Version classiqueVersion mobile
OpenEdition Books

Recent Trends in Narratological Research

John Pier

Story Modalised, or the Grammar of Virtuality*

Uri Margolin


This article deals with areas of non-factivity in literary narrative: their nature, relative weight and aesthetic impact. I begin with non-factive elements embedded in traditional retrospective narratives. These elements include negative facts (the disnarrated and counterfactual conditionals), possibilities not actualised, indeterminacies and hypothetical inferences. In concurrent ("present tense") narratives, the narrated sphere as a whole is poised on the borderline between actuality and virtuality. The story world is in the process of coming into being so that one can indeed define momentary actions or events in it, but not elicit an overall pattern of significance or define any long-range activities and processes. In addition, open conditionals, hypotheticals and possibilities not yet realised are far more central than in retrospective narratives. Prospective (future-oriented) literary narratives are concerned with that which is not yet there at telling time and are hence radically opposed to the factive and completive traditional prototype of narrative. They are dominated by one of four modalities: the doxastic (speaker's belief that something may take place); the hypothetical (a claim that certain possibilities of action exist if some prerequisite condition is fulfilled); the optative (speaker's wish that such and such should befall his addressee); and the deontic (the speaker's instructing his addressee to perform certain actions so that some other events could take place in the narrated domain). We are correspondingly moving from a scenario through an intermediate stage to a script. Specific contemporary literary examples for each of these varieties are then cited and analysed in some detail. The article concludes with a discussion of several possible aesthetic and wider cultural reasons which may account for the employment of such unusual forms by contemporary authors

Texte intégral

  • * An initial, shorter version of this paper was presented at an international conference on literatu (...)

1It is a truth universally acknowledged (or at least a preconception universally enunciated) that literary narratives revolve around that which has already occurred and been completed in the story world, in other words, singular past states, actions and events; that such narratives are presented by the textual narrating voice as known facts; and that they thrive on certainty or factivity. But like all so-called universal truths, this one, too, is only partially valid. In this paper, I would like to point out, firstly, the existence of numerous areas of nonfactivity, even in narratives which conform to the classical model. I will then move on to explore several types of literary narrative for which the classical prototype of pastness+factivity+knowledge holds in part only, or not at all. I will seek to demonstrate that, far from being inconsequential oddities, such narratives constitute a modal system which is complementary or alternative to the restricted classical model.


2A. Any classical, retrospective literary narrative can be viewed as a partial description of an assumedly pre-existent non-actual domain with its entities, states, actions and processes. The set of propositions, whether textually formulated or reader inferred, which have the truth value T/1 with respect to this domain, describes what the case is in the corresponding domain; in other words, this set of propositions represents the facts of the matter, the story told. But oftentimes that which did not happen or exist, when viewed against a background of textually defined possibilities or probabilities, or against the inscribed norms and regularities of the text world, is every bit as significant. Events which did not happen, although they could have, when explicitly referred to by the narrator, define the sphere of the disnarrated (Prince, 1988). The disnarrated can be represented formally by narrative propositions with the truth value F/0 (i. e., it was not the case that p) or as true propositions whose content involves negative facts (i. e., it was the case that not p). The disnarrated may also involve the epistemic and doxastic: that which could have been known or believed by one or more of the narrative agents but wasn't; the deontic: that which should have been done but wasn't; and the bulomeic: that which could have been wished for but wasn't.

3Another way in which negative facts can be expressed in the context of retrospective narration is by means of counterfactual conditionals, such as "if she had been quicker in expressing her love for him, he would have stayed with her." The falsity of the antecedent, indicated by grammatical means, suggests that the consequent is false, as well. The proposition as a whole thus conveys two negative facts and their logical interrelatedness, such as cause and effect or enabling condition and subsequent situation. Although counterfactuals are, strictly speaking, about what was not the case, they still suggest that this could have been the case in the story world had things gone otherwise, thereby leading us to the realm of possibilities missed or not actualised.

