SOURCE: “Fiction and History: A Common Core?,” in International Studies in Philosophy, Vol. XXIV, No. 1, Spring, 1992, pp. 47–63

[In the following essay, Cebik examines the philosophical basis for conflating historical writing and literary fiction, as suggested by White's theoretical model of historical discourse and typological schema.]

In the last decade, a fad has swept across philosophic discussion of narrative discourse. In boldest terms, the fad consists of treating historical and fictional narrative on a par. Each has equal standing before the bar of human knowledge; each has equal if not identical epistemic standing.

The fad has many roots. Deconstruction's casual dismissal of the text releases every set of narrative sentences for subjective interpretation by the reader, making every act of reading one of artistic creation also. Ricoeur attempted to show the “precedence of our narrative understanding in the epistemological order” in his reconstruction of narrative into a metaphysics of time.¹ Whether or not he merely follows the leads of the structuralistic movement or responds to a longer standing impulse stemming from Bergson we may leave to another day's speculation.

THE INDISTINGUISHABILITY OF HISTORY AND FICTION

In this country, the chief proponent of equating historical and fictional narrative has been Hayden White, the father of a significant school of vociferous offspring. Relying upon the same structuralistic heritage as Ricoeur and the deconstructionists, e.g., Jakobson, Halle, Levi-Strauss, and Lacan, White reaches in his own style the same result. This result has not been overtly intentional, since White has placed numerous disclaimers into his essays. In the Metahistory, White separates fiction from history via the inability to distinguish in fiction between chronicle and story, a fundamental distinction for White in history. Oddly, this distinction appears in merely a footnote.² In Tropics of Discourse, White elevates the distinction to a place in the text.³

Still, the basic thrust of White's efforts rests on a willful insistence upon treating the historical text as a literary artifact and as that alone.⁴ He dismisses differences in a sentence, e.g., “I wish to grant at the outset that historical events differ from fictional events in the ways that it has been conventional to characterize their differences since Aristotle.”⁵ Having granted this much, White then proceeds to his main thesis, one underlying the earlier Metahistory and the later essays as well:

Viewed simply as verbal artifacts histories and novels are indistinguishable from one another. We cannot easily distinguish between them on formal grounds unless we approach them with specific preconceptions about the kinds of truths that each is supposed to deal in. But the aim of the writer of a novel must be the same as that of a writer of a history. Both wish to provide a verbal image of “reality.”⁶

This little passage contains a good bit of mischief. On the surface, we find a claim of total indistinguishability immediately qualified by a disclaimer that the distinction is simply not easily made. More significantly, we encounter seemingly illicit activities contained in trying to distinguish between novels and histories: namely, that we must approach them with certain preconceptions.

Disregarding for the moment White's view of the type of preconceptions with which we must approach histories and novels, we can surely ask whether the need to approach such narratives with preconceptions is illicit or inappropriate in any way. Bruce Waters long ago wrestled with similar questions, writing before the ascendancy of the structuralists and their jargon, but after the ascendancy of the positivists and their jargon. He concluded that “It is possible [by political and other means] completely to fictionize history,” for example, to make Sartoris rather than Forrest the Confederate cavalry leader at Shiloh. “In history,” wrote Waters, “we can never get beyond assent.”⁷ Therefore, to see history as in principle different from fiction is to come to history with something in mind: for Waters, a theory or philosophy of history, for White a preconception.

We may grant that both White and Waters have recognized the rapids of having both fiction and history in the mainstream of narrative discourse. Because they are indistinguishable as verbally and visually similar linguistic products, only our approach to a given product makes the difference between them. Indeed, it is not implausible to suggest that at root we do not discover that a work is fiction or that it is history; instead, we decide that question as we approach a work. A book's dust cover may in some cases prompt us. In other cases, we may scrutinize the contents and compare them with much that we learn from other sources. The decision is not arbitrary; neither is it always easy. However, a preponderance of evidence may compel a certain decision. The evidence that a work is history, however, is not identical with the evidence for the history it contains. In the former case, we are deciding between two categorizations for a narrative; in the latter, we are determining the truth or falseness of the history's statements (among other matters).

However we come to decide whether a work is history or fiction, we decide much more than a label. We decide in fact the entire cluster of questions we can pose to the narrative. Some questions we may ask of both fictional and historical narratives. Is it well written? What metaphors occur? Does the narrative hang together with proper connectives? Other questions befit one or the other but not both types of narratives. As Macdonald noted in the 60s, “nothing can count as evidence in favor of a fictional story. And what no fact can confirm none can disconfirm, either.”⁸ Therefore, confirmation of the events of the story is an illicit question to pose to fictional narratives, although it remains central to history. In contrast, we may ask whether the characters of a novel achieve verisimilitude, although the same question put to a history would likely only hide an ironic criticism.

