Robert Piercey
The concept of narrative matters to hermeneutics for two reasons. First, whatever else hermeneutics is, it is the practice and theory of the interpretation of texts. We should expect it to reflect on the nature of these texts, and, since many texts are narratives, we should expect hermeneutics to have something to say about the nature of narrative. But narrative also matters to hermeneutics for a deeper reason. Hermeneutics is not just the interpretation of texts; it is also a philosophical outlook, one that sees the topics of meaning and interpretive understanding as central to the business of philosophy. Like any philosophical outlook, hermeneutics in this sense is driven to reflect on itself, posing questions about its nature and the status of its claims. The questions it poses are analogous to questions that, over the past several decades, have been posed concerning the nature of narrative. Theorists of narrative have raised definitional questions about what narratives are and how they differ from other things. They have also raised epistemic questions about the kind of insight narratives offer, and ontological questions about the relation between narratives and reality. In short, the study of narrative presents specialized versions of many of the questions that have been raised about hermeneutics as a philosophical outlook. Moreover, the answers that have been proposed to these questions offer hints about how to answer similar questions about hermeneutics.
Despite its importance, the concept of narrative has received less attention from hermeneutical philosophers than many other topics. It was generally not an object of explicit study during the nineteenth century, and it received only slightly more attention for much of the twentieth. Gadamer, for instance, makes a few scattered remarks about the activity of narrating (Gadamer 1986, 67, 92), but gives nothing like a fully developed theory of narrative. Narrative came to the forefront of hermeneutics only in the late 1960s, as hermeneutical philosophers began to engage with and react against structuralism. Structuralists such as Roland Barthes attached tremendous importance to the stories through which cultures interpret themselves. Narrative, Barthes claims, is “international, transhistorical, transcultural: it is simply there, like life itself” (Barthes 1982, 251–252). Confronted with the number and diversity of these stories, structuralism adopted an immanent approach to narrative. Bracketing all questions about the reality to which narratives refer, it focused on their formal features alone: the linguistic elements that stories contain and the logic that governs their interaction. Structuralism maintained that questions about authorial psychology are irrelevant to the study of narrative, as are questions about a text’s history and its reception by its audience. From a structuralist perspective, “narrative does not show, does not imitate … ‘What takes place’ in a narrative is from the referential (reality) point of view literally nothing; ‘what happens’ is language alone” (Barthes 1982, 295).
Many hermeneutical philosophers were willing to believe that the structuralist approach offered important insights, and that the illumination generated by narrative has something to do with its formal features. Paul Ricoeur articulated this attitude well when he said that the nature of language “makes not only possible but necessary the mediation of understanding by explanation, of which structural analysis constitutes the most remarkable realization” (Ricoeur 1991, 130). But no hermeneutical thinker could accept a purely immanent study of narrative. Hermeneutics is distinguished by its conviction that interpretation and understanding are historical through and through. They are not merely attained at specific points in history; each “is, essentially, a historically effected event” (Gadamer 1992, 300). Hermeneutical thinkers also tend to insist that understanding is completed only in application: that it “always involves something like applying the text to be understood to the interpreter’s present situation” (Gadamer 1992, 308). A purely immanent approach is silent about the histories that produce narratives and about their reception by their audiences. For these reasons, a different, distinctively hermeneutical approach to narrative seemed called for. Such an approach would define narrative in terms of something more than its formal features. It would explain narrative’s epistemic significance by examining the cognitive acts through which narratives are received and appropriated. And it would explore the ways in which narratives impinge on reality, both by representing it and by transforming it. Above all, a hermeneutical approach would insist that narrative is not a “closed system,” but rather “opens … onto the world” (Ricoeur 1991, 131).
