On Narrative Theory and Its Uses

  • David Herman (Hg.): The Cambridge Companion to Narrative. Cambridge: Cambridge University Press 2007. 310 S. Paperback. USD 31,99.
    ISBN: 978-0-521-67366-2.

There were times in which narrative theory was seen as a highly specialized branch of literary studies, concerned with the classification and analysis of formal features of narrative fiction. Those were the days of what are now often called »classical approaches« in narratology – associated with projects like A.J. Greimas’ typology of actant roles, or Gérard Genette’s development of categories for the representation of time. Since the 1980s, however, interests and goals have shifted, and there have been several more or less closely related developments both inside and outside literary studies. For one thing, ›narrative‹ itself has come into the focus of attention as a general strategy by which humans order their experiences and make sense of them. Secondly, this reassessment of the concept of narrative has opened up interdisciplinary perspectives. No longer do the practitioners of narrative theory see their discipline mainly as a tool box for a descriptivist and ahistorical analysis of narrative fiction. Instead of focussing solely on the texts, they stress the importance of context for their endeavour: on the one hand, »postclassical« approaches emphasize how narratological concepts themselves are always historically and ideologically situated. On the other hand, they suggest that the expertise of narrative theory can be exported to other disciplines and thereby offer a vital contribution to a whole range of interdisciplinary projects.


In the new Cambridge Companion to Narrative, edited by one of the most prominent proponents of »postclassical« narratology, David Herman (Ohio State University), this contemporary understanding of narrative theory, its aims and functions is not only presented and discussed, but organizes the structure of the volume as a whole. The project is highly ambitious: once questions of the definition of narrative have been treated (Part I), the volume aims to combine an introduction into the basic terms, concepts and methods for the analysis of narrative fiction (Part II) with an exploration of narrative in different media (Part III). Finally, exemplary forays into »contexts of narrative study« (Part IV) highlight the potential of cross-fertilization between narrative theory and fields of study like »gender«, »rhetoric/ethics« and »identity/alterity«. Moreover, the book addresses a double audience: it is intended as an introduction to narrative study for beginners, but also as »a unique contribution to the scholarship on narrative« (p. 4), and a guide for advanced scholars in the field. The line-up of contributors attests to the international orientation both of narrative theory as a field of study and of the volume itself, featuring experts from the U.S., England, Germany, Belgium, Canada and Finland.


The individual articles all roughly follow the same principles, which provide a consistent blueprint for the volume as a whole while also allowing for the large degree of flexibility needed to do justice to the wide range of topics covered. Each of the eighteen contributions begins with a brief explication of the central concept or set of issues, and most sketch a diachronic dimension of their topic by including information on the history of this concept, its functions in literary contexts, or, as the case may be, the development of the relevant branches of research. Almost all of the chapters illustrate their main points by applying them to a concrete example. The book also provides a detailed glossary, a further reading section and an index.


As a whole, the Cambridge Companion to Narrative is highly recommendable. It provides a comprehensive survey of recent trends in the study of narrative and serves as a useful introduction to its fields of study.


Part I: On Defining »Narrative«


Part I of the book contains an introduction by David Herman and a chapter on the definition of narrative by Marie-Laure Ryan. Herman gives a brief survey of the companion’s aims and scope as well as some advice for using the book. Moreover, he distinguishes recent trends in narrative scholarship and addresses the definition of narrative.


According to Herman, we cannot expect the definition of narrative to be framed by necessary and sufficient conditions that allow for a clear-cut distinction between narrative and non-narrative texts. Rather, narrative allows for both »membership gradience« and »category gradience«: »A given text can be a more or less central instance of the category, and less central instances will be closer to neighboring text-type categories (descriptions, lists, arguments, etc.) than will be prototypical instances« (p. 8–9). Accordingly, there is, for instance, »descriptized narration« which is located at the border between narration and description. Herman suggests that

core or prototypical instances of narrative represent or simulate (i) a structured time-course of particularized events which introduces (ii) disruption or disequilibrium into storytellers’ and interpreters’ mental model of the world evoked by the narrative (whether that world is presented as actual, imagined, dreamed, etc.), conveying (iii) what it’s like to live through that disruption, that is, the »qualia« (or felt, subjective awareness) of real or imagined consciousnesses undergoing the disruptive experience (p. 9). 1