Given this much alone, we are in a position to question the next step of White's argument, namely that we approach fictional and historical narratives with specific preconceptions about the kinds of truths in which each is supposed to deal. In The Content of the Form, White tried to do away with this part of his formula by transferring his focus from truth to reality. Real events “offered as the proper content of historical discourse” are real “not because they occurred but because, first, they were remembered and, second, they are capable of finding a place in a chronologically ordered sequence.” Fiction, by contrast, has no chronology (or was not remembered) and is thus incapable of yielding both a chronological and a narrative version of the same set of events.⁹

This evasive maneuver, fueled by the time-worn appeal to a chronology and to memory, allows Waters’ potential falsification of history to alter reality. However, the point has seeming weight only so far as we exclude the possibility of there being evidence for the occurrence of events, along with methods for the evaluation of evidence. To date, Collingwood has perhaps explicated best the relationship of questions, evidence, and answers.¹⁰ Claims as to the reality of events are, of course, no more than claims to the effect that they occurred, claims requiring (upon demand or dispute) an array of acceptable evidence. Histories, therefore, necessarily deal in questions of truths, however difficult they may be to answer satisfactorily.

In contrast, fictional sentences necessarily presuppose truths, but not as questions.¹¹ Fictional sentences have sense just because they are instances of generalizations whose truth is not in question relative to the fictional text. Thus, for history it is correct to say that it should deal in truth, where “should” indicates that there is an activity to perform. We cannot say the same for fiction, for its relationship to truth is not an activity, in just the manner that presupposing is not an action, but a logical condition of action.

To be involved in or with truth is not at all the same as being involved in the production of “a verbal image of ‘reality.’” In fact, it is questionable whether the writer of history or the writer of fiction can be correctly described as providing a verbal image at all, let alone one of reality. Certainly, it is hardly ever if at all the intention of either kind of writer to provide verbal images except as matters of style enter into their project. Either may say or think, one imagines, that a writer wishes to choose precisely the right words to create a vivid and unforgettable image of what they are presenting. But that is but one possible thought among many. Their respective projects are not limited by style or to style. White's description of creating a narrative would, by contrast, precisely limit both tasks to matters of style.

The history writer reports, records, interprets and argues, among other things. The story he tells, if he chooses to tell what he tells as a story, is not an image of reality; it is reality. In so far as we are dealing here not with an artifact, but with a human activity, the historian tells us not an image of what happened, but simply what happened. The fiction writer creates his characters and his story. Whether either constitutes a reality is one of the writer's options. Likewise, that we take his work as an image of reality—or of unreality—is one of our options as readers. We may argue with a historian. We may also willfully refuse to believe him or her. But we do not have such options with the fiction writer, even though he or she may write with varying degrees of believability. Moreover, a historian who writes what we take to be unbelievable is not a bad stylist or bad storyteller; he or she has said something we cannot accept as true.

Any accurate description, then, of either history or of fiction cannot survive solely at the level of the literary artifact. This conclusion does not deny any of the stylistic or rhetorical facets White has found in 19th century histories. Instead it affirms that these facets are just that: facets and not the entirety of the work. History cannot be taken as solely a literary artifact except as philosophy or metahistory may restrict themselves to ignorance. The ignorance is not merely whether certain historical facts, findings, and techniques are correct. There must also be ignorance of whether we may have history at all. Then, and only then, would it be the case that we merely tell stories, we merely narrate, and this just to impress at one or another level the hearer or reader.

Both history and fiction, as narratives in a world that recognizes both kinds, are complexes of activities that defy on pain of senseless distortion such restrictive description. As such, they require treatment as activities, not as a collection of literary artifacts. Perhaps the structuralist and post-structuralist turn of thought has failed to realize what other philosophical approaches to language have realized for decades: language is not everything, and language artifacts are even less.

THE PRECONCEPTUAL AND THE AESTHETIC

To take White to task so for his excesses would seem almost fatuous, for he is easy pickings in the stream of counterargument. After all, he is predominantly a historian and not a philosopher given to care in phrasing arguments. Excepting his influence, all this analysis would be otiose did it not reveal a worthier thesis to examine, one presupposed by the passage in question. If history and fiction as literary artifacts are indistinguishable, if they purport to present verbal images of “reality,” and if they have equal epistemic status, why the reduction of history to fiction rather than an equivalent reduction of fiction to history?

White explained the direction of his choice in the Preface to Metahistory and has not changed his central view since:

… I have been forced to postulate a deep level of consciousness on which a historical thinker chooses conceptual strategies by which to explain or represent his data. On this level, I believe, the historian performs an essentially poetic act, in which he prefigures the historical field and constitutes it as a domain upon which to bring to bear the specific theories he will use to explain “what is really happening” in it.

White expands this theme by noting that the prefigurative act is poetic because it is precognitive and precritical in the historian's consciousness, because it is constitutive of the structure of the emergent history, and because it is constitutive of the concepts used to identify objects and relationships.¹² The most interesting question here is where we may root the idea that such prefiguring is in fact a poetic act. The answer—or an answer—lies in White's adoption of a Nietzschean perspective.