In developing their approach to narrative, hermeneutical philosophers drew heavily on resources from phenomenology. Phenomenology offered a way to counter the hermeticism of the structuralist approach, by emphasizing the constituting acts of consciousness that put thinkers in contact with narrative structures. Roman Ingarden was an important model in this attempt. In the 1930s, Ingarden used the techniques of Husserlian phenomenology to describe the experience of reading literature and, in doing so, made several promising suggestions about how to understand narrative. According to Ingarden, a literary work of art is a complex, multilayered intentional object that “has the source of its being in the creative acts of consciousness of its author and its physical foundation in the text set down in writing” (Ingarden 1973, 14). It is not reducible to either: the work is intersubjective and ideal, and transcends all the intentional acts through which subjects relate to it. At the same time, the literary work is essentially incomplete. It contains “places of indeterminacy” (Ingarden 1973, 13) that must be filled in by its readers, and this filling-in is “not sufficiently determined by the determinate features of the object” (Ingarden 1973, 14). A literary work’s existence is fully realized only when it is read, but different readers realize it in different ways. Ingarden’s view of the literary work proved an important inspiration for hermeneutical accounts of narrative. Both Gadamer (1986, 27) and Ricoeur (1988, 167–169) drew on Ingarden in attempting to explain how narratives can be ontologically independent of their audiences but still completed only in application. Another important inspiration for hermeneutical accounts of narrative was Wolfgang Iser. Iser’s phenomenology of reading also explored the conscious acts that put readers in contact with literary works. However, Iser paid special attention to the temporality of these acts. According to Iser, a literary work differs from many intentional objects in that it “can never be perceived at any one time” (Iser 1978, 108). Readers are forced to adopt a “wandering viewpoint” on the text, grasping it in a series of phases, “each of which contains aspects of the object to be constituted, but none of which can claim to be representative of it” (Iser 1978, 109). As a result, reading is an exercise in “consistency-building” (Iser 1978, 118). Readers try to synthesize the phases of their reading into a “consistent interpretation,” and such an interpretation “cannot be exclusively traced back either to the written text or to the disposition of the reader” (Iser 1978, 119). Its goal is a “consistent gestalt” that “endows the linguistic signs [of the narrative] with their significance” (Iser 1978, 121). Iser’s emphasis on consistent Gestalten proved seminal for hermeneutical accounts of narrative, since it highlighted the link between literary structures and the experience of temporality. It also suggested a way of explaining how narratives transform their elements by bestowing on them the significance that comes from being part of a plot.
In short, hermeneutical approaches to narrative are distinctive in a number of ways. They do not dismiss the insights offered by a structural analysis of narrative. But they combine these insights with strategies drawn from phenomenology, and often other influences as well. Balancing all of these influences is a challenge, and this challenge gives rise to a number of difficult questions.
The first such question concerns the nature of narrative. What is a narrative, and what distinguishes narratives from other things? There are a few claims about these matters that nearly everyone would accept. One is that narratives are symbolically mediated representations of actions. “Action” is a crucial term here. Mountains, glaciers, and other natural phenomena may change over time, and these changes may be described in speech or writing. But we would probably not call such a description a narrative. Narratives represent not just any events, but actions: changes intentionally brought about by conscious beings and capable of being justified with reasons. As for the notion of symbolic mediation, it is closely linked to that of language, but not identical with it. While narratives typically represent actions with spoken or written words, not everyone would agree that narratives must be linguistic. Some would argue that selves or other extra-linguistic entities deserve to be called narratives as well. Another claim that is generally accepted is that narratives are more than mere chronicles. They are “significant” rather than “plain” (Walsh 1958, 31), to use W. H. Walsh’s terminology. A narrative is not just a list of actions, but a structure that shapes actions by placing them in a larger whole. Minimally, this larger whole is a structure with a beginning, middle, and end. By placing actions in this larger structure, a narrative gives them a significance they would not otherwise have. Finally, nearly everyone would agree that these criteria are not sufficient to make something a narrative, and that further conditions must be met. But there is little agreement about what these further conditions are.
A common suggestion is that narratives must contain causal links. The thought here is that the events described by a narrative cannot be entirely unrelated. At least some of them must be depicted as effects of other events in the narrative. This point is captured in E. M. Forster’s famous remark about plots: “The king died and then the queen died” is not a plot, but “The king died, and then the queen died of grief” is (Forster 1962, 87). Louis Mink echoes this point when he says that narratives describe events that stand in “causal connection with events already established in the narrative” (Mink 2001, 215). Mink adds, however, that only certain causal connections mark something as a narrative: namely, connections that are relevant to the story as a whole. We therefore need a criterion that distinguishes relevant links from irrelevant—a criterion Mink admits he cannot give. However, not everyone agrees that narratives must contain causal links. W. B. Gallie, for instance, argues that they need only display “logical continuity” (Gallie 2001, 43). By this, he means that it must be possible for us to retrospectively say of a later event in a narrative that it “required, as its necessary condition, some earlier one” (Gallie 2001, 43). To be sure, one of the ways an event can require an earlier event as its necessary condition is by being its effect. But there are other ways. The earlier event might merely have “occasioned or evoked” the later one, “or, at the very least, made it possible” (Gallie 2001, 43). Even if causality is the most common source of logical continuity, it does not seem to be a necessary feature of narratives.
