French or German, and in Spanish only in 1991 (Candido 1991), it has never been read by the people who would recognize its worth had they seen it. The essays presented here will make his contribution more widely available.

This brief introduction discusses the problem of  "literature and society"; describes Candido's perspective on it, and especially the way he combines the analysis of works of literature with  an understanding of the nature of society; tells something about Brazil, and about Antonio Candido's life and work; and, finally, takes up the question of his Brazilianness, how being a writer and thinker in a developing country appears in his writing.

The art-society problem

 We can see the strength of Candido's approach  by comparing it to common practice among those sociologists of art and culture, and literary critics and analysts, who are interested in the idea that art somehow is influenced by or connected to the society it is produced in. (See the summary in Hennion 1993.) Such analysts typically assert such a connection, then demonstrate it by pointing to congruences between specific works of art, specific aspects of those works, and specific aspects of the society in which the works were made.

The tricky part of all such formulations is the verb used to describe the connection, the term which specifies how the society actually does something to the work or how the work incorporates whatever it gets from the society. If we ask how the congruences between society and work the analyst has discovered have come about, the usual answer is that the latter reflects the former. "Reflection" envisions the literary work as a mirror, whose surface picks up and sends back whatever is placed before it.  But this vague metaphor simply asserts that there is a connection, leaving the specific mechanisms unexplored.

Analysts, recognizing this weakness, have proposed many ingenious substitutes for "reflect": "influence," or "be congruent with," or "resonate with," among others. The connection is often asserted visually, by drawing arrows between the words "Art" and "Society," the arrows serving as a still more vacuous description of what is going on. (Michael Lynch 1991 provides a telling critique of this sort of thing.) Neither the verbs nor the pictures do the job. Each evades giving a detailed analysis of the process by which the world shows up in the work, and is just a way (in this reproducing a common fault of scholarly writing) of alluding to something without having to take the intellectual responsibility for making an assertion so concrete that investigation could prove it wrong. The vague words reveal the simplicity and inadequacy of the sociology.

Some writers in this mode spend most of their time on the work itself. Their kit of analytic tricks makes possible subtle analyses of the contents of literary works, the excavation of social references and influences in the details of plot, characterization, and literary style. But these writers seldom have a similarly detailed understanding of the social context of the work, nor an adequately detailed theoretical framework for creating such an understanding. They have a sophisticated literary theory, but a crude, homemade sociology.

Practitioners in this style, having alleged that such a non-specific connection as a "reflection" exists, then make an analysis which, as I've suggested, is no more than an ingenious searching out of congruences between social phenomena and details or elements in the work under analysis, a demonstration, for instance, that the work of literature being analyzed deals with themes characteristic of the period and place in which it was written. The congruences do not demonstrate any actual connection, only similarities, and their number is limited only by the cleverness of the analyst.

Another version of this "art and society" approach, favored by sociologists who want to make use of the insights into and data about society to be found in fiction, is to treat the work as a kind of reportage. The work reflects society as a journalist's account does. Since many authors have had just such intentions, many literary works lend themselves to being taken as such reportage, something sociologists would do, if only they wrote that well and were as imaginative. Social analysis in fictional form also allows the introduction of the detail of case studies in a way that seems more difficult in social science texts (although it has been done, e.g., in W. Lloyd Warner's Yankee City books (Warner et. al. 1941-1959) as well as in the detailed accounts of such ethnographies as Street Corner Society (Whyte 1955) and Tally's Corner (Liebow 1967) .

So:  Stendhal's The Red and the Black becomes a report on social class and mobility in France of the period, James Farrell's Studs Lonigan a report on the Irish community of Chicago in the Twenties, and so on. This becomes more interesting, but simultaneously far more speculative, when, say, Kafka's The Trial is read as another version of Max Weber's description of the "iron cage" of modern bureaucratized societies. In either case, the interest of the results depends on the knowledge of social reality which went into the work's construction, and the ingenuity of the fictionalization. Like reportage, the more detailed the description, as in Farrell's description of Chicago's Irish community, the greater the necessity for point-for-point accuracy in the translation to fiction. Unlike reportage, such works can evoke a generalized mood which need not be accurate in the details, as Kafka's book is not an accurate description of a specific bureaucracy, only of the the essence of the phenomenon. (Since Weber's descriptions were "ideal types" rather than empirical reports, the comparison raises an interesting question about the connection between two such generalizing strategies.)