Nietzsche is neither the ultimate nor the proximate source of White's poetization of narrative prefiguration. The idea grows throughout the 19th century, with roots in both German and English Romanticism. Closer to White, as he generously recounts in an extensive footnote, Jakobson, Benveniste, and others collapse distinctions that would leave poetry distinguished from other rhetorical modes of discourse under a unifying collection of “poetic tropes.”¹³ Neither ultimate nor proximate, Nietzsche nevertheless serves as a focal source. He enunciates a view which—as interpreted by White and others—captures the core of the thesis in question. If Nietzsche is wrong, then so too are his followers and successors who would subsume the epistemic under the artistic or aesthetic.

White's analysis of Nietzsche in the Metahistory indeed foretells much of his own perspective on history. White sees Nietzsche has having changed the linguistic rules of the historiographic game through a critique of its artistic component. The goal is for history to become once more an art.¹⁴

To Nietzsche the form, meaning, and content of all science and all religion were aesthetic in origin, products of a human need to flee from reality into a dream, to impose order on experience in the absence of any substantive meaning or content. He held all “truths” to be perversions of the original aesthetic impulse …¹⁵

What is needed by Nietzsche, as filtered through the eyes of White, is an art aware of its metaphysical purpose, which is not to imitate nature, but to supplement it and overcome it.¹⁶ Even the impulse behind philosophy is an aesthetic one; that is, it has its origins in the desire to impose form on the world.¹⁷

Where White draws a line distinguishing himself from Nietzsche is in the German's attempt to release aesthetic sensibility from morality and wed it to the will. Thereby, claims White, we turn “life itself away from that knowledge of the world without which it cannot produce anything of practical benefit to anyone.”¹⁸ Instead, White envisions historians freed to conceptualize history “in whatever modality of consciousness is most consistent with their own moral and aesthetic aspirations.”¹⁹ However, if the foundation of all constructs, including historical constructs, is aesthetic, then moral freedom and whimsy are indistinguishable. Moreover, if the motivation for dream creation is escape from reality, then all histories are rebellious or revolutionary in the senses developed by Camus in The Rebel, or they are mere fantasies. Despite these consequences, White views without variance the construction of prefiguring conceptual structures for historical narratives as sheerly aesthetic, poetic, artistic: a function of some deeper consciousness of the historian.

Calling the prefiguring of narrative structure a function of a deeper consciousness presents problems of its own. As literary artifacts, histories (and narrative fictions) have structures of event presentation and connection that we may say the narrative presupposes. But as earlier noted, presuppositions are logical conditions of making sense; they are not necessarily the product of conscious (or unconscious) effort. Only if, like Nietzsche, we view them as functions of will do we make them products of individual thought and activity. It is in this context that Nietzsche posed the following question: “Suppose we want truth: why not rather untruth? and uncertainty? even ignorance?”²⁰ From here it is an easy step to the declaration that “the falseness of a judgment is for us not necessarily an objection to a judgment. …²¹

As a figure of speech, Nietzsche's counterpoise of “untruth” to “truth” might be interpreted as merely neutralizing the entire category of “truth” from applicability to foundational preconceptions. If this is all Nietzsche is after, then the point is unexceptionable. For Nietzsche to be wrong, we must take his remarks very literally, so literally that epistemic consequences flow from his remarks. White reads him just this way.

However, prefiguring or imposing form upon nature or history occurs at a level that logically precedes the actual investigation of nature or history. The level is indicated in the mode of criticism White applies to it: it results in benefit or detriment for humankind or human individuals. There are no methods of investigation, no possible evidence to tell us that a given formulation is true or false. If “truth” applies at all, it is at best in the Heideggerian sense of truths of Alethera, which have no opposition of falsity.

What Nietzsche—and Heidegger, for that matter—are wrestling with here is a way to formulate the notion of fundamental constructs that inform perception itself, let alone description and explanation. Depending upon whom one reads, we have greater or lesser freedom in creating these constructs. For the problem at hand, the philosophic difficulty stems from calling these constructs art. They are not fictions in the way that a novel about outer space or dragons is a fiction. They are not stories just to the degree they may be able to establish and limit the shape stories may have. They are not poetry just to the degree that they defy being set aside like poetry. They are not truth, untruth, or falseness just to the degree they may determine what propositions can possibly qualify as being true or false.

In short, preconceptual constructions of this order are and can be neither art nor nonart: they logically precede the distinction. Nietzsche's and White's importation of the language of art, the aesthetic, and rhetoric to this level of preconceptual construction is as misleading as Heidegger's attempt to preserve the language of truth within the same arena. Preconceptual constructions are neither truth nor untruth, neither fictions nor nonfictions. If in fact Nietzsche viewed them as art, then he was as trapped in his own metaphor as all those who have followed him, White included.

Dropping the terms “art,” “aesthetic,” and “poetry” from the category labels for preconceptual constructs carries with it the elision of innumerable implications drawn almost strictly from those terms. The reality-forming function of such constructs does not carry over into fictional narratives just because they also happen to be artistic fictions. The question of whether novels or other narrative fictions have any necessary reality-forming functions remains open to be settled on its own ground.

