In modern Western literary theory, literary fiction is an exceedingly important basic category; many of the modern features of literature are grounded upon it. Yet it is precisely this problem that has provoked endless dispute. And postmodern theory’s dissolution of the boundary between the fictional and the non-fictional has thoroughly shaken the modern conception of literature. Still, if we set out from the speech-act theory of literary discourse and re-examine these problems, then not only can literary fiction itself be elucidated more thoroughly, but a series of much-debated basic problems of literary theory—and even the conflicts among the social-historical, the aesthetic, and the cultural interpretations of literature in the past—can also be given a more pertinent and reasonable account.
In works such as Speech Acts: An Essay in the Philosophy of Language and Expression and Meaning: Studies in the Theory of Speech Acts, Searle further deepened Austin’s view of “doing things with words” in ordinary discourse and clearly pointed out that whether one speaks or writes, one is performing a very definite speech act—the “illocutionary act.” How, then, does fictional discourse—a special form of discourse excluded from Austin’s analysis of ordinary discourse—do things? Put more concretely and precisely: “The words and other elements in a fictional story have their ordinary meanings, and yet the rules that attach to these words and other elements and determine their meanings are not followed. How can these two things both occur at once?” (Searle, “Fictional Discourse,” 141). On this, with his consummate analytic skill and rigorous logic, Searle advanced a series of important views, which also provoked dispute among scholars.
Searle held that, unlike ordinary discourse that does things with words, fictional discourse is the author’s pretended performance of illocutionary acts. Take, for example, the common story opening: “Once upon a time there was a faraway kingdom where there lived a wise king who had a beautiful daughter.” The literal sense of this sentence is clear, yet it is non-serious; it does not satisfy the conditions laid down by semantic and pragmatic rules. The author neither takes responsibility for the truth of the proposition expressed by his assertion nor genuinely believes what he says; he lacks sincerity. That is to say, the fictional discourse above is, on its face, an act of assertion, yet in fact does not follow the constitutive rules an assertion ought to possess; the author asserts under pretense, or behaves in action as though he were asserting—a pretended illocutionary act. This pretended illocutionary act is an imitation of the ordinary act of doing things with words; it is parasitic, non-serious, and bound up with the author’s intention. Searle points out that the constitutive principles of truthfulness and sincerity that non-fictional discourse requires the author to follow are, in essence, the principle that words connect to the world; he regards this as the “vertical rules” that establish a link between language and reality. Once the author follows the vertical rules, his discourse establishes a connection with reality and thereby possesses the capacity to refer to the real world. And the very possibility of fictional discourse arises because it imitates non-fictional discourse while drawing on a series of extralinguistic, non-semantic conventions, conventions that break the connection between word and world established by the constitutive principles above. Searle calls these conventions of fictional discourse “horizontal conventions.” He says: “The pretended performance of illocutionary acts constitutes a work of fiction; this pretended performance is in fact made successful by performing the kind of speech act whose intention is to invoke the horizontal conventions, conventions that suspend the illocutionary commitments that generally inhere in any utterance” (“Fictional Discourse,” 145).
Against Searle’s view that fictional discourse is the author’s assertion under pretense, a non-serious illocutionary act, Genette proposed a correction. One should, he held, regard Searle’s so-called pretended assertion from a more flexible angle. He says: “To produce a false assertion (or to pretend to produce one) does not, as a matter of course, exclude the following possibility: in producing an assertion (or pretending to produce one), one is in fact accomplishing another act—the act of producing a work of fiction” (115). On the surface, fictional narrative always likes to put on the garb of assertion; yet in reality this very state is the use of non-verbal intimations, demands, requests, suggestions—inviting the reader to spread the wings of imagination and enter together into the world of the fictional work. Concretely, by virtue of conventional usage—namely, Searle’s extralinguistic, non-semantic horizontal cultural conventions—a story opening like “Once upon a time …” that seems to be an assertion is in essence an announcement that a fictional story has begun. The author of a fictional work, authorized precisely by cultural convention, performs an announcement: “to produce a work of fiction.” Such a mode of discourse is, literally, a non-serious, pretended assertion; yet at the non-verbal level it accomplishes another act, a serious act of construction and announcement, which, by constructing a fictional world and proclaiming this act, influences reality. The author writes seriously, produces the fictional work seriously, seriously creates an artistic world distinct from the real world, and seriously announces to the world his own distinctive illocutionary act.