Each of these conditions is meant to exclude certain text types that do not belong in the narrative category. – As this conceptualization not only reflects recent trends in narrative theory but, more generally, definitions of narrative are taken to provide the foundation for the discipline as a whole, we will consider it in some detail here. 2


First, we have some doubts as to how Herman applies the very idea of prototypical concepts to narrative. Items that fall under a prototype concept, as Herman rightly implies, only tend to instantiate the prototype’s features. Thus no single feature is taken to be necessary for the application of the concept. Herman, however, seems to think that the three features mentioned in his definition (roughly, temporal order, disruption and the conveyance of experiential quality) are necessary after all – his discussion of the features is framed in terms of necessary and sufficient conditions, which appears to be alien to the very idea of prototype concepts. 3


Second, Herman maintains that what counts in favour of thinking that narrative is a prototype concept is the fact that there are both membership gradience and category gradience. This, however, hardly constitutes a strong reason, for both phenomena can be reconciled with the idea that the concept of narrative is governed by necessary and sufficient criteria and hence is not a prototype concept. Most concepts that are governed by necessary and sufficient criteria allow for borderline cases insofar as the concepts employed in the definiens are themselves to some extent vague. Moreover, some of the texts that appear to be borderline cases of narrative may be taken to be mixed cases that combine narrative and non-narrative elements. Neither borderline cases nor mixed cases compel us to think that narrative is a prototype concept.


We believe that in order to settle the question of what type of concept ›narrative‹ may be taken to be, one needs to consider both the context of the definition and its purpose. What needs to be kept apart is whether one attempts to give a lexical definition of a concept that is widely employed in everyday usage, or whether one is trying to shape such a concept in order to make it suitable for a scientific purpose. This last operation will inevitably involve some stipulation, that is, it yields a new concept, and it follows its own rules and criteria. 4


Marie-Laure Ryan takes up this idea when she points to the fact that the term »narrative« is hardly used outside academic contexts; rather, we are dealing with a technical term »designed by narratologists« (p. 32). Ryan specifies a »toolkit« including eight features for »do-it-yourself definitions« of ›narrative‹ (p. 30). They are:


1st Narratives must be about a world populated by individuated existents;


2nd This world must be situated in time and undergo significant transformations;


3rd The transformations must be caused by non-habitual physical events;


4th Some of the participants in the events must be intelligent agents who have a mental life and react emotionally to the states of the world;


5th Some of the events must be purposeful actions by these agents;


6th The sequence of events must form a unified causal chain and lead to closure;


7th The occurrence of at least some of the events must be asserted as fact for the storyworld;


8th The story must communicate something meaningful to the audience (p. 29).


The toolkit approach draws attention to the fact that once we conceive of ›narrative‹ as a technical term we are likely to get a bundle of different – and partly stipulated – concepts of narrative. 5 However, Ryan combines this useful suggestion with the claim that narrative is a prototype concept and that there are degrees of narrativity. In fact, the toolkit list is meant to explain both the prototypical structure of the concept and its gradability. Our quarrels with this idea stem from the following sources:


Ryan’s suggestion is that the degree of narrativity depends on how closely a text resembles »prototypical cases that everybody recognizes as stories« (p. 28). This in turn can be measured by how many of the conditions mentioned in the toolkit are fulfilled. However, the fact that a text may resemble a prototype to different degrees does not necessarily mean that ›narrative‹ is a grade term. Consider the term ›bird‹, which is often cited as a prime example of a prototype concept. The fact that robins are clearer instances of ›birds‹ and resemble the prototype more closely than emus does not make ›bird‹ a gradable concept. An animal either is a bird or it is not. Generally speaking: gradability does not follow from prototypicality. 6 Nor does the fact that people have more or less grave difficulties in classifying a given text as narrative or non-narrative, depending on how many (and which) of the criteria the text fulfils (and on how clearly it does so), imply anything about whether a concept is gradable or not. People may be inclined to say that ›this text is more narrative than that‹, just as they may be inclined to say that a robin is more ›birdy‹ than an emu, but this might just be a somewhat awkward way of putting it, and maybe it is not all too revelatory with regard to the concept’s logical structure. 7 Consider, finally, a term that is truly gradable, like ›warm‹. Such terms characteristically come as adjectives that allow for comparatives. Something can be warm, or warmer, or warmest among a group of other things. The question of whether ›warm‹ is a grade term has nothing to do with our difficulties in deciding whether something is warm or not, and it is not a matter of how many conditions for warmth are fulfilled in a specific case. 8


So, what does follow from the fact that we have a toolkit of criteria which a given text may fulfil? It follows, we think, that we have many different concepts of narrative, depending on which of the criteria we use for the definition of the concept at hand.


Part II: »A Starter Kit«


Part II (»Studying Narrative Fiction: A Starter Kit«) contains chapters on »Story, Plot, and Narration« (H. Porter Abbott), »Time and Space« (Teresa Bridgeman), »Character« (Uri Margolin), »Dialogue« (Bronwen Thomas), »Focalization« (Manfred Jahn), and »Genre« (Heta Pyrhönen). These chapters provide very useful introductions into the problems at hand, often delineating and explaining scholarly controversies that evolved around some of these concepts. To name but one especially noteworthy example, Manfred Jahn’s chapter on focalization succeeds in providing a survey of the concept’s history, its elaboration in ›classical‹ structuralist narratology, more recent developments and refinements in ›postclassical‹ approaches, and a case study. The chapter closes with »a task sheet for analyzing focalization« (p. 105–106). A similar addition would also have been welcome in some of the other chapters, as it serves as a practical guide for the analysis of narratives. Margolin’s chapter, for example, veers in the other direction by stressing the philosophical side of fictional characters and, although illuminating in itself, does not provide much information as to character analysis. Abbott’s chapter then nicely steps in as his case study is on »trying to understand Heathcliff« (p. 45–49). In this as in some other cases the connections between the individual chapters could have been more clearly marked by the insertion of cross-references.


Part III: Other Media


Part III (»Other Narrative Media: A Selection«) leaves the territory of narrative fiction and explores the potential of narrative theory for the analysis of »Conversational Storytelling« (Neal R. Norrick), »Drama and Narrative« (Brian Richardson), »Film and Television Narrative« (Jason Mittell), and »Narrative and Digital Media« (Nick Montfort). It is impressive to see how in the restricted space allotted to each of the individual chapters – none of them exceeds 15 pages – they all manage to convey a sense of the central issues connected with each subject, and especially the modifications to narrative theory necessary to deal with them (Jason Mittell’s chapter is particularly successful). We would have welcomed a brief editorial introduction to this part, which could have made explicit the criteria of selection and addressed the underlying concept of ›medium‹. It seems that while this term is adequate to capture differences between digital works and film/television, the shift in focus to everyday conversations might be more satisfactorily described in terms of, for instance, disciplinary differences, while ›drama‹ is, first of all, a different genre (like ›poetry‹, which is also left aside). This is hardly a major point of criticism, as the selection does succeed in illustrating the wide range of application, but the lack of clarity surrounding the notion of ›medium‹ is a stumbling block in Brian Richardson’s chapter »Drama and Narrative«, in which the differences between the level of the dramatic text and that of performance are not as explicit as one could wish. Nick Montfort’s chapter on »Narrative and Digital Media« presents additional »dimensions« for the analysis of computer-based narratives and thus offers the prospect of insights into new options of multimedia experiences of narrative. It comes as a bit of a disappointment, however, that the main example used in this chapter is that of a purely text-based computer game created in 1985, which passes over some of the interesting questions about the role of graphic and sound effects as well as multi-player virtual environments. These quibbles notwithstanding, Part III offers a compact and helpful introduction to the inter-medial dimensions of narrative study.