While even this modest account of narrative faces problems, many accounts are considerably more ambitious. A moderately ambitious one appears in Ricoeur’s three-volume work Time and Narrative. Ricoeur sees narrative as one of the most fundamental ways human beings have of making sense of experience—a “primary cognitive instrument” (Mink 2001, 213), as Mink puts it. According to Ricoeur, what makes narrative so fundamental is that it humanizes time. Time is essentially aporetic. Our lives unfold in time, but, as Augustine discovered, we can say nothing satisfactory about what it is. Ricoeur thinks we respond to this aporia by emplotting events, relating them to earlier and later events in written or spoken stories. The result is a virtuous circle in which “time becomes human time to the extent that it is organized after the manner of a narrative; narrative, in turn, is meaningful to the extent that it portrays the features of temporal experience” (Ricoeur 1984, 3). For Ricoeur, therefore, narrative is a form of mimesis. It represents actions, and by doing so, it both depicts and makes intelligible the time in which those actions occur.
Ricoeur argues that this mimetic activity takes place in three spheres, and that as a result, “narrative” names three distinct but related forms of organization. The first, which Ricoeur calls mimesis1, is a structure that inheres in actions even before they are described in language. Actions are linked to the goals for which they are performed, and to the agents who perform them. They are also temporally related to earlier and later actions. Before we explicitly narrate them, actions are already embedded in implicit stories. Ricoeur’s second activity, mimesis2, is the construction of explicit stories: the conversion of actions into a plot. A plot combines, and thereby mediates that which it combines. It “draws a meaningful story from a diversity of events or incidents” (Ricoeur 1984, 65). It also yields a unified “temporal whole” (Ricoeur 1984, 66) by taking events that are scattered across time and grasping them together. The third stage of the mimetic process, mimesis3, corresponds to what Gadamer calls application. It is the process through which a narrative’s audience appropriates it, or makes it concrete by bringing it to bear on its own situation. Mimesis3 “marks the intersection of the world of the text and the world of the hearer or reader” (Ricoeur 1984, 71). This is the stage at which narratives open onto reality.
Hayden White gives an even more ambitious account of narrative. Like Ricoeur, White sees narrative as a primary cognitive instrument that bestows meaning on actions by incorporating them into larger structures. But in two respects, White goes further than Ricoeur. First, White argues that all narratives are explicit. There is no mimesis1, no story implicit in events before they are brought to language. All stories are told: they are “verbal fictions, the contents of which are as much invented as found” (White 1978, 82). This is not to say that we always recognize that a story has been invented. Historical narratives, for example, can seem so natural that we think we have read them off events themselves. But this is an illusion. The relationships which “appear to be inherent in the objects inhabiting the field” have actually “been imposed on the field by the investigator in the very act of identifying and describing the objects that he finds there” (White 1978, 95). White’s second departure from Ricoeur is his belief that the significance bestowed by a narrative results not from generic features of all narratives, but from features specific to a certain genre of narrative. Every story is a specific kind of story: “Romance, Comedy, Tragedy, [or] Satire” (White 1973, 143). Narratives in different genres structure events in very different ways, and in doing so, they bestow very different sorts of significance on those events. Events have to be “made into a story by the suppression or subordination of certain of them and the highlighting of others, by characterization, motific repetition, variation of tone and point of view, alternative descriptive strategies, and the like” (White 1978, 84). This is the case with the “true” narratives we call histories as well as the fictional narratives we call literature. Historical narratives employ “all of the techniques that we would normally expect to find in the emplotment of a novel or a play” (White 1978, 84). Again, they do not always let on that they are doing so. On the contrary, historical narratives tend to suggest that theirs is the only possible way of telling their particular story. (This is one of the key themes of White’s Metahistory.) But in White’s view, there are always other ways. The type of plot used to organize a series of events does not inhere in them, but is imposed by the narrator.
