Heterocosmica: Fiction and Possible Worlds - Review

Lubomir Dolezel. Baltimore and London: The Johns Hopkins UP, 1998. xii + 339 pp. $45.00 hardcover.

The timing of the publication of this book tells a deceptive story. Heterocosmica appeared in the wake of a number of other books on the application of possible worlds theory to literary semantics (Pavel, Ronen, Maitre, Semino, and myself), but none of these books (with the possible exception of Pavel's) would have been what it is, nor arguably written at all, without the inspiration provided by Dolezel's earlier work in this area. This is not to say that Heterocosmica is a mere gathering of older essays. The ground-breaking ideas proposed by Dolezel in the seventies and eighties form the core of the book, but they have been revamped, expanded, illustrated with new materials, adapted to a wider audience (one not as conversant in logical semantics and analytical philosophy as the readership of the original articles), and above all developed into a comprehensive theory.

The main body of the book consists of two parts, Narrative Worlds and Intensional Functions. Each of them is introduced by a "starter terms" section devoted to the extended definition of basic concepts. The chapters that compose the two parts are labeled in the table of contents as either analytical (i.e. straight discussion of literary texts), theoretical, or theoretical with examples. Though the book is relatively restrained in its use of technical language, a glossary at the end of the book should help readers cope with the conceptual rigor of Dolezel's use of terminology.

Dolezel's theoretical assumptions, stated in the prologue, "From Nonexistent Entities to Fictional Worlds," can be summarized as follows:

1. Fictional worlds are possible worlds constructed by language through a performative force granted to imaginative literature by cultural convention.

2. Fictional worlds are not representations (mimesis) of the actual world but autonomous realities called into being through the unrestricted creative power of fictional language. The limits of the fictionally possible are the limits of the expressible, or imaginable.

3. Fictional worlds differ ontologically from the real world through their incomplete nature. Because it is impossible for the human mind to think up an object (much less a world) in all of its properties, every fictional world presents areas of radical indeterminacy (i.e. ontological gaps).

4. Fictional language can be referential without entering into a mimetic relation to the real world. Through this position, Dolezel distanciates himself both from the neo-Saussurian, deconstructionist view that language can only refer to itself, because all realities are language-made, and from the position of "one-world" philosophers (Frege, Russell, Searle), who hold that language can only refer to actual, i.e. autonomously existing individuals. In Dolezel's model, reference to an entity does not presuppose its language-independent existence.

The first part of the book provides an exploration, bordering on a typology, of the basic constituents of narrative worlds. Whereas narrative grammars (such as Prince's) formulate the minimal conditions of narrativity by means of a syntactic pattern of events, Dolezel describes these conditions in terms of what philosophers customarily call the "furniture" (i.e. inventory of existents) of a possible world. Not surprisingly, the minimal conditions for the development of narrative action is the introduction of one individual in a fictional world. The one-person world suffers from severe dramatic restrictions, since it excludes human competition, but Dolezel's discussion of three "one-person worlds" (chapter one) shows that interpersonal conflict can be fruitfully replaced with themes such as the attempt to establish civilization in the wilderness (Robinson Crusoe), the challenges of outdoor life, (Hemingway's "Big Two-Hearted River") or the struggle of the individual with the phantasms of his own mind (Huysmans, A Rebours).

In the second chapter, "Action and Motivation," the one-person world is expanded into a multi-person universe through a theory of action inspired by analytical philosophy, especially by the work of G. H. von Wright. In this chapter Dolezel proposes a rigorous discussion of such key narrative concepts as actions, events, motivation, motivational forces, and the narrative role of unintentional events. Despite the stated focus of the book, this theory is no less relevant to the study of factual narrative than to literary fiction. Of particular interest is Dolezel's classification of action on a scale that leads from the fully rational to the impulsive and akratic (akrasia is an "incontinence or weakness of will" [70] that tempts characters to act against their own interests), all the way to the irrational and the insane. I wonder, however, if these last two categories wouldn't often benefit from being described as a course of actions that proceeds rationally from the character's personal beliefs. In this case, irrational action would not be a case of random behavior, but a matter of conflict between private worlds and the consensual world. Recognizing a realm of subjectively explainable action would not eliminate irrationality from literature, but on the contrary, bring into sharper focus the realm of the radically irrational, such as the behavior of the court in Kafka's The Trial. The dynamics of the multi-person world are illustrated in chapter 3 by discussions of Dostoevsky's The Idiot, Dickens' Little Dorrit, and the work of Milan Kundera.