4Possibilities not actualised, alternatives not implemented or roads not taken, although they were available at a given point in the action sequence, form a thematically significant part of retrospective narration, defining a set of virtual scenarios with respect to the actual story world. For example, "She could marry a rich old man, but chose instead to marry a poor young one." In fact, the significance — psychological, moral, deontic, axiological or whatever — of a narrative agent's actions often emerges only when it is confronted with the zone of alternative options open to this agent. The objective existence within the story world of such possibilities is easiest to establish in retro-narration, where the narrator is able to survey the whole field of action in a synoptic manner in the light of later developments and final outcomes. Positive and negative facts, counterfactuals and possibilities which existed but were not realised thus jointly define what was the case in the narrative reference world.

5Other existents, properties and relations, actions and events in the narrated domain may be explicitly referred to by the narrating voice in a mode of doubt and uncertainty: X may or may not have happened, existed or been thus and so; the course of events was precipitated by A or Β or C, but no one knows for sure by which one of them; some say that p was the case, others maintain that it was q, but it is impossible to decide between these conflicting claims. In all such cases, we are dealing with narrative propositions which lack both definite confirmation and disconfirmation, with assumptions and conjectures left open, with uncertainties and indeterminacies. The truth value of such propositions or disjunctions is hence indeterminate, since their truth or falsity in the story world is deliberately left undecided and undecidable. Differently put, each of the propositions in question is modalised as "possibly p," which is of course logically equivalent to "possibly not-p." While the nature and scope of narrative indeterminacy varies in different texts, we are all familiar with a whole genre, that of the conte fantastique, which is defined by a built in, unresolved hesitation between a natural agency (often psychological) and a supernatural one as the motivating force behind central events in the story. This hesitation inevitably spills over into the very ontological landscape of the narrated domain, which may or may not include supernatural entities. But even here, indeterminacy is not total: it is limited to certain events and entities and is defined relative to a stable framework of determinacy, embodied in the conjunction of facts and negative facts: A existed, Β did not exist, and C may or may not have existed.

6Indeterminacy may also be expressed by means of narrative hypothetical inferences such as "If anyone was there, s/he would see the great volcano erupt." As opposed to counterfactuals, the narrator is not able to assert the truth or falsity of either antecedent or consequent, so both are left in limbo as interrelated events which may or may not have taken place in the story world. For a further clarification of the relative role of indeterminacy and its status as merely epistemic or irreducibly ontic, we need now to correlate narrative and temporality. When a narrated course of events is textually presented as anterior, as having been completed before the moment of narration, we are obviously dealing with narrative retrospection, with a reconstruction of what has gone on earlier, with a configuring of earlier states and events into a totality with global coherence. All the facts, both positive and negative, are in, so to speak; the relevant time frame is completed or closed, and certain knowledge can be available to the narrator. It is hence a fact that some things have occurred, that others have not, and that still other things may or may not have occurred. In other words, it is an objective fact of the narrated domain, not a mere logical possibility, that something has possibly happened, that a narrative agent may have been thus and so, that event A may have caused event B. Technically speaking, the merely possibly-possible is ruled out. The narrator and the reader are not able to resolve certain indeterminacies, to assign a truth-value to certain propositions, but this is due to an irreducible lack of information, not the inherent indeterminacy of the story world itself. In this sense, traditional narrative does not share the ontology of quantum mechanics, and all indeterminacies are, in the final account, considered merely epistemic, stemming from the incomplete information state of the narrating voice. In a similar vein, inferences made by the narrator or narrative agents are assumed to have been objectively satisfied or not, and suppositions or future projections referring to moments within the narrative's time span to have been realised or not. The temporal boundedness or closure of the narrated domain also deprives any wishes, hopes and commands the narrator may utter with respect to this domain of their illocutionary force: after all, we can no longer influence that which has already taken place.