A second set of questions concerns the epistemology of narrative. Hermeneutical philosophers take seriously the fact that when we follow a story, we often have the sense that we have learned something, or gained some sort of insight into the events described. This insight seems to be of a different kind than that offered by the natural sciences, which illuminate events by subsuming them under general laws. Narratives, by contrast, are concerned with particular occurrences, and as Dilthey argued, they offer a sort of insight that has “individuals and their deeds as its elements” (Dilthey 1989, 158). But there is little agreement about what this insight is. Some argue that narratives explain, in that they reveal why certain events happened. Arthur Danto argues that stories just are explanations. “In certain contexts,” Danto claims, what people typically want and expect, when the need for explanation is felt, is simply a true story” (Danto 1985, 233). Danto’s view is most readily applied to historical narratives, but also has implications for fictional narratives. Following Aristotle, we might say that historical narratives explain what has happened, whereas fictional narratives explain the kind of thing that might happen. As for what it is about narratives that allows them to explain, Danto suggests that stories describe changes, and that narrative form simply is the form of an explanation of change. Stories tell us that X is A at time T1; that something happens; and that as a result, X is no longer A at time T2. Others reject Danto’s view, claiming it is contradicted by the experience of following a story. Gallie, for example, denies that we seek explanations from narratives. He argues that when we read or hear a story, we normally follow it easily and unreflectively. We simply “get” its point, and we do not seek explanations unless the story breaks down: “It is only when things become complicated and difficult—when in fact it is no longer possible to follow them—that we require an explicit explanation of what the characters are doing and why” (Gallie 2001, 41). In Gallie’s view, narrative yields insight, but there is almost nothing positive to say about the nature of this insight.
Ricoeur gives a particularly elaborate account of the insight narratives offer. He argues that historical and fictional narratives yield different sorts of insight, but by means of a common form. History seeks a true account of the past and, specifically, a true explanation of why certain past events happened. General principles are part of such explanations, but only a part. What history studies are “the thoughts, feelings, and actions of individuals in the specific context of their social environment” (Ricoeur 1984, 196, emphasis added). Historical narratives are constrained by a demand for documentary evidence. But they are still creative. Their plots are never simply read off of events themselves, though they are prefigured by them. Even seemingly non-narrative history, such as that offered by the Annales school, involves active, poetic configuration. This does not make historical narratives fictions. “Historical knowledge,” Ricoeur argues, “proceeds from our narrative understanding without losing any of its scientific ambition” (Ricoeur 1984, 92). Fictional narratives differ from historical ones in that they are not constrained by a demand for evidence. But they still reveal something about reality. Fiction is an “ethical laboratory” (Ricoeur 1984, 59) in which we encounter different experiments in living, different existential possibilities we might explore in our own lives. It presents the reader with imaginative variations on her own existence, and ultimately gives her “a self enlarged by the appropriation of the proposed worlds that interpretation unfolds” (Ricoeur 1991, 301). History and fiction offer different kinds of insight, but there is an “interweaving” (Ricoeur 1988, 245) between the two narrative forms. History tells the truth about the past while remaining poetic; fiction invents while nevertheless saying something true.
A final set of questions concerns the ontological status of narrative. At issue here is whether narratives present reality as it is, or invariably distort it. Such questions are most pressing for historical narratives, though versions of them can be raised about fictional narratives. Questions about the relationship between narrative and reality do not arise for structuralist accounts, which focus exclusively on narratives themselves and ignore what might lie beyond them. Nor do they arise for the thinkers Andrew Norman calls “anti-referentialists” (Norman 2001, 181)—those who claim narratives do not even purport to refer to reality, but perform some other function. John McCumber, for instance, flirts with the anti-referentialist view when he argues that historical narratives “cannot claim truth,” but are “organizing devices” validated by “the fact that they can order a diversity of material” (McCumber 2005, 82). Hermeneutical accounts of narrative, by contrast, generally accept that narratives at least purport to refer to reality, whether or not they succeed in doing so. To say that a narrative refers to reality is to say more than that the events it describes actually happened—or, in the case of fictional narratives, that its elements successfully claim truth in some other way. It is to say that narrative form can disclose something about the actual course of events: that the organization found in a narrative can reflect features of reality outside it. One’s view of this matter is strongly shaped by how ambitious one’s account of the nature of narrative is. The thicker one’s account of narrative, the more likely one is to see narrative form as a distortion imposed on reality rather than something continuous with it. The thinner one’s account, the more likely one is to think narrative can tell the truth about reality.