The logical description of action is complemented in chapter 4, "Interaction and Power," by a discussion of its social dimensions. It is through the existence of power relations that the multi-person world differs most significantly from the one-person world. While power is "perhaps the strongest motivational factor in the multi-person world" (106), the "most representative motivational complex is the erotic cluster" (104). It would be too easy, and too plainly mimetic, to explains the universality of the literary theme of sexual desire through its prominence in human life. For Dolezel, the importance of erotics as a narrative force is due to its ability to subsume and mirror all the motivations of human actions: "The power factor in its 'assertion of mastery' [ Millet, 249] over the partner links erotics to the activities of domination, especially politics. The cognitive constituent of the cluster introduces into erotic acting practical reasoning, plans, designs, and strategies, often articulated in social scripts of courting, bonding, and so forth" (104).

Rounding up this section, a chapter on "Narrative Modalities" updates and expands an influential paper by the same title published in 1976. Here, Dolezel tightens the relations between possible worlds theory and narrative semantics through a description of the various types of "modal constraints" that structure narrative worlds. Four systems are distinguished: the alethic, made up of the operators possible, impossible, and necessary; the deontic (permitted, prohibited, and obligatory), the axiological (good, bad, indifferent), and the epistemic (known, unknown, believed). The alethic system is shown to be responsible for the division of the population of fictional worlds into groups of different abilities (gods versus humans, or the seeing among the blind), as well as for the categorization of fictional worlds as a whole as realist, fantastic, or nonsensical. Constraints of the deontic type generate plots of obligation, crime, and punishment, the axiological system underlies stories of quest and moral dilemma, and the manipulation of the categories of the epistemic system produces mystery stories, narratives of learning (the Bildungsroman), comedies of errors, and the all-important function of deceit.

Though the discussion of intensional and extensional semantics in "Starter Terms II" serves as introduction to the second part of the book, it is really the pivot of Dolezel's conception of literary meaning. The terms of intension and extension (the former a technical neologism patterned after extension, and not to be confused with intention) are used in the philosophy of language to denote roughly the same concepts as Frege's notions of sense and reference. Extension, writes Dolezel, "is the meaning constituent of a linguistic sign that directs the sign toward the world" (136). The extension of the term "Robinson Crusoe's father," for instance, is a specific individual in the fictional world of the novel. Intension, on the other hand, is the informational content carried by an expression. It may not matter much in purely practical communication whether we call a certain individual "the father of Robinson Crusoe" or "Mr. Kreutznaer" (as he is named in the novel) since the two terms have the same extension, but each of them says something different about the individual in question. The exploitation of such nuances is an important dimension of the aesthetic function of language; it determines what Dolezel calls the "texture" or the literary text: "Being fully determined by its texture, intensional meaning is affected by any textural change; it is non-paraphrasable, it slips through the net of 'interpretants,' it is lost in retelling" (138). The distinction between extensional and intensional meaning implies that the semantic analysis of literature can be pursued on two, perhaps even on three distinct levels. Purely extensional semantics - which can be understood, retrospectively, as the subject matter of the first part of the book - is the description of what narratologists call story. Concerned with existents, plot and themes, it captures a level of signification that remains largely unaffected by discourse variations. Intensional semantics, on the other hand, is the analysis of how narrative discourse structures the reader's perception of the fictional world. This broadly defined domain encompasses two distinct areas: the patterns of signification that vary with retelling but survive in reasonably faithful translations, and the shades of meaning that depend on language and culture. Dolezel's discussion of what he calls "intensional functions" is clearly tilted toward the type of meaning that depends on the telling but easily survives translation. For literary purposes (as opposed to purely logico-semantic ones), "[i]ntensional function is redefined as a global regularity of texture that affects the structuring of the fictional world" (139). To illustrate the idea, Dolezel invokes the case of the mode of reference to fictional individuals in Robinson Crusoe. From the viewpoint of purely extensional semantics all mentioned characters are simply existents, sharing the same ontological status of"incomplete imaginary entities," but the intensional texture of the novel creates a distinction between the characters designated by proper names ("Robinson," "Xury," and "Friday") and those referred to by definite descriptions: "the English captain's widow" or "Friday's father" (140). This aspect of texture does not seem to do more than divide the cast of Robinson Crusoe into major and minor characters, but in Dolezel's later reading of Kafka's The Trial, the implications of the textural effect of the mode of reference are far more significant. Because the two men who kill Joseph K. are nor referred to by institutional function, as are all other representatives of the Court, Dolezel concludes that they are not Court-appointed executioners, but the incarnation of the random violence that permeates all of society as the result of the rule of the Court (196).