For many sociologists and sociologically minded literary historians, the alternative to a search for congruences between the work and the society has been to see art works as the products of networks of cooperation among all the people whose actions are necessary to the final product being whatever it is, and to see further how being made in such a network shapes works. (See the discussion in Becker 1982.)  Hennion (1993)calls this a "repeopling" of the worlds of art, a bringing back of all the intermediaries involved in the making of art works, and notes that this leads to the analyst applying all the conceptual resources of a sociology of work and institutions to worlds of artistic activity. (There is a substantial overlap between this point of view and Bruno Latour's analyses  of scientific activity as the operations of networks of a variety of actors, e.g. in Latour 1987.)

The strategy is exemplified in Sutherland's analysis of the way writing for serial publication led, in Victorian England, to the episodic plots of Thackeray and the early Dickens—authors did not bother to develop complex plots for novels they would not finish if sales were not sufficient–and a wordy style, understandable if you are paid by the word. (Sutherland 1976)

Leading to a complex understanding of how literary works acquire their characteristic features, this is an order of fact to which Antonio Candido often appeals (as in certain aspects of his analysis of Dumas' Count of Monte Cristo). But such analyses are usually not accompanied, and certainly not in the work of sociologists, by any deep analysis of the literary text itself. This is the place to confess that I am a chief offender in this respect (see, for example, the analyses in ).

Sociologists usually lack the command of languages and the breadth of literary and artistic experience needed to do justice to works of fiction and, if they have those, usually do not have the (not tremendously arcane) tools of literary analysis. So we learn from them about the conditions of production but not much about the way authors work with available materials under those conditions to produce specific works. The results are long on general features of the work: we learn that authors write lengthy, wordy novels in three volumes because fiction at the time was mainly distributed through lending libraries which relied on books so composed to maximize rental income. But such analyses are typically short on analysis of the details which give works their distinction and unique qualities. Why did Emile Zola emphasize stairways so much in his accounts of working class Parisian life? And here is where Antonio Candido provides what is generally missing (the example is from Candido 1993, pp. 55-94): because the image of the stairway allowed Zola to use a new fact of urban France, the presence of tall tenements in which such workers lived, to embody the precariousness of social ascent and descent characteristic of workers' lives.

This example exemplifies the distinctive analytic stance Candido brings to the debate, his insistence on the importance of the details of the work's structure as well as the details of the structure of the society for understanding the relations between the two.  For Candido, literature and sociology are not two different fields of study. The chief thread in his scholarly life has been "the search for an open and integrative mode [of analysis or criticism] which rises above [conventional academic divisions between social science and the humanities] in order to arrive at a more coherent point of view." (Velho and Leite 1993, p. 28).

Academic sociology and Marxism accentuated my tendency to study the social and ideological aspects of literature, and I created some mixtures which constituted my way of attempting the greatest possible amplitude and avoiding dogmatisms, without losing my guideline. I found this guideline above all in the notion of structure, linked to those of process and of montage, as terms of an integrated vision. (Velho and Leite 1993, pp. 38-9)

So he is alert to the infrastructure of literature: the numbers and distribution of readers, and the results for the productive apparatus of publishing, as in his essay (in this volume) on literature and underdevelopment. But he is equally alert to the more subtle relations between social practice and literary work, as when he remarks ironically, of The Count of Monte Cristo, that the bourgeoisie could have nourished their children with massive doses of the book:

Not as furtive or marginal readings, soon transformed into a pretext for playing; but also, and above all, as assigned readings, analyzed by the teacher, as a major part of the school program, to the end of transfusing them into the thought and action of every hour. (Candido 1993, p. 10)

What Candido writes, then, is not just critical analysis, and not just sociology, not just general speculations on the connections between abstractions, but a combination which finds the operation of society in the way the work is put together, a combination heavily influenced by social science but focused (in the anthropological rather than the sociological style) on specific cases.