Likewise, the question of history's epistemic status remains its own and not tied to the fortunes of fictional narrative. It is neither prima facie true nor reflectively evident that “the aim of the writer of a novel must be the same as that of a writer of a history.”²² The respective aims of each type of writer may be—and usually are—as complex as the particular projects which engage them. All that remains true of both of them is that they employ narrative.

The ultimate confusion, common to White and Ricoeur, that produces a morass of misconception lies in thinking that all elements of narrative form are preconceptual constructions. Constructions many may be, but hardly preconceptual. What is preconceptual—or more correctly presupposed by any narrative construction—is narrative discourse and its requirements for simultaneous temporal and content relations built into sentential structures. To say this, however, is no more than to say that the logic of narrative discourse is presupposed by any actual narrative discourse.

In contrast, to the very extent that we can analyze narrative forms into rhetorical types—whether or not we agree with White's particular list—we can also consciously and rationally choose the rhetorical form for our narrative. That possibility makes the selection of rhetorical form, if not purposive, at least functional.

Functionality subjects rhetorical form to criteria, not of truth, but of evaluation and judgment. We may evaluate a narrative form relative to its content as effective, adequate, complete, and cogent to any degree from negligibly to wholly so. We may judge narrative forms as good or bad, as right or wrong for us as individuals, as communities, as a people. What we can choose, evaluate, and judge according to function we can also teach others how to use. Choosing a narrative form, then, is not either preconceptual or aesthetic in the requisite sense precisely because it is functional and therefore can be purposive.

Indeed, White's Nietzsche can call the imposition of order aesthetic in origin only if he can postulate a prearticulate—and indeed, prenarrative—activity of conjuring informed worlds. Although Nietzsche's attempt to relate Homeric epic poetry to dreaming suggests that at least early on he may have believed such a view,²³ it explains virtually nothing in narrative literature. As Macdonald noted, “Like the content of dreams, the objects of fiction may presuppose, but do not compete with, those of ordinary life. Unlike those of dreams, however, they are deliberately contrived.”²⁴ Not only may the objects of fiction be deliberately contrived, so too may be the rhetorical structures. Contrarily, the impulse to dream, to order reality in sleep pleasantly or otherwise, is not aesthetic. It simply is, if it is an impulse at all.²⁵

More recently, White has tried to avoid the language he coined in Metahistory, opting instead to “dissolve the distinction between realistic and fictional discourses” by dropping ontological presumptions about their respective “referents.” White now prefers to adhere to the body of semiological theory that encloses language within a “systematic substitution of signifieds (conceptual contents),” thereby yielding a “system of meaning production” so that individuals can live in an imaginary—an “unreal but meaningful”—relation to “their real conditions of existence.”²⁶ No less self-contradictory than his earlier formulations—in this case, with respect to whatever may be real—White's new foundational statement simply transfers the problem of vacuous contrast to different terms.

First, it will not do to substitute “realistic” for “historical” in naming types of discourse, since the realistic is the feigned, and this label prejudices the case before any justification for the label begins. Second, and more significantly, terms like “imaginary” and “unreal” become as illicit as the term “aesthetic.” A theory that eliminates any possible justification for calling something real equally loses reason for terming anything unreal. Likewise, without grounds for determining that something is nonimaginary, we cannot warrant any claim that something is imaginary. Meaningful relations “to social formations” generally require that we be able to sort out, however imperfectly, the erroneous from the correct, the real from the feigned, the natural from the imaginary. Thus, White's socialized individuals either live inherently meaningless lives or the terms of the theory carry nothing of their everyday content into their theoretic employment. At the theoretic or foundational level, activity is no more imaginary than it is aesthetic or poetic.

In the end, White's three reasons for labeling so-called prefigurative acts poetic turn out to be reasons for nothing at all. To the degree that such acts are preconceptual—constitutive of concepts used to identify objects and relationships—they can be neither conscious nor aesthetic, and to the degree they are aesthetic, they cannot be preconceptual. Even if some aspects of the imposition of rhetorical form are also constitutive of the emergent history, the process is one of overt selection and not precognitive constitution: it is a thoroughly conscious, functional, and often purposeful act. At this level, White's four quartets of historical rhetoric reduce to a simple heuristic typology, useful but devoid of profound theory.

THE STRUCTURE OF NARRATIVE THEORY

The fad of equating history and fiction as rhetorical and therefore artistic, poetic, or imaginative enterprises flounders upon the simple bad habit of tossing everything not susceptible of truth into the barrel of art and the barrenness of the aesthetic. That practice only gives art and the aesthetic a bad name, if not a like smell. The number of things and activities unrelated to truth is legion. Most are not art. Most are not metaphysical. Most are important, or at least interesting. Some are useful to boot. The rhetoric of narrative falls somewhere within the latter categories on this list.

Free of the bias to split our world into things susceptible of truth and those not susceptible of truth, we may take a quite different view of the various ways of looking at narrative. The following outline represents one sort of prolegomena to future narrative theory. It is not the only sort; rather, this particular outline serves the present purpose of trying better to understand the role of the preconceptual and the presuppositional in several of the facets of narrative. To that end, we may employ the language of types, perhaps even of Weberian ideal types.