The next two chapters propose further examples of what Dolezel means by "intensional function." Chapter 6, "Authentication," offers an illuminating discussion of the convention that enables fictional language to endow certain individuals and facts with fictional existence, as well as of the discourse phenomena that challenge this convention. Dolezel describes the authentication function as dyadic: "entities introduced in the discourse of the anonymous third-person narrator are eo ipso authenticated as fictional facts, while those introduced in the discourse of the fictional persons are not" (146). This formulation holds in its broad lines for traditional narratives, but as Dolezel demonstrates, twentieth-century literature has found mind-boggling ways to subvert the authority of the third-person narrator: internal contradictions, impossible worlds, games of give-and-take, and self-referential voiding. I wonder, reading this chapter, if the repeated subversion of the structure of narrative authority will eventually lead to a complete inversion of the convention, through which first-person narrators will be able to disauthenticate the declarations of third-person narrators. Dolezel makes a convincing case for the foundational role of the authentication function in the phenomenology of reading fiction, but precisely because of the importance of this role, I have my reservations about assigning this function to the domain of intensional semantics. Authentication is not a particular technique of presentation of what exists in a fictional world, but the very principle that enables fictional discourse to have an extension at all. It is therefore the precondition for all types of semantic inquiries. As such it belongs to the pragmatic domain, together with the convention that establishes the performative power of fictional discourse.

Chapter 7, "Saturation," offers a much more convincing example of intensional function. Here the function under scrutiny is the distribution of explicit, implicit, and non-existent ("gaps") information in the textual representation of a fictional world. Dolezel rejects as "mimetic" Wolfgang Iser's claim that the reading of fiction involves a filling in of informational gaps. As already stated, he regards incompleteness as the distinctive feature of fictional existence, and by filling the gaps, the reader would reduce the ontological diversity found in fictional worlds to "the uniform structure of the complete, Carnapian world" (171). Dolezel further believes that a filling of gaps would neutralize the effect of the strategies of showing and hiding that regulate the disclosure of narrative information. It is not insignificant, for instance, that Goethe's Elective Affinities "suppresses the material and organic levels and constructs explicitly only the mental and spiritual levels" (184). This instance raises the question of the location of gaps: are they part of the texture, or do they belong to the fabric of the fictional world itself? Are they, in other words, an intensional or an extensional phenomenon? Dolezel's position regarding the incompleteness of fictional worlds locates them in the extensional domain, but his concept of texture, with its variable degrees of saturation, allows them to be part of the intensional domain. One way or another, however, gaps can only be apprehended against the background of a full world: the completeness of the fictional world compared to the incomplete texture if we regard gaps as an intensional phenomenon, the fullness of the real world compared to the incompleteness of the fictional world if gaps belong to the ontology of the fictional world. The importance of this question is foregrounded by the case of Goethe's Elective Affinities. How can we explain that despite the lack of textual information about the organic and the material, the reader does not imagine Goethe's characters as disembodied minds floating ghostlike in the fictional world? The theory to which I personally subscribe is one that makes readers imagine that Goethe's characters have bodies (this is a pragmatic implicature of their being human), but does not require of them to form a mental image of these bodies. At the same time, however, readers retains the right to fill out the domain of the corporeal beyond textual specifications, if this mental imaging deepens their immersion in the fictional world and does not contradict the textually prescribed. The phenomenology of reading is a deontic system, regulated by the operators "forbidden," "permitted," and "obligatory," and by imagining the fictional world beyond the textually explicit and its implications, readers are merely making use of their right to explore the permitted. As long as they remain aware of where the obligatory ends and where the permitted begins, they will retain an appreciation of the density of the representational texture.