The anthropological idea of culture, implying the ideas of totality and organicity, influenced my way of analyzing literary works. . . .  [O]ne cannot, for example, do a comparative literary analysis by taking (let's say) the function of money in Machado de Assis, in Dostoyevsky and in Balzac, and making a comparison pure and simple. It is necessary to consider the work of Machado as a whole to see how money functions in it. It will certainly function differently than in Dostoyevsky and Balzac, seen as totalities in which it is inserted. . . . I believe that anthropology attracted me to the degree that it had links with the literary attitude and allowed me to better satisfy my taste for the concrete. The anthropologist relies more on intuition, on empathy, on personal experience, and can do individualized case studies. (Velho and Leite 1993, p. 34).

To sum up Candido's approach in a formula (a dangerous effort in the case of so subtle a thinker), we might say that authors use the material of social observation and analysis as the basis of the structure of a work more than of its content. The most successful works, artistically, are those in which the form exemplifies the nature of the social phenomenon which furnishes the matter of the fiction. Not "reflection" or "congruence," but the active effort of an author to create a form which successfully embodies a social analysis or understanding. Not judgments based on a fiction's success in reporting facts which support an author's social theories, but judgments of how well the author has created a form which augments and complements our understanding of the subject matter. He gives a memorable example in his analysis of the fragmented narrative technique of Conrad's Lord Jim:  ". . . the effectiveness of Conrad's art is not due to the simple proposal of an attitude of life, but to the fact of translating that attitude into a method of narration, which becomes an indissoluble part of what the novelist means since, in the end, it is what he effectively says."

Another example. In his analysis of the Memorias de um sargento de milícias, an early 19th century Brazilian novel (Candido 1993, pp. 79-103),  he identifies the duplicity of Brazilian society of the time with the constant movement of the book's characters from high to low. That movement, which is the structure of the book, emphasizes the interdependence of upper and lower strata of a stratified society, of the respectably moral and the disreputably poor and deviant, of the highest human aspirations and the lowest conduct. (This will recall, for those acquainted with the so-called "Chicago School" of sociology, Everett Hughes' discussions of the dependence of "good" people on others who will do their dirty work (Hughes 1984, pp. 87-97), a theme Candido takes up in several essays in this volume.)

Brazil and its intellectual tradition in the human disciplines

Antonio Candido, having made his career in Brazil, writes and thinks in a way that embodies much that is Brazilian, so a little background is in order for those not familiar with that country. Brazil is larger than most North Americans and Europeans realize.  Covering more than three million square miles, it has a population of about 140 million, according to 1991 estimates, and is as large as the rest of South America combined.  But it is less known elsewhere than countries like Mexico or Argentina, almost surely because its national language is Portuguese rather than, as in the rest of South America, Spanish.

Spanish, like French, German, and English, is one of the several languages that can be called a world language, one that is understood beyond the borders of its home country. Scholars who work in such a language address each other in it and expect (sometimes quite unrealistically) scholars everywhere else to be able to read their work. Portuguese, on the other hand, like Dutch or Danish, is not a world language. Scholars who work in such languages may write for each other in them, but when they address the world they do so in one of the world languages; they know that "no one" reads their native tongue.  Being relatively small, those countries cannot support a large scholarly output in their own language.  But Brazil, a very large country, has an educated class large enough to support publication and diffusion of knowledge in the national language and so, though Brazilian scholars are typically at home in at least French and English, they do not produce their major works in those languages.  Scholars in the major world languages seldom read Portuguese, so a vicious circle creates an ignorance elsewhere of Brazil's sizeable, varied, and interesting intellectual output. (My speculation about the reasons for this situation may be arguable, but the fact is not.)

As a result, Brazilian scholarship, while fully cognizant of and embedded in the traditions of Western thought, maintained, at least until the 1930s, a strong tradition of scholarship in the humanities and social sciences as "letters."As Candido explains (Candido 1965, p. 130):

The powerful magnet of literature interfered with the sociological tendency, giving birth to that mixed genre of essay, constructed in the confluence of history with economics, philosophy, or art, which is the most Brazilian form of investigation and discovery in Brazil . . . . It is no exaggeration to say that this kind of essay—which combines more or less felicitously imagination and observation, science and art—constitutes the most characteristic and original feature of our thought.