Initially, we may constitute fiction and history as two types along a continuum of types in a methodological typing of narratives. Within the methodological typing of narrative arise many of the considerations regarding truth that we have had occasion to note in reviewing White's conflation of the rhetorical and the epistemic. However, methodology includes many more problems: for example, the nature and justification of historical constructs and the nature and use of conceptual evidence in history and in fiction.²⁷ The huge range of historical constructs from the level of the assertion of the ordinary facts to the level of what Ankersmit calls narratios begs for a common methodological language that transcends the simple idea of true statements and leans more toward Dewey's notion of warranted assertability.²⁸ Similarly, the question of “conceptual evidence,” a phrase coined by Danto, is only partially concerned with the truth of our claims; it is equally, if not more significantly, concerned with the use of such evidence in both history and fiction.²⁹

By treating paradigmatic history and paradigmatic fiction as types on a continuum, we achieve the ability to eliminate false methodological problems occasioned by mixed types. Historical fiction, such as War and Peace, and fictionalized history, such as I, Claudius have proper places on the continuum. We cannot view them merely as extensions of fiction on the simplistic grounds that they employ the findings of history and contribute nothing to the ongoing doing of history. Indeed, writing such an intermediate type of work may involve doing history as much as it involves the creation of a fiction. Only a superiority complex that might deny merit to the contributions made by amateur and local historians would also deny the label of doing history to certain activities of those we call novelists. Methodologically, such disputes are irrelevant.

As noted earlier, the methodological distinction between history and fiction consists in the collection of sensible questions we may bring to the work—either as an activity in which we are engaged or as a product which we read. Collingwood noticed that, although we may raise questions about certain questions in the course of questioning (relative presuppositions), questioning others (absolute presuppositions) ends the inquiry by putting its sense in jeopardy.³⁰ The decision to treat a work as history or as fiction can be conscious and for reason, but relative to that decision, the collection of questions constituting historical and literary inquiries is presupposed as one condition of making the decision sensible, as an absolute presupposition of the decision. Only in the process of actual inquiry may we sensibly question the questions we put to history or to fiction, for only then have we the possible means of answering them in terms of better questions for which we can have evidence. We still have far to go in sorting out the methodological presuppositions of narrative.

One advantage of calling this typology methodological rather than epistemic is that the label places questions of truth and knowledge within the framework of the activities that go into producing historical and fictional narratives. Giving priority to questions of truth and knowledge, on the other hand, has often bent the practices of both history and fiction to the requirements of a preset theory. Thus arose the unnecessary concerns for history's inability to meet correspondence requirements for truth and equally otiose concerns that somehow fictional narrative might be a form of lying. Many such spurious anxieties disappear when one gives priority to the nature of the activity.

We may also typologically distinguish argumentative forms of narrative or arguments within narrative, as most notably did Dray.³¹ The argumentative typology treats the narrative as a vehicle of explanation. Dray inherited the attempt to transform all written history into disguised or overt attempts to explain why events occurred by subsuming them under general laws, however loosely or probabilistically they might be formulated. His rejoinder, often misinterpreted as an anti-covering-law bias, was to expand the number of types of explanations making use of the narrative form.

Among Dray's types were explanations showing the rationale of actions, explanations showing how events could have occurred contrary to first appearance, and explanations of what events were or amounted to as an exercise in “colligation.” Passmore expanded the list of explanatory types—at least for “everyday life”—and argued that such explanatory types occurred in history more usually than covering law explanations.³² The function of an explanation, on Passmore's ordinary language account, is to clear up a puzzlement. An explanation can only occur, therefore, if a puzzlement—a question of a certain order—preexists the account, and the account succeeds if the puzzlement disappears. The type of explanation required corresponds to the nature of the puzzlement involved.

At first sight, any explanatory typology seems fit only for historical narratives. First, as Collingwood and many others have noted, history begins with questions—at least one, but usually more. History answers questions—puzzlements, if one likes—by adducing evidence from among the relics of the past. Second, the requisite puzzlement presumes logically that one already knows something of what happened, enough to raise the puzzle in the first place. Indeed, philosophical questions concerning explanation have largely focused upon the presuppositions of historical explanations and the situations that call for an explanation of one sort or another.

First sight yields an illusion. Historical narrative is not the narrative of historians, but any narrative utilizing the methods of question, evidence, and answer to settle puzzlements. When a Sidney Sheldon apprentices himself for three years to learn the life and ways of Italian sculptors in preparation for his novel on Michelangelo, his efforts constitute an explanatory investigation (within limitations of sculpting in the modern era) to answer the puzzle of what life would have been like for the Renaissance artist. How Sheldon presented his findings departs from the norm for historical writings, but then his work never pretended unto Library of Congress cataloging with history books.