7B. When a narrated course of events is being presented as taking place concurrently/simultaneously with the act of its narration, we are dealing with on-the-scene reporting, observation or inspection, verbalising as one watches, not with retrospection; with the stepwise, on-line real time construction of a sequence, with a gradual figuring out of what is the case as it evolves, not with its integration into a totality in hindsight. The narrated domain is a world in the process of becoming or unfolding, progressively taking shape as it is being narrated, so that the overall situation emerges in stages. The situation is viewed from within as a series of unconfigured particulars. It is in progress, in the midst of a process, and does not yet form a bounded whole. It is not possible as yet to elicit an overall pattern from the succession, to determine results and define the contribution of individual events or actions to the global course of events. The dominant use of progressive and imperfective present tense verb forms in such narratives is a grammatical encoding of this ontic and epistemic situation. In simultaneous narration, factivity is inevitably reduced. While punctual, momentary actions and events can be reported with certainty, long-range, multi-staged ones, as well as most cause and effect relations, remain within the realm of mere possibility, since they have not yet reached their completion at the moment of utterance. Their outcomes are still undecided, and they may or may not succeed, be accomplished or achieve their goals. And the same applies to the negation of processes and complex actions, that is, the assertion that they have definitely failed to achieve their final stage. As for possibilities, most of them are further modalised, in that the narrator cannot often determine in situ whether they actually exist in the story world or are merely logically conceivable. Corresponding to this dramatic decrease in factivity, open conditionals or hypotheticals possess their full force: given that p has occurred or is occurring, q may or may not follow, but this is undecided as yet at utterance time. Similarly, whenever the text has a marked narrator or speaker, his or her wishes and hopes at every given moment regain at least part of their illocutionary force, especially with respect to the outcomes of processes not yet completed at utterance time. Narration in the present is thus located on the cusp or borderline between factivity and indeterminacy, actuality and virtuality, objective and speaker-envisioned possibilities of knowing.

8C. However, this is not the final stage. Narration can also be prospective, dealing with posterior events, with that which has not yet occurred at speech time: a prediction, prognosis, scenario, projection, conjecture, wish, plan, and so on. Now here there is as yet no fact of the matter, positive or negative, to be experienced, known and reported. Prospection, construction and pre-figuring are now what the textual voice or personalised narrator is engaged in. The story world is purely virtual: a mere potentiality or possibility being projected, entertained and described, not a range of actual facts to be reported upon. While the semantic distinction between fact, negative fact and possibility remains, all three categories are subjected unexceptionally to the global "possibly" operator. Much is possible, but nothing has been decided as yet. While in past narratives the virtual was subordinated to and defined by the actual, and while both exist in an uneasy balance in concurrent narration, the would-be actual is now definitely put in the service of the potential. The fact/possibility, determined/undetermined hierarchy has by now been decidedly reversed. The governing frame or narrative matrix-clause has been shifted from the constative or affirmative to the speculative. Barring prophecies with supernatural warrant, the narrator of a future scenario cannot claim any certain knowledge of his reference world, and obviously no first hand experience of it. Belief or doxa necessarily replaces narrative certainty. All assertions in a future narrative thus have the indeterminate truth value at utterance time, and are problematic assertions in the technical philosophical sense. The temporal position (not yet) and the modal (mere possibility) are essentially intertwined in future discourse, the narrator's epistemic limitations being ontically grounded. Conversely, a process which began with concurrent narration now reaches its completion: hypotheses and conditions are truly just that, while the narrator's wishes, desires and instructions regarding the projected future domain possess their full illocutionary force.


9The essential intertwining of the temporal location of events and their modal status in future narratives opens the door to a wider array of variously modalised narratives, all future-oriented and all opposed to the factive, completive, classical prototype. Let me define them by means of a semi-serious phrase each and then proceed to a fuller discussion and illustration. Here we go: (1) The future possible simpliciter or "once upon a time there will be"; (2) This may happen (to you/one) if something else takes place; (3) May this happen to you; (4) Do this and trigger the following course of events. The dominant modalities of these kinds of narratives can be ranged on a scale as follows. First, the doxastic, the speaker's belief that the following may take place. Second, the hypothetical, claiming that certain possibilities of action exist for one/whoever if a certain prerequisite condition is fulfilled. Third, the optative, the speaker's wish that such and such should befall his addressee. Finally, the deontic (understood in the wide, linguistic sense), the speaker imposing obligations on his/her addressee, instructing him/her to perform or initiate certain actions so that certain events could develop in the narrated domain. In terms of dominant mental faculties, the movement is from the cognitive via the emotive to the imperative. We are correspondingly moving from a scenario, in the sense of an outline of a potential course of events, through an intermediate stage to a script, in which the textual addressee is given instructions which need to be fulfilled in order for an envisioned project to be realised. While all four varieties are inherently opposed to the concept of narrative as a factive report of what has already occurred, the report as a description of temporal situations, regardless of their ontic and epistemic status, still dominates future-oriented narratives. But it is framed by increasingly different types of discourse: inferential-conditional, emotive and imperative. The dominant type of illocutionary act (to put it in John Searle's terminology) moves correspondingly from the representative (assertive) through the expressive (emotive) to the directive. The presence of a textually marked speaker or addressee is optional in future and conditional narratives, while it is built into the other two — the optative and imperative. In the last two, the addressee, as both recipient of the message and its topic entity, is naturally the essential constituent. The standard grammatical forms by which the dominant modalities of the four kinds of narrative are expressed are the future indicative, the present conditional, the present subjunctive and the imperative respectively.