When we look back at the early stages of modern, professional American sociology, we see works of sociological research: The Polish Peasant in Europe and America, and the classic monographs produced at the University of Chicago under the influence of Robert E. Park.  When Brazilian scholars look back, they see works whose virtues are as much literary as scholarly and scientific, exemplified (to mention books known outside Brazil) by Gilberto Freire's massive trilogy on Brazilian social history (1986) and Euclides da Cunha's Rebellion in the Backlands (1944) or (to take an example not known abroad) Sergio Buarque de Holanda's Raízes do Brasil (1956, also see Candido 1985, pp. 109—38).

While academic disciplinary lines exist in Brazil, they are fuzzier and more permeable than North Americans and Europeans are accustomed to.  The history of the social science disciplines, and their relations to  government and academic institutions, are different, as are Brazilian traditions of intellectual life and production. (More detail is provided in Becker 1992.) Writers respond to the specific constraints of Brazilian intellectual and academic life, and draw on its unique traditions. As a result, literary ideas, and standards, permeate Brazilian social science in a way no longer imaginable in American sociology.

Antonio Candido's work, resisting the contemporary move toward disciplinary specialization in Brazil, belongs to this tradition and continues to mix, as Candido says (Velho and Leite1965, p. 30) of this tradition, "Imagination and observation, science and art." It cannot, thus, be read as if it were a sociology of literature or literary theory or any of the specialized genres that now characterize Western scholarship in these fields. Candido does not construct arguments, in the modern professional style; he explores a work, seizes on its details, places the work in the context of the readers and literary world of its day, all to the end of achieving a deeper understanding of the mutual connections of work and setting.

Some biographical notes

Antonio Candido is a man of letters in an older, inclusive sense, not only a critic of specific literary works and a writer on more general literary matters, but someone who writes analytically about his society and others, about historical and political questions, often basing his social analyses on literary materials.

His childhood gave him some incredible gifts for a writer on social and literary topics.

I am the child of a studious physician, dedicated to his profession, but with strong interests in philosophy, history, and literature, and of a mother who had only the brief instruction in the nuns' schools of that time, but who was very intelligent and cultivated. . . . It was a house full of books, many with beautiful bindings, and I grew up among them. . . . One nice day, when I was about nine years old, my brother around seven and the youngest six, my father gave us a two volume Larousse Universel, saying: "Play with this." And we began to play, to look at the colored plates with maps, uniforms, mammals, reptiles, butterflies, fish, etc. Since I wanted to understand, I exerted myself, questioning my mother and gaining, between mistakes and lucky guesses, a growing understanding. After spending a year coloring the wigs of historical characters, putting mustaches on Roman emperors, a goatee on Louis XIV, and similar things, we had acquired some familiarity with the entries and learned a little French, reinforced by mother's Berlitz lessons. . . . (Velho and Leite 1993, p. 30)

I have always been an indiscriminate devourer of books, from the age of nine until my sixties, and I believe that I thus accumulated much information, not least because I read compendia, dictionaries, and encyclopedias. (p. 40)

But Candido didn't go to school, at least not for very long; instead, his mother tutored the children at home. When he was eleven, his parents took him to Paris for "a year which was decisive for my cultural development." The children's French teacher, in addition to giving them lessons at home, took them on Saturdays to "museums, churches, institutions or, sometimes, to the Comédie Française, explaining the plays to us beforehand, in a way that was accessible to children of our age. . . . With this strong saturation and our stay in France, I became more or less bilingual." (P. 30)

Back home, they were taught English and introduced to English literature by a teacher trained in a North American institution in Brazil. They were also in close touch with an Italian friend of their mother, who spoke Italian and sang opera with them, and who introduced Candido to the ideas of socialism.

He was trained as a scholar when the "French mission," a group of French scholars (of whom Claude Levi-Strauss is the most well-known, but others of whom—especially Roger Bastide and Jean Maugüe—had a greater influence on him), was organizing the curriculum and teaching the first classes at the then almost new University of São Paulo. Brazil had traditionally turned to Paris as a cultural center, and modern universities, when they were finally created in the Thirties, got a jump start from the sizable number of French scholars who were sent to bring French civilization to such outposts. Candido was part of the first generation of professionally trained social scientists in Brazil, among whom the outstanding name in sociology is Florestan Fernandes.