Likewise, we may mistakenly treat mixed narratives as excessively historical. Many early readers of Malraux's La condition humaine treated the work as principally historical, down to the existence of his characters. Only later did we discover that Malraux's involvement in Chinese revolutionary movements in the 1920s did not extend to the details of the Shanghai insurrection: he had not even visited the city when he wrote the novel. The historicity of a novel's details is open to questions for which evidence may permit an answer. Malraux's novel, so it turns out, was not a historical explanation for the insurrection's failure. Nonetheless it may well be an explanation of the mentalities of revolutionaries.

Besides showing the relevance of explanatory types throughout the methodological spectrum, the examples also demonstrate the distinction between typologies. Although there may be no absolute separation between the methodological and explanatory typologies, one may deal with explanatory types without necessarily addressing matters of methodology or of rhetoric. Moreover, truth is significant only within certain types of explanations: it fails to be a fundamental concern or presupposition of explanatory narrative in general. The criteria for explanations include accuracy, adequacy, and relevance; and truth is but a part of one of them (accuracy).³³ The consequences of the explanatory typology thus do not compete with those of methodology.

WHITE'S RHETORICAL TYPOLOGY

We may also, as did White, distinguish rhetorical types among narratives without ever invoking the distinction between history and fiction. We may overlook that distinction simply because the typology we are addressing is not methodological. As was the case with the explanatory typology, a rhetorical typology does not compete with the methodological.

White wishes to think, however, that the rhetorical typology competes with the explanatory. He characterizes emplotment, formal argument, and ideological implication as modes of “explanation.” It turns out, nonetheless, that emplotment simply identifies “a story of a particular kind.”³⁴ Argumentation consists not in actual modes of argumentation, but favoring certain sorts of narrative connectives among the many sorts to be found. Indeed, only White's mechanistic (cause-effect), organicist (teleological), and contextualist (colligatory) modes specify types of connectives. If no one type dominates over others, White calls the work “formist.”³⁵ Dray and others have enumerated far more connectives than White envisions, and the formist catch-all cannot do justice to them.

Ideological implication, to the degree that it is ideological, also fails to meet the explanatory requirement of answering to a puzzlement. In fact, a narrative (or other piece of writing) becomes ideological only after any puzzles have been solved. White's own account of Marx's “grammar” illustrates the point unambiguously. If Marx had a puzzle about the nature of human existence, then—according to White—he solved it out of pieces that were not historical. Rather, they were elements of the human condition: “the impulse to satisfy needs,” “the capacity to reproduce,” and “the constitution of modes of production.”³⁶ Only thereafter does White's Marx write his “histories” through these principles. His histories show these principles in action: the essence of ideological writing or, less kindly, propaganda.

None of these criticism invalidate White's categories as rhetorical types. They only reduce their scope from an overarching and all-encompassing theory of historical and fictional narrative to simply one among many typologies applicable to narrative. Whether the well-worn derivative quartet of category quartets provides the best set of ideal types for the analysis of rhetorical functions is open to question, just as are the types within the other typologies.

Without the excess baggage of preconceptual and precritical deep poetic consciousness at work, the rhetorical typology has some promise of showing what is presupposed by each identifiable type. Rhetoric's constitutive function of concept selection with respect to object identity and relationship characterization does not require appeal to poetry.³⁷ Instead, it requires attention—as noted earlier—to function, purpose, and achievement.

Moreover, not everything constitutive is either preconceptual or presuppositional. Some constitutive matters concern simply the formative rules of an enterprise.³⁸ Unlike formal games, such as baseball and chess, many human activities have no rule books. Nonetheless, we can explicate rules that govern activities, or as Rawls calls them, practices. Indeed, without the practice of baseball, one may throw a ball, run, etc., but one may not balk or steal a base.³⁹ The act of stealing a base or balking presupposes the rules that constitute the practice (or game) of baseball, but the game or practice does not presuppose a particular set of rules. To enter the game is to enter the rule-governed practice which, in part, gives sense to the very language within which we describe objects, events, and relations. Just as we may consciously and for reason decide to enter a game of baseball, so we may consciously and for reason enter into the task of creating a tragic, mechanistic, radical narrative (even if we do not know these particular type-words).

What gives the enterprise an air of unconscious (or deep conscious) choice is that on occasion, we may give this aspect of the work no thought at all. We just write our narrative and the result turns out to be tragic, mechanistic, and radical. No recourse to hidden consciousness is required by such situations. Such narratives may emerge because over the directions that each category represents, they are the only directions one has habitually used, that one has been taught, or that one knows. Alternatively, the subject matter itself might suggest them, or our theoretical commitments might dictate them.