Part IV: Further Contexts


The fourth and last section (»Further Contexts for Narrative Study«) can in a way be regarded as the linch-pin of the volume: in presenting a wide range of fields of inquiry in which issues of narrative can play a central role, it puts to the test the practicability of the categories as they were introduced in Part II, and poses the question of the cultural relevance of narrative inquiry in general. The contexts explored are »Gender« (Ruth Page), »Rhetoric/ethics« (James Phelan), »Ideology« (Luc Herman/Bart Vervaeck), »Language« (Michael Toolan), »Cognition, Emotion, and Consciousness« (David Herman) and »Identity/alterity« (Monika Fludernik). While three of the chapters foreground issues that play an important role in the telling of narratives as well as in their analysis (»Gender«, »Ideology« and »Identity/alterity«), Michael Toolan’s chapter on »Language« is more concerned with the contribution another discipline – in this case linguistics – can make to the study of narrative texts. James Phelan’s »Rhetorics/ethics«, in turn, deals with the methodical and thematic consequences of a specific understanding of narrative as an act of communication, while David Herman’s »Cognition, Emotion, and Consciousness« presents a new angle on the study of consciousness in narrative – in terms of the analysis of narrative fiction’s potential to give inside views of characters’ minds and emotions, but also in terms of a tracking readers’ mental processes as they »build up a model« (p. 251) of these cognitive and emotional states. It would have been very interesting to read some comments about the relation between and potential compatibility (or incompatibility) of the various approaches – again, we missed an editorial guide explaining the selection and the relation between the individual approaches. For example, it seems obvious that there must be some kind of connection between »Rhetoric/ethics« and »Ideology«, but the links as well as the differences between the two approaches are not made explicit. While this may make it hard especially for beginners to form a clear idea of the discipline, its tendencies and paradigms and how they are linked, Part IV still makes for fascinating reading, as on less than 100 pages it manages to introduce a large range of thought-provoking insights into contemporary practices of narrative study.



A shorter version of this definition can be found in the glossary, p. 277–8.    zurück
For the claim that the definition of narrative may serve as a foundation of narratology see Tom Kindt / Hans-Harald Müller (eds.): What Is Narratology? Questions and Answers Regarding the Status of a Theory. Berlin, New York: Walter de Gruyter 2003.    zurück
What needs to be carefully distinguished is (a) what features a given prototype concept necessarily has in order to be this prototype concept (that is, what features secure the identity of the concept, or its intensional structure), and (b) what features an item must have in order to fall within the prototype’s extension. In his discussion, Herman apparently blurs the two questions. – There are also some general shortcomings of the prototype theory of concepts. For one, there is doubt as to whether prototypical theories can explain category membership at all. We will not explain these worries in any detail here but rather refer to the excellent introduction to the subject in Stephen Laurence / Eric Margolis: Concepts and Cognitive Science. In: S.L. / E.M. (eds): Concepts. Core Readings. Cambridge, London: The MIT Press 1999, p. 3–81, see especially p. 27–43.   zurück
For the distinction between lexical and stipulative definitions see Richard Robinson: Definition. Oxford: Oxford University Press 1954.    zurück
As one can see, Ryan’s criteria only partly overlap with Herman’s criteria mentioned above. The very fact that there are two different definitions in the Companion indicates that there is more than just one concept. However, the two definitions compete only insofar as they are taken to be lexical definitions of the same concept. Nothing prevents us from stipulating many different concepts of narrative.    zurück
Nor, of course, does prototypicality follow from gradability. There are many gradable concepts which do not exhibit a prototypical structure.    zurück
In the case at hand, people’s difficulties in classification can be explained by the fact that some of the defining features of birds are not easily observable. There’s more to birds than meets the eye.    zurück
The gradability of a concept typically is not a matter of whether certain conditions that are necessary and sufficient for the application of the concept are fulfilled, but rather depends on the fact that some constitutive feature or other allows for degrees. If this is true, then even if ›narrative‹ were a gradable concept, its gradability would presumably call for a different explanation than the one Ryan provides.    zurück