The classic British anthropology of Malinowski, Radcliffe-Brown, Evans-Pritchard, and others always inspired Candido more than sociology. Mariza Peirano (1990. P. 43), an anthropologist who has written about Antonio Candido's work in social science, comments that "American sociology, for example, seemed to him totally uninspiring [he used the English word]: 'Social surveys, the atrophied sociology of American universities.'" In the end, he found sociology boring—"If I study a primitive culture, I end up preoccupied with the human problem of the person who is before me. How he walks, how he sings, how he dances, how he sees the world. At the other extreme, that of sociology, I don't see any person at all. I see that 7,283 people use Kolynos toothpaste."

So he changed fields, teaching sociology for almost 15 years but eventually becoming a professor of comparative literature. He now insists that he is not a sociologist—"If you were to ask me what I am, principally, I would say, italicizing it, 'teacher'" (Velho and Leite 1993)—but is known still for his dissertation, a major sociological investigation of a rural community in the state of São Paulo (Candido 1964). During his years as an academic sociologist, he always wrote literary criticism, as well as publishing his classic work, The Development of Brazilian Literature (Candido 1959).

I taught sociology without being a sociologist, I had no training in literature and taught literature, I had a close call with medicine, I studied law without using what I learned, I read a little of everything with no method. You can understand why I consider myself someone with irregular and heterodox training, but, modesty aside, productive. (Velho and Leite 1993, p. 35)

Though Candido is serious about his renunciation of sociology as an academic field, we can't ignore the connections to sociology, broadly conceived, in his work. His sociological thoughts are not the truisms so often uttered by humanists with only a vague understanding of the social sciences, but deeply sociologically insights into social organizations, processes, and practices, as we see in a comment on the situation of Latin American writers:

 . . . the problem of publics presents distinctive features, since it is the only group of underdeveloped countries whose people speak European languages (with the exception, already noted, of the indigenous groups) and have their origins in countries which today still have underdeveloped areas themselves (Spain and Portugal). In these ancient mother countries literature was, and continues to be, a good of restricted consumption, in comparison with the fully developed countries, where publics can be classified according to the kind of reading they do, and such a classification permits comparisons with the stratification of the entire society. But, as much in Spain and Portugal as in our own countries, there is a basic negative condition, the number of literates, that is, those who could eventually constitute the readers of works. This circumstance brings the Latin American countries nearer to the actual conditions of the old mother countries than are, in relation to theirs, the underdeveloped countries of Africa and Asia, which  speak different languages than those of the colonizers and confront the grave problem of choosing the language in which to display literary creation. African writers in European languages (French, like Léopold Sendar Senghor, or English, like Chinua Achebe) are doubly separated from their potential publics; and are tied either to metropolitan publics, distant in every sense, or to an incredibly reduced local public. (Candido 1995, p. 123)

In this analysis, Candido combines several fundamental social facts to outline the situation of Latin American writers. They live and work in underdeveloped countries. But they write in the language of a mother country which is "their own," one in which literature is, as it is where they live, a good of restricted consumption, relatively low rates of literacy leading to small, class-restricted audiences. Embedded in this network of class, linguistic, and national connections, their problem of finding readers, topics, and language differs from that of their counterparts in Africa and Asia.

Throughout his long career as teacher and writer, Candido has also been an active participant in the political and cultural life of his country. A lifelong socialist, he has been concerned with political and social issues. The concern has been practical, as in his active participation in the founding of the PT [Partido dos Trabalhadores], the Brazilian Workers Party. He was a "bad" militant, he says, because he had no taste for politics, but was nevertheless very active, in his youth, in the former Socialist Party, working in elections and giving workers' classes.

Marxism was always a strong influence for me, and I believe that the dialectic march is visible in my critical work. Nevertheless, I was not a Marxist properly speaking. . . .

Marxism is a totalizing philosophy, and to be a Marxist is more or less like being a Catholic: you have to pass all the reality of the world, of being, of action, through the sieve of doctrine. Now beyond the general suggestion of method . . . Marxism was important for me above all in the field of politics. I think, for example, that without concepts like the class struggle one cannot understand social reality correctly. But I was soon convinced that it was not a closed doctrine, but rather an instrument of great analytic and practical power, which must always be adjusted in the light of new knowledge. (Velho and Leite 1993, p. 40)

His political concerns also led him to investigations of political thought and action, nowhere better illustrated than in the long memoir "Teresina and her Friends," in which he describes the political life of the woman who taught him and his brothers Italian and opera, to whom circumstances gave no opportunity for meaningful political action but who nevertheless embodied for him a "socialist way of being."  Deeply sociological, the analysis of Teresina's life separates the problem of political orientation from that of political action, and shows how the latter depends on what the situation makes available to the actor.