10Now, however, the time has come to leave the abstract system aside for a while and look at a few concrete examples. The American linguist Suzanne Fleischman states (Fleischman, 1982, 30) that we can regard discourse about the future as a temporal screen onto which humans project a variety of modalised notions, such as possibility and uncertainty rooted in the present, as a projection of one's experiential present hopes or apprehensions, and not as an objective ontological category. This linguistic insight is beautifully borne out by Christine Brooke-Rose's novel Amalgamemnon (1984), a first-person narrative consisting of various scenarios constructed by the speaker as to her possible future life course following her impending forced early retirement from her university post due to cutbacks. The only factive component of the narrative concerns the speaker's present state of anxiety and her ongoing activity of projection and speculation. All narrative propositions refer to non-realised states and events: future or future conditional, expressed by verbs in the future tense or in the modal auxiliary (may, could, should, would, might). There are hence no grounds for including or excluding any particular projected sequence from the story world or for choosing between alternatives, the entire world of Amalgamemnon being left dangling in an ontological limbo (McHale, 1992, 219). Character discourse, as distinct from that of the narrator, does include references to past and present events, but only in the interrogative or in the negative, thus preserving, as Brooke-Rose herself says, the notion that nothing can be said to have happened or to be happening. Since the future is essentially indeterminate, several alternative probable branches grow forth from the common present node, and the resultant canonic structure of the narrative is "possibly" (a or b or c).

And now from future to conditional.

11As we know, all open conditional sentences (if p, then q) consist of an antecedent and consequent (protasis and apodosis, in traditional grammar) whereby the antecedent proposition explicitly sets up a possible future situation or event while the consequent describes what may probably/likely follow from it. In other words, both p and q are entertained as mere possibilities or suppositions, not as facts. The maximally conditional, hence minimally factive, story would accordingly consist of a description of a merely possible future initial state of affairs, with what follows left completely open or unspecified, that is, a problematic supposition or premiss with no conclusion, or, quite simply, "if p, then what?"

12This is precisely the case of the would-be story generated by the chapter headings of Calvino's If on a Winter's Night a Traveller (1979): "If on a winter's night a traveller...leaning from a steep slope looks down in a network of leaves...what story down there awaits its end?" The reverse case, that of a deleted but easily recoverable initial assumption and a detailed description of what could probably/possibly follow, is provided by the Swiss-German author, Ε. Y. Meyer, in his novel In Trubschachen (1973). The novel consists of a potential narrative, recounting on a day-by-day basis the activities of an anonymous traveller, designated as man (= indefinite singular pronoun, roughly equivalent to the French on), who could, might, or would spend a brief Christmas vacation (December 27 to January 3) in this small village in the Emmental. The deleted premiss is quite banal: "Should one go on vacation in Trubschachen at this time of the year, then one would...." The activities of this man, from the inbound to the outbound train trip, are recounted almost entirely in the German Konjunktiv form: "Man würde" — as are the setting, circumstances, and co-agents: "ein Wind wäre aufgezogen, der Himmel würde sich bedeckt haben," or "die meisten der Reisenden würden den Zug rechtzeitig genug bestiegen haben," and so on. While the activities "one" engages in range from the generic and the typical, applicable to every traveller/whoever (such as carrying one's luggage) to the specific and the unique (such as reading this particular passage of this specific book at this hour), their rendering is relentlessly hypothetical, as if saying "here is a description of a possible situation and course of events, of what one could do," and not a report or assertion of actualised, factive events. While the pronoun man itself may be recuperated as either first-or third-person narrative, the associated actions and situations cannot be naturalised into a traditional narrative form, since they are possibilities entertained, not facts asserted. Meyer has also judiciously chosen the würde form, which expresses a degree of probability located between the absolute müsste and the weaker könnte.

