Telling the Truth

The Theory and Practice of Documentary Fiction

Barbara C. Foley

• Publisher: Cornell University Press
• Year: 2018
• ISBN:
  9781501722905
• Topic: Literatura
• Subtopic: Crítica literaria histórica

Barbara Foley here focuses on the relatively neglected genre of documentary fiction: novels that are continually near the borderline between factual and fictive discourse. She links the development of the genre over three centuries to the evolution of capitalism, but her analyses of literary texts depart significantly from those of most current Marxist critics. Foley maintains that Marxist theory has yet to produce a satisfactory theory of mimesis or of the development of genres, and she addresses such key issues as the problem of reference and the nature of generic distinctions. Among the authors whom Foley treats are Defoe, Scott, George Eliot, Joyce, Isherwood, Dos Passos, William Wells Brown, Ishmael Reed, and Ernest Gaines.

PART I

THEORY

1

The Documentary Novel and the Problem of Borders

Belief in fiction cannot be a matter of degree. We either accept the incidents of a story as if they were true, or we are aware of them as fiction. There can be no halfway house, no keeping an open mind, no suspending our judgement until further evidence is available.

—Vivienne Mylne

In this book I shall be arguing that the documentary novel constitutes a distinct fictional kind. It locates itself near the border between factual discourse and fictive discourse, but it does not propose an eradication of that border. Rather, it purports to represent reality by means of agreed-upon conventions of fictionality, while grafting onto its fictive pact some kind of additional claim to empirical validation. Historically, this claim has taken various forms. The pseudofactual novel of the seventeenth and eighteenth centuries simulates or imitates the authentic testimony of a “real life” person; its documentary effect derives from the assertion of veracity. The historical novel of the nineteenth century takes as its referent a phase of the historical process; its documentary effect derives from the assertion of extratextual verification. The documentary novel in the modernist era bifurcates into two distinct genres. The fictional autobiography represents an artist-hero who assumes the status of a real person inhabiting an invented situation; its documentary effect derives from the assertion of the artist’s claim to privileged cognition. The metahistorical novel takes as its referent a historical process that evades rational formulation; its documentary effect derives from the assertion of the very indeterminacy of factual verification. Finally, the Afro-American documentary novel represents a reality submitting human subjects to racist objectification; its documentary effect derives from the presentation of facts that subvert commonplace constructions of reality. In all its phases, then, the documentary novel aspires to tell the truth, and it associates this truth with claims to empirical validation. If it increasingly calls into question the possibility of truth-telling, this skepticism is directed more toward the ideological assumptions undergirding empiricism than toward the capacity of fictive discourse to interpret and represent its referent.

Clearly the documentary novel, as I define it in this book, is not a minor subgenre that can be readily relegated to the margins of novelistic production in any given era. On the contrary: in the seventeenth and eighteenth centuries, the documentary novel is closely aligned with writing that Lennard J. Davis calls the “news/novel discourse”;¹ in the nineteenth century, the documentary novel intersects with the major tradition of realism; in the early decades of the twentieth century, it participates in the principal concerns of modernism. Much writing in the entire domain of Afro-American prose fiction has a pronounced documentary quality. Thus central texts from each phase in the history of the novel (for example, Moll Flanders, Pamela, Waverley, Henry Esmond, Orlando, A Portrait of the Artist as a Young Man, Native Son) can be adjudged to be documentary novels. But while the documentary novel overlaps with the mainstream tradition of the novel, it is not identical with this tradition. Rather, the documentary novel is distinguished by its insistence that it contains some kind of specific and verifiable link to the historical world. (Whether or not this link succeeds in being “extratextual” in a larger sense remains to be seen.) It implicitly claims to replicate certain features of actuality in a relatively direct and unmediated fashion; it invokes familiar novelistic conventions, but it requires the reader to accept certain textual elements—characters, incidents, or actual documents—as possessing referents in the world of the reader. The documentary novel is not superior to other modes of fictional discourse in its capacity for assertion—all fictions assert their propositional content with equal force and sincerity, I believe—but it does raise the problem of reference for explicit consideration. To investigate the truth-telling claims of the documentary novel is thus to illuminate the assertive capacities of fiction in general.

As we shall see in the literary-historical portions of this book, the documentary novel’s shift in representational strategy reveals the inevitable historicity of generic definitions. Factual and fictive discourses are not immutable essences but are historically varying types of writing, signaled by, and embodied in, changing literary conventions and generated by the changing structures of historically specific relations of production and intercourse. As M. M. Bakhtin has remarked, “The boundaries between fiction and nonfiction, between literature and nonliterature and so forth are not laid up in heaven. Every specific situation is historical. And the growth of literature is not merely development and change within the fixed boundaries of any definition; the boundaries themselves are constantly changing.”² In examining the documentary novel’s protean identity, I have had to abandon many prior conceptions about what constitutes fiction, the novel, history, and the elusive quality that I am terming the “documentary” effect. Modes of discourse do not remain within “fixed boundaries”; they change as much as do the modes of social and political representation in the worlds that they take as their referents.

I have discovered, nonetheless, that the distinction between fictional discourse and its various nonfictional counterparts—history, journalism, biography, autobiography—has remained a qualitative one. The need to distinguish between narratives held to be imaginary and those held to be directly representational would seem to be not a post-Cartesian phenomenon, testifying to the alienation of subject from object, but an abiding feature of discursive production. Even in the seventeenth and eighteenth centuries, when the documentary novel possessed its most ambiguous generic identity, the issue was not that prose fiction simply blended into purportedly veracious kinds of writing but that its primary locus, the romance, could not effectively assert the kind of truth that the early pseudofactual novelists wanted to tell. The pseudofactual novel’s ambiguous generic status does not mean that writers and readers of the time inhabited an ontological haze but that they felt obliged to simulate veracious discourses if they hoped to appear credible to their readers. To say, as Bakhtin does, that the borderline between fiction and nonfiction is “constantly changing” does not mean that writers have not routinely respected such a borderline; it means, on the contrary, that writers have composed their fictions in contradistinction to one or more acknowledged forms of nonfictional writing. Fiction, I would propose, is intrinsically part of a binary opposition; it is what it is by virtue of what it is not.