Two analytical chapters complete the book. In "Modern Myth," Dolezel examines how twentieth-century literature preserved the dyadic structure of myth despite the general loss of faith in the sacred. In the works of Kafka and of the Russian novelist Andrej Belyj, the dualism of the natural and supernatural is transformed into a hierarchic structure in which an irrational, invisible world of power exercises total control over the members of a visible domain. The last chapter is labeled epilogue, but rather than drawing conclusions and summing up the book, it opens a new topic: intertextuality and postmodern rewrites of famous narratives. In a very useful taxonomy, Dolezel distinguishes three types of rewrite: transposition "preserves the design and the main story of the protoworld but locates them in a different temporal or spatial setting"; expansion "extends the scope of the protoworld by filling its gaps, constructing a prehistory or posthistory, and so on"; and displacement "constructs an essentially different version of the protoworld, redesigning its structure and reinventing its story" (207). The three types are illustrated, respectively, by Ulrich Plenzdorf's The New Suffering of Young W., a rewrite of Goethe's Werther set in postwar Germany; by Jean Rhys's Wide Sargasso Sea, which tells the story of the first wife of Mr. Rochester, the man who marries Charlotte Bronte's Jane Eyre, and by J.M. Coetzee's Foe, a novel devoted to the "real story" of Robinson Crusoe. Dolezel explains the relations between the new and old versions of fictional characters through the concept of "transworld identity lines," or counterpart relations. Possible world philosophers identify counterparts of the same individual through the "rigid designator" of the name, which picks a specific individual in every possible world, but to accommodate the relation between Wiebeau, the hero of Plenzdorf's novel, and Goethe's Werther, Dolezel relaxes the conditions of transworld identity, so that characters may be rebaptized as they migrate from one fictional world to another. It seems to me, however, that this step is not needed. The hero of a transposition is not bound to the original character by relations of transworld identity, but by relations of functional analogy within similar narrative patterns. While transworld identity allows two counterparts to experience widely different fates (Hitler winning World War II), functional analogy requires similar destinies. Moreover, if the lines of transworld identity allowed renaming, a given individual could end up with two different counterparts in a fictional world: think for instance of a sequel to Flaubert's Madame Bovary, in which Emma's daughter suffers the same fate as her mother. In this world Emma's counterpart would be both Emma and Berthe. To this Dolezel might of course reply that if multiple counterparts are imaginable at all, they are indeed fictionally possible.

Through its contribution to possible worlds theory, Heterocosmica strengthensan approach to narrative that offers one of the very few alternatives in contemporary literary studies to deconstruction and its various offshoots. But it is above all a perceptive, elegantly written, wide-ranging, constantly provocative, and, in spite of its philosophical sophistication, reader-friendly discussion of crucial interpretative issues from which every student and teacher of narrative should benefit, regardless of theoretical loyalties and particular thematic interests.

Marie-Laure Ryan

References

Dolezel, Lubomir. "Narrative Modalities." Journal of Literary Semantics 5.1 (1976): 5-14.

Iser, Wolfgang. The Implied Reader: Patterns of Communication in Prose Fiction from Bunyan to Beckett. Baltimore: Johns Hopkins UP, 1974.

Maitre, Doreen. Literature and Possible Worlds. Middlesex: Polytechnic P, 1983.

Millet, Kate. Sexual Politics. New York: Doubleday, 1970.

Pavel, Thomas. Fictional Worlds. Cambridge: Harvard UP, 1986.

Prince, Gerald. A Grammar of Stories. The Hague: Mouton, 1973.

Ronen, Ruth. Possible Worlds in Literary Theory. Cambridge: Cambridge UP, 1994.

Ryan, Marie-Laure. Possible Worlds, Artificial Intelligence, and Narrative Theory Bloomington: Indiana UP, 1991

Semino, Elena. Language and World Creation in Poems and Other Texts. London: Longman, 1997.

Marie-Laure Ryan is an independent scholar and 1999 Fellow of the Society for the Humanities at Cornell University. She is the editor of the collection Cyberspace Textuality: Computer Technology and Literary Theory (Indiana UP, 1998) and is presently working on a book-length project tentatively titled Immersion versus Interactivity: Literature as Virtual Reality.

COPYRIGHT 1998 Northern Illinois University
COPYRIGHT 2000 Gale Group