13Marguerite Duras' La Maladie de la mort (1982) is a novel in which both antecedent and consequent regularly occur. This would-be story is presented by a voice which invokes, in the conditionnel présent, certain possibilities regarding an encounter between an unnamed vous (singular masculine) and an elle and the stages of their relationship, and it then describes scenes from this relationship in the present tense, thus alternating repeatedly between pouvoir-être (envisaged possibility) and the assertion of être (what is), between si and ainsi. In other words, the sentences in the conditional set up certain possibilities or potentialities for action whose envisioned consequences are explored by scenes in the present tense. However, the factive sections — and the text as a whole — are invariably framed by a conditional, thereby making the narratively actual conditional upon the realisation of an open, hypothetic possibility. For example: "vous devriez ne pas la connaître...vous pourriez l'avoir payée," where the modal verbs and the conditional ending both convey the idea of hypotheticality; and further on: "chaque jour elle viendrait... chaque jour elle vient," etc. The use of vous+conditionnel can be also construed, however, as a set of cinematographic or scenic directions for an actor who is about to perform the male role, as a programme for staging or a script, as well as a story sous condition. Under this interpretation, the conditional nature of the events is linked to an anticipated creation of a ludic world or game of make-believe, described in the present tense passages. Because of the possibility of the generalised vous being appropriated by actual male readers, a third, related interpretation is also conceivable: that the text is in fact a series of guidelines to the reader on how to imagine from beginning to end a love affair with an unknown woman. On these last two readings, the doxastic clearly shades into, or is even displaced by, the deontic: what the vous ought to do to bring the story about in a make-believe domain of individual imagining or of public staging.

14The voice with whom a future-oriented narrative discourse originates may not only draw inferences with respect to potential situations and courses of events, but also express his/her attitude towards such imagined situations: hoping and wishing that they come about; wishing that a given individual be of a certain nature or that s/he do, experience and say certain things (imprecation). In so far as the individual in question is personalised and specified as an addressee, we get a blend of the referential (the story being invoked), the affective (the speaker's attitude towards his/her projected reality) and the conative, as the very formulation of these propositions is meant to have an impact on the addressee's future destiny. While in Duras' case the deontic reading was purely optional, we now are clearly faced with a narrative which is inherently on the borderline between scenario and script. Jean-Michel Raynaud's Pour un Perec lettré, chiffré (1987) is a rare example of an extended text predicated upon the speaker's wishes with regard to his addressee: "que vous concluiez..que vous insistiez, que vous fassiez remarquer, que vous soyez, que l'université ait organisé cette conférence, que la nuit soit belle," and so on. The necessity, at least in French, of employing repeatedly not only the same verb form (subjunctive present), but also the same phraseological mold, prevents us from being absorbed in the would-be reality, insisting on its purely wished-for rather than actual status. The projected or wished-for nature of the life story is also reflected in the very title of the book, which includes the word pour (towards). The text as a whole thus describes a project, not an accomplishment.

15It is only a small step from a narrative consisting of wishes with respect to the addressee to one in which s/he is put under obligation to perform certain acts and thereby give rise to the story in which s/he plays the/a major role. Narratives whose framing clause consists of imperatives are an obvious case in point, since they are by definition future-oriented, directed to an addressee, and express the speaker's desires with respect to the domain of reference. In six of the stories in Lorrie Moore's collection, Self-Help (1985), a female character, who is either a teen-ager or an adult, is instructed how to behave in situations involving a divorced mother, a lover, a husband, an intended literary career. The purpose of the behaviour is the realisation of specific states or the achievement of certain goals. Fulfilment of the initial instructions gives rise to the development of the corresponding personal interrelations, described in a series of scenes in the present or future tense, with further instructions being interspersed as the sequence of the projected events progresses. We thus have an intercalation of instructions and attendant projected action sequences which, together, yield a possible biography of the recipient of the instructions as regards those items of behaviour that are either at issue in the instructions or result from them. This intercalation of imperative and indicative is in fact inevitable, since no narrative, actual or virtual, can consist of instructions alone. It is only the (non-)fulfillment of these instructions and its further results, expressed in the indicative mood, which can form a story of any kind. This is somewhat analogous to the antecedent-consequent relation discussed earlier. Two more points about instructions are worth noting. First, instructions always look ahead to what is not yet there. The title of the collection (Self-Help) as well as the titles of its individual stories ("A Guide to," "How to") remind us that the point of departure consists of a script yet to be performed, not a fait accompli. Second, their illocutionary force notwithstanding, instructions may or may not be carried out, and even if carried out conscientiously, they may not yield the expected results. Any results expected from following a set of instructions are a conjecture, not a fact. And indeed, even though the "you" closely follows the textually-inscribed instructions, all ensuing affairs and relations in Moore's stories invariably end in failure.