The deceptively simple conclusion of this essay—in which Candido shows how a seeming correlation between political moderation and anti-fascism in the Italian community of São Paulo is, after all, historically contingent and not to be taken as indicating a general relationship—embodies a profound understanding of a classical sociological phenomenon, the shaping of social action by contingency.

Candido's style will come through to the English-speaking reader only imperfecly in these translations. It is (he attributes this to his French training (Velho and Leite 1993, p. 37)) characteristically lucid, clear, free of jargon, aimed at a discerning reader who will take the time to follow the line of thought.

While I was teaching sociology, I was against rebarbative language, precisely because this language is at times just a conventional dressing up of banalities. I always thought that, since sociology is a humanistic discipline, it would be a good idea to use, by preference, common language, adopting only those technical terms which are indispensable, which anyway soon become clear, like mobility, stratification, acculturation, etc. . . .

 I always acted in the field of literary studies as I had previously in sociology, preferring also common language and using indispensable technical terms parsimoniously. . . . (Velho and Leite 1993, p. 38)

Candido writes in a way whose naturalness, avoidance of pompous constructions and ostentatious displays of scholarship, and easy exploration of bypaths recalls conversation. He pays close attention to details, using them to reach fundamental characteristics of the work in question, as in his discussion of the national origins of the names of characters in Dino Buzzatti's The Tartar Steppe (contained in "Four Waitings" in this volume). We find ourselves sharing the experience of a careful, subtle reader. (I’ve used some thoughts found in Arrigucci 1992.)

Candido constructs an argument deliberately, without the paraphernalia of scientific "hypotheses," preparing the ground carefully before presenting a conclusion which you have not foreseen but which nevertheless follows clearly, even obviously, from what you have already learned. Thus, the minutely detailed stories of Teresina's Italian socialist friends lead inevitably, though not foreseeably, to his conclusion: though it might seem obvious that milder and more tentative socialists would have resisted the appeal of Mussolini's Fascism where the more aggressive radical socialists could not, in fact these connections are historically contingent and could have easily turned out otherwise.

A world writer

Despite what I have said about his Brazilianness, it would be a mistake to think of Candido as representing a South American or Brazilian or Latin American point of view, or to think that his most important or most characteristic work is that which deals with literature from that part of the world or with the problems of Third World development.

We can think of the intellectual life of the Western tradition as a conversation to which people from all over the world contribute their thoughts and discoveries—in literature, in philosophy, in criticism, in social analysis.  Candido certainly participates in this conversation as someone fully rooted in his time and place, Brazil in the mid-twentieth century, personally involved in the development of a Brazilian literary modernity, helping to shape its concerns through his critical practice and historical and analytic writing. He has wanted to see Brazilian literature develop in a way that would entitle it to stand alongside work from the major centers of Europe and North America, and has devoted himself to establishing literary studies in Brazil on a firm footing.

Thus rooted in his country's literary history and life, he is nevertheless fully involved in the literary conversation of the contemporary world. He writes about world literature, especially that of the Western tradition (Europe and the Americas, but also including the literature of former colonies in the languages of the metropole). He writes about topics of world interest, such universal topics as vengeance or catastrophe, as well as the common topics of contemporary critical writing: class, social change, political engagement.

He draws deeply from the world's resources. His substantial knowledge of world languages and literature is today uncommon; he is at home not only in Portuguese, but in Spanish, French, Italian, and English as well. And not only in the languages, but in the literature and history associated with them, so that he draws examples from all those societies, bringing to the conversation a breadth of comparison seldom available to contemporary critics. And not just literature, but film, drama, and the other arts as well.