As we have seen, Hayden White’s account of narrative is particularly ambitious. He argues that every narrative belongs to a particular genre, and that a narrative’s genre gives it highly specific features. Each genre has its own ways of organizing events through “characterization, motific repetition, variation of tone and point of view, alternative descriptive strategies, and the like” (White 1978, 84). So it is not surprising that White is skeptical about narrative’s ability to reflect reality. He argues that every story is a tragedy, or a satire, or an instance of some other genre, but that reality is not tragic or satirical. To that extent, narratives are fictions, and when they suggest that their structure inheres in events themselves, they distort reality. Even the relationships described by historical narratives “exist only in the mind of the historian reflecting on them” (White 1978, 94). White is eager to add that narrative’s fictional character “in no way detracts from the status as knowledge which we ascribe to historiography. It would only detract from it if we were to believe that literature did not teach us anything about reality, but was a product of an imagination which was not of this world but of some other, inhuman one” (White 1978, 99). But the fact remains that, in White’s view, narrative form, as such, distorts reality. Stories are told rather than lived.
Other philosophers argue that narrative distorts reality in one sense, but in another sense reflects it accurately. They typically argue that narratives make explicit a structure that is implicit in events themselves, but that is transformed in being explicated. Alasdair MacIntyre advances such a view when he claims that “stories are lived before they are told,” and that “it is because we all live out narratives in our own lives and because we understand our own lives in terms of the narratives that we live out that the form of narrative is appropriate for understanding the actions of others” (MacIntyre 2007, 212). In particular, MacIntyre thinks the unity displayed by stories has its analogue in the unity of a human life. The difference is that lives do not, as it were, come pre-unified. Rather, we strive to make them unified, and it is often only in retrospect that we can see their unity. But in doing so, we are explicating something implicit in life, rather than imposing a form on a reality that lacks it. Ricoeur’s position is similar. Recall that, in his view, narrative is a form of organization through which actions are related to other actions, to the agents who perform them, and to the time in which those agents act. Sometimes this organization is made explicit through what Ricoeur calls mimesis2: the crafting of written or spoken stories. Sometimes it is left implicit. But the implicit structure that Ricoeur calls mimesis1 is a structure that actions really do have. It is not that of an explicit story; it might even be called a pre-narrative structure rather than narrative as such. But it is because actions have this structure that explicit stories are suited to describe them. Narrative does not distort events, but brings out something incipient in them.
An even stronger view is advanced by David Carr, who maintains that narrative “is not merely a possible way of successfully describing events; its structure inheres in the events themselves” (Carr 1986, 117). He presents his view as a foil to Ricoeur’s, which he considers too timid. Carr claims that the prefiguration that Ricoeur calls mimesis1 “is not itself narrative structure” (Carr 1986, 119), but merely an anticipation of it. When this structure is explicated through mimesis2, the resulting story “redescribes the world: in other words, it describes it as if it were what presumably, in fact, it is not” (Carr 1986, 120). Carr thinks reality possesses narrative structure in a much stronger sense. The key to seeing this is to recognize that the reality described by narratives is not that of physics. It is “human reality” (Carr 1986, 121): the sphere of human experience, intentions, and actions. Human reality does seem to display many of the features of written and spoken stories. The experience of time is mediated by the structures Husserl calls protention and retention, and these structures make time into an organized whole with beginnings, middles, and ends. Similarly, actions display something very much like narrative form. When we act, “we quite explicitly consult past experience, envisage the future, and view the present as a passage between the two. Whatever we encounter within our experience functions as instrument or obstacle to our plans, expectations, and hopes” (Carr 1986, 122). It is true that we cannot view our own lives from the standpoint of omniscient narrators. But this is as true of many written stories as it is of lived ones. In any case, “we are constantly striving, with more or less success, to occupy the story-tellers’ position with respect to our own lives” (Carr 1986, 125). The difference between narrating a written story and narrating a life is one of degree, not kind.
Narrative is difficult to define because the notion can be used to refer to many different things. However, its different meanings all orbit around a core meaning connected with the symbolic representation of action. The insight offered by narrative is hard to characterize because stories are hybrids that explain without subsuming. But few would dismiss the epistemic pretensions of narrative altogether. Even White, who calls narratives verbal fictions, thinks these fictions can nevertheless say something true. Finally, the relationship between narratives and reality is difficult to clarify, because narratives describe an order that inheres in things themselves but that is transformed in being brought to language. However, few would deny that narratives disclose reality in some way, even if it is a challenge to say what that reality is. In short, narrative is a thoroughly heterogeneous notion, but in that way it reflects the necessary, productive heterogeneity of hermeneutics itself.