16Let me conclude with some brief remarks about three related issues: (A) textual devices which enhance the non-factive status of the foregoing kinds of narrative; (B) the compensatory narrative patterns they employ; and (C) possible reasons for the emergence of such nonstandard narrative forms.

17(A)The non-factive nature of the narratives discussed in this article is further enhanced by their use of indeterminate referents: a traveller, man, vous, elle, "you," "him," and (in Meyer's novel) by the reference to all co-agents by role rather than by name, such as der Lehrer, die Wirtin. At least some of the virtual events referred to in all of these works are presented, not as unique possibilities, but as clusters of alternatives. Brooke-Rose's narrator projects diverse alternative possible futures for herself, from pig farming to international terrorism; the woman in Moore's story "How" is told to "begin by meeting him in a class, in a bar, at a rummage sale"; Meyer's man could possibly encounter, on his inbound trip, "an invalid, a deaf person, a blind person, a deaf and mute person." In Meyer's novel, all speech, including that of the man, is rendered exclusively in the indirect mode (oratio obliqua), which in German is always in the Konjunktiv form. This pervasive morphological feature helps to further enhance the merely potential quality of the surrounding narrated context, which is also portrayed in this form, as we recall. In Duras' story, the narrator addresses the vous towards the end of the narrative, after elle has disappeared, and raises the possibility that the whole love affair is now being recounted by the vous to some audience as something that could not actually have happened, or as something that the vous himself has invented. With this one stroke, Duras conjures up at least one more level of embedded indeterminacy: the possibility of a possible story-telling carried out by a possible character. Duras' text is followed by three pages of authorial commentary, all in the conditional, discussing various possibilities and manners of staging or filming the foregoing text, possibilities of transforming one kind of "as if into another mode of make believe.

18(B)Even in these texts, action, action sequence and the logic of action — all traditionally viewed as central to narrative — are not dead, however. After all, even a virtual story is still structurally a verbal representation of an action sequence, and this is borne out by a whole array of elements and patterns. First of all, future projections share with past completive narratives the capacity of describing long and complex sequences of actions. Both punctual and durative events as well as short-term and long-term actions can find their place in a modalised story, which, in this sense, has the same range of options as traditional narrative. The texts I have been examining also narrate their projected stories in a strictly chronological order, thereby satisfying the most basic condition of action description. In Brooke-Rose's case, each alternative projected life story also loops back to the present moment of speculation and narration. Thirdly, although the factive and determinate is subordinated to or embedded in the non-factive, it is still there. In Duras and Moore, conditional and imperative frames are followed by scenes in the present, while the climactic scene of Meyer's novel, where "one" loses his way at night and is almost buried in a snow drift, is rendered in a blow-by-blow historical present. Fourth, the stories told are often familiar in their discourse models, following the literary stereotypes of a spy novel, thriller or cheap romance in Brooke-Rose, of the unhappy love affair in Duras and Moore, and of the pedantic Swiss tourist guide book in Meyer. In Moore's case, the text type of the collection as a whole is a parody of the American favourite: a primer, guide or "how to" manual — especially how to succeed in love, sex, business, etc. Fifth, much of the material utilised by these texts consists of familiar world knowledge and schemata of everyday behaviour. Brooke-Rose builds on international terrorism and kidnappings in the seventies and on the endless prognosticating chatter (economic, social and political) in the mass media; Meyer on the realia of Swiss everyday life, from common dishes to the SBB timetable; Moore on the pervasive role of TV, fast food and pets in middle-class American life. Raynaud's case is much more radical. His book, published in 1987, several years after Perec's untimely death, is in fact a fictionalisation in retrospect of actual events, utterances and literary activities of Perec's intellectual life. But, as Raynaud says, "since this biography misses Perec, it gives rise to several subjunctives." What is merely wished for in the fictional domain has already taken place and gone in the actual world, and the writer's wish now, as he writes, is to turn the past life and person into a timeless literary monument, an enduring character and text which will capture and immortalise Perec.