The last writer in English who commanded such a wealth of intellectual resources and had such a breadth of interest was the North American critic Edmund Wilson. Oddly enough, Wilson, the literary writer, did more direct reporting on society, in his wonderfully detailed and researched pieces of reportage that amounted to a kind of anthropology or sociology (see Wilson1956, 1960, 1966), than Candido, trained in sociology, whose Os Pareceiros do Rio Bonito, while masterful, is his only work in that genre. But Candido differs from Wilson, who was in so many ways a 19th century anachronism, in being rooted in the 20th century, at home with its problems instead of resenting them as a reminder of the lost privileges of class. And Candido is a more generous spirit, with greater human warmth than the egomaniacal Wilson.

Though Candido takes so much from the world of literature, that world has had little chance to accept the important return of those gifts he offers. Being Brazilian, his language is Portuguese and he did not, as others with his knowledge, interests, and desire to participate might have done, as Brazilian intellectuals so often did in earlier times, move to Paris (or, later, New York) and undertake a career in those central places on the literary map. He stayed in Brazil, wrote in its language, and devoted much of his effort to its literature, unfamiliar (with a few exceptions) to non-Portuguese speaking readers. And so his work is almost unknown elsewhere. His remarks on Machado de Assis, the great Brazilian novelist of the late 19th and early 20th centuries, and on the problems of Latin American writers generally, speak of the difficulties and opportunities contained in his own situation:

. . . at the deepest levels of creative elaboration (those which involve the choice of expressive instruments), we always recognize as natural our inevitable dependence. Besides, seen thus, it is no longer dependency, but a way of participating in a cultural universe to which we belong, which crosses the boundaries of nations and continents, allowing the reversibility of experiences and the circulation of values. And when we in turn influence the Europeans, through the works done by us (not through the thematic suggestions our continent presents to them to elaborate in their own forms of exoticism), at such moments what we give back are not inventions but a refining of received instruments. (Candido 1995, p. 130)

Candido brings to the world conversation a point of view which, rooted in the experience of his country, in its literature, its history, and its social problems, provides new perspectives on literature and its relation to society. He does not just apply the already developed perspectives of the western intellectual community to some new material, or show that Latin Americans too can master these methods. Rather, like Jorge Luis Borges, whose literary inventions altered twentieth century literature, he introduces something new, something the rest of that community can now take up and use, something which enriches the common conceptual stock, which arises from his unique mixing of a knowledge of the world's literature and the situation of a country which is the intellectual descendant, as he says, of a poor relation of contemporary European thought.

A fundamental stage in overcoming dependency is the capacity to produce works of the first order, influenced by previous national examples, not by immediate foreign models. This signifies the establishment of what could be called, a little mechanically, an internal causality, which makes the borrowings from other cultures more fruitful. . . . We know, then, that we are part of a broader culture, in which we participate as a cultural variant. (Candido 1995, p. 131)

Selecting the essays

In selecting essays for this volume, from the large and varied body of Antonio Candido's work, I have tried to exemplify both the unity and the variety of that work, always looking for the essays which best embody his solutions to the literature and society problem.

So, first of all, the essays on Western literature—on the theme of vengeance in Dumas' The Count of Monte Cristo and the themes of catastrophe and survival in the novels of Joseph Conrad, as well as the extended analysis of four major works dealing with the situation of the worn-out society seeking, in conflict with the outside, cultural and spiritual renewal, in "Four Waitings."

Then the studies of Brazilian literature. Here the problem is complicated, because few people will have the patience to read lengthy analyses of works totally unfamiliar to them, as most Brazilian literature is to English-speaking readers. And yet to avoid these topics would only contribute to the problem. Here I chose a middle ground, with an essay on the Brazilian writer most well-known outside Brazil, Machado de Assis, an essay designed to introduce Machado's work (much of it available in English) to non-Brazilians, and a longer essay, which fully exemplifies both Candido's analytic method and his gift for critical paraphrase and summary, on the 19th century novel of the malandro, Memorias de um sargento de milícias.

Political themes are integral to Candido's work and so I chose the short essay on the role of police and the long memoir about Teresina and her friends, in which political thought is dealt with explicitly and directly.

Finally, in addition to his gifts as an analyst of literary works, Antonio Candido is (despite his insistence otherwise) is a first-rate sociological theorist, and so I have included essays on criticism and sociology and on the situation of literature and writers in the developing world.