To some, this argument will seem trivial or self-evident. As I indicated in my prefatory remarks, however, recent years have witnessed a wholesale assault upon the idea that fictional and nonfictional writing can—or should—be qualitatively differentiated. The theoretical arguments for the nonqualitative position come mainly from poststructuralism, but many writers and critics have added their voices. In this prolegomenon to my central theoretical and historical discussions, I shall confront the principal claims of my adversaries by arguing on logical grounds for the superiority of a qualitative view of mimesis. (By “qualitative” I mean different in kind rather than in degree.) In so doing, I run the risk of appearing to endorse an ahistorical or essentialist view of mimesis. Certainly it is true, as we shall see, that many defenses of the uniqueness of mimesis do in fact reify the realms of fictional and nonfictional discourse and deny their continually altering character. I believe, however, that a qualitative approach to the matter of defining fictionality is consonant with the premises of a materialist literary theory, so long as we remember that binary oppositions are dialectical oppositions as well. In order to counter the objection that my entire analysis of the theoretical and historical features of documentary mimesis is based on a fallacious premise about the distinctness of discursive kinds, I shall defend my procedure here, trusting that the full validation of my argument will emerge in the chapters that follow.³

The view that fictional and nonfictional discourse cannot be qualitatively distinguished ordinarily rests on one of three arguments. The first of these, which I shall term the “spectrum” argument, centers on the claim that the significant qualities of factuality and fictionality inhere in separate facets of a literary work, rather than in any informing paradigm, and that the task of criticism is to assess the impact of these upon the work’s rhetorical effect. Thus Paul Hernadi advocates a “microstructural theory of poetic discourse,” which holds that any given literary work possesses aspects of various discourses and should be analyzed not as a text unified by a single generic frame but as a unity of multiple components reflecting the richness of literary discourse in general. Hernadi repudiates the investigation of “generic conventions as reflections of historically conditioned preferences of writers and readers.” Instead, he asserts, “The finest generic classifications of our time make us look beyond their immediate concerns and focus on the order of literature, not on borders between literary genres.”⁴ Scholes and Kellogg, in The Nature of Narrative, reach a similar conclusion from the opposite direction, for they insist that it is such “historically conditioned preferences” that furnish the logical basis for a nonqualitative definition of mimesis. The novel has historically synthesized two narrative impulses, they argue, one directed toward the “empirical,” or historical, and the other directed toward the “fictional,” or imaginary. Empirical and fictional are blended tendencies, rather than distinctive kinds; history and fantasy stand as the poles of a narrative spectrum, with different narrative forms such as autobiography, realism, and romance occupying positions at various points along the scale. The “recording of specific fact, the representation of what resembles specific fact, and the representation of generalized types of actuality,” declare Scholes and Kellogg, are all to be aligned along the empirical part of the narrative spectrum.⁵

A second type of nonqualitative argument, which I shall term the “family resemblance” argument, is based upon an invocation—albeit somewhat simplified—of Ludwig Wittgenstein’s theory of linguistic reference. In his discussion of the definitive characteristics of concepts such as games, Wittgenstein points out the difficulties involved in the attempt to delineate a limited set of criteria that descriptively include all activities commonly held to be games and exclude all other activities: “Consider for example the proceedings we call ‘games.’ . . . What is common to them all?—Don’t say, there must be something common, or they would not be called ‘games’—but look and see whether there is anything common to all. For if you look at them you will not see something that is common to all, but similarities, relationships, and a whole series at that.” Drawing an analogy between games and the physical traits shared by the members of the same family, Wittgenstein concludes, “We see a complicated network of similarities overlapping and crisscrossing; sometimes overall similarities, sometimes similarities of detail.”⁶ For Wittgenstein, no fixed set of properties defines the term “game,” just as no fixed set of physical characteristics is shared by all members of the same family. For Wittgenstein, language seduces us into believing that certain words denote actually existing sets of relations, when all that these words really denote is concepts used to order the world of things.

Applied to the problem of defining mimesis, the “family resemblance” argument proposes that qualitative distinctions between factual and fictional discourse are founded upon fallacious logic. Morris Weitz, for example, invokes Wittgenstein when he asserts that all aesthetic categories, including the theory of literary kinds, are concepts with blurred edges and that it is therefore impossible to formulate a clear description of mimetic discourse. Asking himself whether an experimental fiction utilizing new kinds of referential procedures can be classified as a “novel,” he observes, “What is at stake here is no factual analysis concerning necessary and sufficient properties but a decision as to whether the work under examination is similar in certain respects to other works, already called ‘novels,’ and consequently warrants the extension of the concept to cover the new case.” Every such classificatory decision is, however, necessarily an ad hoc decision:

“Art,” itself, is an open concept. New conditions (cases) have constantly arisen and will undoubtedly constantly arise; new art forms, new movements will emerge, which will demand decisions on the part of those interested, usually professional critics, as to whether the concept should be extended or not. . . . Art, as the logic of the concept shows, has no set of necessary and sufficient properties, hence a theory of it is logically impossible and not merely factually difficult.⁷

Charles Stevenson, also invoking Wittgenstein, introduces a mathematical model to solve the problem of fictional classification. Since fictionality...