19(C)Why would contemporary authors employ such unusual forms to begin with? I can think of several good reasons. In the broadest terms, the production of (alternative) scenarios for one's own future, or for that of a given group or even of mankind as a whole, is an ever increasing obsession of Western culture. All such projections are of course purely virtual at the time of their production, a fact that is further enhanced by the symbolic mode through which most of them are generated, namely, computer modelling, where pure virtuality is the name of the game. In this respect, literature is part and parcel of a wider process, which, according to some postmodern thinkers, may lead or has already led to the very replacement of the actual by the virtual as the dominant factor of our lives and our conception of the world.

20However, more specifically aesthetic and literary reasons are also at issue. Imagining what may happen simpliciter or if some condition is met, playing with different alternative scenarios, wishing or commanding that some state of affairs should come into being are all activities within the fictional sphere which are analogous to the very act of literary creation, where everything is, initially, a merely potential story, and where the reality constructed is always non-actual and always amenable to further modification through manipulation by its creator. And then there is the parallel preoccupation with the reader and his/her role, one of the central concerns of postmodern literature and literary theory alike. One specific form this general issue has assumed concerns the possiblity of the actual reader occupying the slot defined by the main narrative agent of the story in front of him/her. Making the events lie ahead of narration time (or providing a governing narrative frame consisting of scene-setting for potential events), making these events largely stereotypical and habitual and leaving their protagonist non-specific, hence, potentially each and every you/one, are all conducive to readerly imaginative identification, to making it true of oneself in the game of make believe that one is a member of the narrated domain, that one is acting out the textually inscribed scenario or script. Reduced factivity and specificity may thus be compensated for by the emergence of a near ideal set of conditions for readerly involvement and readiness to participate in a vicarious experience, to assume the textually inscribed role for a game of make-believe which may last as long as the reading act itself. Although all the texts I have discussed precede the advent of computer-generated, virtual-reality activities, from hypertext to cyberspace, they seem to be almost prophetic in this respect. One could hence definitely expect to see in the future more and more narratives written under the sign of virtuality. This, too, is naturally merely possible, but quite likely, indeed.


Works cited

Primary sources

Brooke-Rose, Christine, 1984. Amalgamemnon. Manchester: Carcanet Press.

Calvino, Italo, [1979] 1981. If on a Winter's Night a Traveller. Trans. William Weaver. Toronto: Lester & Orpen Dennys.

Duras, Marguerite, 1982. La Maladie de la mort. Paris: Les Editions de Minuit.

Meyer, Ε. Y, 1973. In Trubschachen. Frankfurt a. M.: Suhrkamp Verlag.

Moore, Lorrie, 1985. Self-Help. New York: Alfred A. Knopf.

Raynaud, Jean-Michel, 1987. Pour un Perec lettré, chiffré. Lille: PU Lille.

Secondary sources

Fleischman, Suzanne, 1982. The Future in Thought and Language. Cambridge: Cambridge UP.

McHale, Brian, 1992. Constructing Postmodernism. London and New York: Routledge.

Prince, Gerald, 1988. "The Disnarrated." Style 22. 1, 1-8.

Notes de fin

* An initial, shorter version of this paper was presented at an international conference on literature and philosophy held at Tel Aviv University in March 1996. I wish to express my gratitude to Professor Monika Fludernik, Freiburg University, whose perceptive critical comments helped me improve upon the original version.


Professor of Comparative Literature at the University of Alberta, Edmonton, Canada. His research is focused on literary theory and methodology and on narratology. Publications include close to forty articles in collective volumes as well as in international journals such as Journal of Literary Semantics, Language and Literature, Language and Style, Poetics Today, Semiotica and Style

© Presses universitaires François-Rabelais, 1999

Conditions d’utilisation :