In what sense, then, do the rules of rhetoric prefigure the resultant narrative? Were all narratives the consequence of active, rational, strategic decisions, then prefiguration would constitute a temporal term. Since they are not, White resorts to an appeal to the deep consciousness of the narrativist to perform a poetic act. Of course, no act at all is required (even if one may on occasion be performed because such an act is possible). Rather, the elements of prefiguration are no more than the rules presupposed by the nature of the particular type of narrative one writes, that is, the rules regarding emplotment, preferred connective, and (if any) ideological implication. Whether we intend to write or have already written a narrative showing how past events led up to the present situation, the rules specify the use of narrative connectives such as “cause-effect,” “forcing,” “influencing,” “prompting,” “reacting to,” and a host of others.⁴⁰

To the degree that we can adduce and formulate the rules for making narratives with certain functions, we cannot classify such rules as preconceptual, but only as presupposed. The preconceptual indicates a level of formative operation that determines the logic of concepts and conceptions. To see with Marx the human being or human society as essentially economic alters what it is to be human and social. Any analysis of the change presumes (or discovers) a premarxian concept of what is human and what is social and also tracks the alterations necessitated by Marx's works (and by postmarxian writings as well). Such an analysis may be historical in the sense of tracing the significant changes in the ways in which we use these concepts and find them meaningful before and after Marx. Equally, such an analysis in the heat of debating Marx may also serve to provide reasons for favoring adoption or rejection of the proposed changes. Sometimes we make changes in concepts by working out the implications of proposals that do not seem initially conceptual, as in the continuing western social unrest that accompanies the development of what follows from declaring all men or humans to be equal. As well, we emplace such changes through convincing rhetoric, the sanction of “science,” and the refusal to consider any other foundational principles, as did Marx in his writings and debates. 1989 events in China suggest that we may even enforce a narrative account by social ad baculum measures. To the degree that any such account may include conceptual proposals, an enforced proposal may revise the very way in which we conceive experience itself. Missionaries backed by soldiers showed remarkable success in making for some the inconceivable believable.

CONCLUSION

What emerges most clearly from this examination of narrativist fallacies is that, contrary to White, the prefiguring of a narrative is not a matter of deep consciousness or even unconsciousness. Nor is it necessarily a poetic act—nor even an act. Shaping a narrative may be simply (but not merely) the process of making strategic decisions about choosing the narrative form that will best accomplish the purpose of the narrative. Whether or not chosen, the very act of writing a narrative presupposes (within the rhetorical typology) rules of emplotment, connection, and implication. Except for cases of conceptual or conceptualization proposals, none of this activity need occur at a preconceptual level.

If anything, structuralist linguistics obscures the merits of the rhetorical typology by shifting attention to unnecessary special appeals. We gain nothing in appreciating the artistry of Michelet by subsuming all of history under the aesthetic. Perhaps we may even lose a good bit, as we confuse the power to convince readers alogically, i.e., rhetorically, with the aesthetic, which includes much that is mundane and not designed to convince. As with history, only the rarest of art has ever succeeded in altering the way that even some people see and experience the world.

Moreover, tropic theory and typology has and requires other than artistic or aesthetic grounds for determining the success of its ventures, especially in the realm of tracking the history of literary and historical styles. Nothing in linguistic theory can decide among the various typological foundations for portraying the sweep and shifts of style, foundations which include Jakobson's diadic metaphoric-metonymic poles, White's more kantian preference for quartets, or Brady's more hegelian triadic and dialectical scheme.⁴¹ Independently of each other, White and Brady adopt their larger collections of basic tropes for their power to overcome anomalies of fit created by a mere dualism. Contrary to normal applications of fundamental theory, structuralist poetics and philosophy of history find that the successful fit of tropic categories with the facts of literary and historical style (respectively) provides the criteria for selecting and evaluating theoretic foundations.

In the end, one common core to history and to fiction, as they are often but not exclusively written, is narrative discourse. Without recourse to confusions between history and fiction, especially with respect to their epistemic status, the typologies suggested here largely correct some of the errors inherent in White's misentropic treatment of narrative's rhetoric. One fears, however, that fads of theory—like common colds—run their course in their own time. The best cure may not be therapy but patience.

Notes

1. Paul Ricoeur, Time and Narrative, Vol. 2, Chicago, Chicago University Press, 1985, p. 7. See White's appreciation of Ricoeur in his essay “The Metaphysics of Narrativity” in The Content of the Form, Baltimore, The Johns Hopkins University Press, 1987, pp. 169–184.

2. Hayden White, Metahistory, Baltimore, The Johns Hopkins University Press, 1973, p. 6.

3. Hayden White, Tropics of Discourse, Baltimore, The Johns Hopkins University Press, 1978, p. 82.

4. Ibid., pp. 81–100.

5. Ibid., p. 121.

6. Ibid., p. 122.

7. Bruce Waters, “The Past and the Historical Past,” The Journal of Philosophy 52, 1955, pp. 266–269.

8. Margaret Macdonald, “The Language of Fiction,” in Philosophy of Art and Aesthetics, ed. F. Tillman and S. Cahn, New York, Harper & Row, Publishers, 1969, p. 622.

9. White, The Content of the Form, p. 20.

10. See my “Collingwood: Action, Re-enactment, and Evidence,” The Philosophical Forum, 2, 1970, pp. 68–89, for a review of Collingwood's ruminations on questions, evidence, and answers in history.