There are two major omissions. One of Candido's most important works is The Development of Brazilian Literature: Decisive Moments. (Candido 1959) For the reasons already suggested, his description of a literature almost all of which would be totally unfamiliar to readers of this volume weighed against including any of this work. I regret even more not using any of the short essays from In the Classroom (Candido 1985), which deal with Brazilian poetry. The difficulties of translating poetry are well-known and, in this case, the problem would be doubly difficult, even were it within my powers, since his analyses make so much use of the sounds and rhythms of the Portuguese language, most of whose subtleties are integral to the effects he analyzes.

With these provisos, I believe that these essays give readers a good understanding of the breadth and scope of Candido's work, and of the solutions he has provided and modeled for us to the perennial problem of art and society.

References

Becker, Howard S. 1982. Art Worlds. Berkeley: University of California Press.

—. 1992. "Social Theory in Brazil." Sociological Theory 10:1-5.

Buarque de Holanda, Sergio. 1956. Raizes do Brasil. Rio de Janeiro: José Olympio.

Candido, Antonio. 1959. Formacao da literatura brasileira: momentos decisivos. São Paulo: Martins.

—. 1964/1987. Os Parceiros do Rio Bonito: Estudo sobre o caipira paulista e a transformação dos seus meios de vida. São Paulo: Livraria Duas Cidades Ltda.

—. 1965/1985. Literatura e Sociedade: estudos de teoria e história literária. São Paulo: Companhia Editora Nacional.

—. 1985. na sala de aula. São Paulo: Atica.

—. 1991. Critica Radical. Translated by C. Selection, Bibliography, Translation and Notes by Márgareta Rusotto. Caracas: Biblioteca Ayacucho.

—. 1993a. Antonio Candido on Literature and Society. Translated by H. S. Becker. Princeton: Princeton University Press.

—. 1993b. O discurso e a cidade. São Paulo: Livraria Duas Cidades.

da Cunha, Euclides. 1944. Rebellion in the Backlands. Chicago: University of Chicago Press.

Freire, Gilberto. 1986a. The Mansions and the Shanties: The Making of Modern Brazil. Berkeley: University of California Press.

—. 1986b. The Masters and the Slaves: A Study in the Development of Brazilian Civilization. Berkeley: University of California Press.

—. 1986c. Order and Progress: Brazil from Monarchy to Republic. Berkeley: University of California Press.

Hennion, Antoine. 1993. La Passion musicale: une sociologie de la médiation. Paris: Métailié.

Hughes, Everett C. 1971. The Sociological Eye. Chicago: Aldine.

—. 1984. "Good People and Dirty Work." Pp. 87-97 in The Sociological Eye, edited by E. C. Hughes. New Brunswick: Transaction Inc.

Latour, Bruno. 1987. Science in Action. Cambridge: Harvard University Press.

Liebow, Elliot. 1967. Tally's Corner: A Study of Negro Streetcorner men. Boston: Little Brown.

Peirano, Mariza. 1990. "O Pluralismo de Antonio De Candido." Revista Brasileira de Ciéncias Sociais 12:43.

Sutherland, J.H. 1976. Victorian Novelists and Publishers. Chicago: University of Chicago Press.

Velho, Gilberto and Yvonne Leite. 1993. "Os vários mondos de um humanista (interview with Antonio Candido)." Ciéncia Hoje 16:28.

Warner, W. Lloyd and Paul Lunt. 1941. The Social Life of a Modern Community. New Haven: Yale University Press.

Whyte, William Foote. 1955. Street Corner Society: The Social Structure of an Italian Slum. Chicago: University of Chicago Press.

Wilson, Edmund. 1956. Red, black, blond, and olive; studies in four civilizations: Zuni, Haiti, Soviet Russia, Israel. New York: Oxford University Press.

—. 1960. Apologies to the Iriquois. With a study of the Mohawks in high steel by Joseph Mitchell. New York: Farrar, Straus and cudahy.

—. 1966. Europe without Baedeker: Sketches Among the Ruins of Italy, Greece and England, together with notes from a European Diary, 1963-1964. New York: Farrar, Straus and Giroux.

Zolberg, Vera. 1990. Constructing a Sociology of the Arts. Cambridge: Cambridge University Press.