11. See my Fictional Narrative and Truth, Lanham, MD, University Press of America, 1984, especially pp. 111–126.

12. White, Metahistory, p. x. See also White's more extensive remarks on pp. 30–31.

13. Ibid., pp. 31–33, n. 13.

14. Ibid., pp. 277 and 279.

15. Ibid., p. 332.

16. Ibid., p. 343.

17. Ibid., p. 368.

18. Ibid., p. 374.

19. Ibid., p. 434.

20. Friedrich Nietzsche, Beyond Good and Evil, trans. W. Kaufmann, New York, Vintage Books, 1966, Paragraph 1.

21. Ibid., Para. 4.

22. White, Tropics of Discourse, p. 122.

23. For example see Nietzsche's remark, “At this stage artistic urges are satisfied directly, on the one hand through the imagery of dreams. … ; on the other hand, through an ecstatic reality …” The Birth of Tragedy, trans. F. Golffing, Garden City, NY, Doubleday Anchor Books, 1956, p. 24.

24. Macdonald, “The Language of Fiction,” p. 625. White's own attempt to deal with differences between everyday or historical treatments of objects and stories and of fictional counterparts is too weak to require noting within the text. He bases his remarks on the distinction between a historian “finding” stories and a fiction writer “inventing” stories. Then he notes how much “invention” plays a role in the historian's attempt to tell a story. If one begins with a silly distinction, such as that between finding and inventing, of course the result can get no further. Metahistory, pp. 6–7.

25. Curiously, for White's Nietzsche (although perhaps not Nietzsche's Nietzsche), the dream urge is functional (“a human need to flee from reality”) and not free. And freedom from function remained to Nietzsche's day a prerequisite for the artistic and the aesthetic.

26. White, The Content of the Form, p. x.

27. For remarks on the former problem, see my “Understanding Narrative Theory,” History and Theory, 25, 1986, pp. 65–70; and on the latter question, see Fictional Narrative and Truth, pp. 116–119.

28. See John Dewey, Logic: The Theory of Inquiry, New York, Henry Holt and Company, 1938, pp. 3–22.

29. Arthur Danto, Analytical Philosophy of History, Cambridge, The University Press, 1965), pp. 122–123.

30. R. G. Collingwood, An Essay on Metaphysics, Oxford, Clarendon Press, 1940, p. 31. See also my Concepts, Events, and History, Washington, D.C., The University Press of America, 1978, pp. 87–88.

31. See William Dray, Laws and Explanation in History, Oxford, Oxford University Press, 1957, especially Chapters V and VI, as well as Chapter 2 of his Philosophy of History. Englewood Cliffs, NJ, Prentice Hall, 1964, for a summary of his work to 1964. Note his remarks on explanatory and descriptive histories, pp. 29–32, and compare them to Frank Ankersmit's idea of a narratio in Narrative Logic, The Hague, Martinus Nijhoff, 1983. See also Dray's “Colligation Under Appropriate Conceptions” in Substance and Form in History, ed. L. Pompa and W. Dray, Edinburgh, University of Edinburgh Press, 1981, pp. 156–170.

32. John Passmore, “Explanations in Everyday Life, in Science, and in History,” in Studies in the Philosophy of History, ed. G. Nadel, New York, Harper and Row, Publishers, 1965, pp. 17–18.

33. Michael Scriven, “Truisms as the Grounds for Historical Explanations,” in Theories of History, ed. P. Gardiner, New York, The Free Press, 1959, p. 446. Accuracy involves many things in addition to truth.

34. Metahistory, p. 12.

35. Ibid., pp. 14–21.

36. Ibid., pp. 297–299.

37. Ibid., p. 31.

38. The notion of constitutive rules, of course, owes to the work of Rawls, “Two Concepts of Rules,” The Philosophical Review, 64 (1955), pp. 25–29) and Searle (“How to Derive ‘Ought’ from ‘Is',” The Philosophical Review, 73 1964, pp. 55–56).

39. Rawls, “Two Concepts of Rules,” p. 25.

40. Mercifully, the rhetoricians of history do not, as did the covering law theorists, try to reduce all narrative connections to causes and effects. Unmercifully, they subject us to the monotony of metonymy.

41. Compare the justifications for altering Jakobson's dualism in White, Ibid., pp. 32–33 (plus the preceding pages devoted to quartets of differing levels of “explanatory affect.”) and in Patrick Brady, Structuralist Perspectives in Criticism of Fiction. Berne, Peter Lang, 1978, pp. 105–106. Despite their interests in differing literatures and centuries, the explanation for differences between the White and Brady schemes lie elsewhere. Whether or not White's more synchronic orientation toward the 19th century as a sort of unit and Brady's diachronic interests in the movements within 18th century literature hold the key to their respective kantian and hegelian perspectives is beyond the scope of this essay to decide. However, it is of interest to note that Brady does not reject the Jakobson dualism for the purposes of criticizing individual works, but only for the sake of placing the Rococo (contra Durand) between Baroque Classicism and Romanticism in a temporal schema of “affirmation/negation/reversal.” Compare Ibid., pp. 100–101.