At this point we may give a relatively clear definition of literary fiction. What we mean by literary fiction is precisely what Genette calls producing a work of fiction, creating an imagined fictional world—not falsity in the literal or thematic sense, nor Searle’s assertion under pretense. To put it another way: literary fiction is the construction of a fictional world in the sense of utterance performativity, not fiction in the sense of representation (a discrepancy between statement and fact). The many disputes about literary fiction have their root precisely in the confusion of two different kinds of fiction. Consider, for instance, this sentence from Flaubert’s Sentimental Education: “On the 15th of September 1840, about six o’clock in the morning, the Ville-de-Montereau, just ready to start, was sending up great whirls of smoke at the quay of Saint-Bernard …” If we regard this sentence in isolation, the meaning it states is wholly acceptable; it is faithful to the truth of experience and may, on its face, be taken as non-fictional—even the “great whirls of smoke” may perfectly well accord with reality. Yet cultural convention tells us that, for ordinary discourse or other applied discourse, “great whirls of smoke” is altogether a linguistic excrescence; it conveys a “useless” piece of information, and to delete it would better accord with the principle of discursive economy. The appearance of “great whirls of smoke” precisely announces, in its “visualizing” descriptive posture: here begins the construction of a literary world of imagination. Although the meaning stated by the phrase above may not be false, it nonetheless summons the reader and demands that the reader take part in constructing a literary fictional world; the whole sentence is thereby dragged into the literary fictional world, participates in the construction of the fictional world, and becomes a constituent of it.
We may further draw the following view: the fictional activity of literary utterance performativity has no necessary connection with whether the content stated by the utterance is false; it often depends on cultural convention. When discourse, instead of referring to the existent world, performs an act of construction and proclaims that it will construct a fictional world, discourse necessarily becomes non-self-sufficient. For language alone is powerless to construct a world; it is always full of the “spots of indeterminacy” of which Ingarden speaks and the “blanks” of which Iser speaks, requiring the reader to fill them. Hillis Miller says: “As I have said, each work opens a magical world, and unless one reads the work one cannot approach it; reading is the affair in which a person, holding nothing back, throws all his thought, feeling, and imagination into it and, within himself, re-creates the world together with the words” (Miller, “On Literature,” 118). Hence, as the discourse that produces a fictional world, literary discourse must invite the reader to participate in joint construction. While literary discourse announces the construction of a world, it also exercises another illocutionary act: it makes a narrative gesture or other non-verbal act, and, using cultural convention, summons the reader to imagine a world together with the language.
The “referential illusion” of which Barthes speaks has its root precisely in the distinctive performativity of literary discourse. Barthes held that the signifier of literary discourse joins secretly and directly with the referent, constituting the “concrete detail” of the work’s world at the level of the sign, while the signified is expelled from the sign—so that any possibility of developing the “form of the signifier” is also excluded. This is the “referential illusion”: as the signified of denotation, the real is deleted in the realist statement; as the signified of connotation, the real returns to the realist statement. The real, in realist literature, is no more than a sense of verisimilitude. Hence he says: “For the real, discourse bears no responsibility whatever: in the most realist novel, the referent has no ‘reality’ … in essence, what realist theory calls the ‘real’ is merely the code of representation (of signification): it is not a code that can be put into practice; the truth of story is not realizable” (Barthes, 164).
Apart from performative sentences, what an ordinary constative sentence first draws attention to is what it states—its literal sense. The content of a statement often conceals the performative feature of the sentence. Because the concept used in stating appeals to human consciousness, while its performativity is often non-verbal, it is easily overlooked by consciousness; this leaves the discourse’s capacity to do things in a latent state, or in a state where, though it has taken effect, it has not been noticed by consciousness. Hence, when Austin advanced the view of doing things with words, it could come as something fresh, even startling. Literary works, by various means, strengthen and highlight the performativity of discourse, letting the act of doing things with words—especially the act of construction—become the dominant feature of literary discourse. This is one reason literary discourse is more imagistic and vivid: it does not lay rows of corpse-like, hollow concepts before the reader’s eyes, but creates a world of flesh and blood and desire and invites the reader to live within it in person; and it is also one reason literary discourse has more illocutionary force and can better move the heart and infect the reader. When literary discourse, by its illocutionary force, draws the reader to participate in constructing the fictional world, then all the discourse in the work—including literally fictional or non-fictional discourse—is swept into this fictional activity, becoming a constructive force and constituent of the fictional world, set in two mutually parallel dimensions distinct from the real world. In the fictional context, the truth of statement is stripped away; the apparent reference to reality becomes “illusion,” a decoration of considerable verisimilitude within the fictional world; discursive reference has degenerated into reference internal to the fictional world. In Searle’s terms, literary discourse, by virtue of “horizontal conventions” (cultural conventions), proclaims and summons the reader to take part in the fictional activity of constructing the literary world; the reader dwells in a fictional world separated from the real world; the “vertical rules” connecting word and reality must then be temporarily suspended, at which point they have ceased, for the moment, to be in force. Hence our other view: the performativity that is the characteristic of literary discourse possesses a distinctive force; when it unfolds fictional activity, it carries out what Marcuse calls “aesthetic tyranny”—altering the character of the various kinds of discourse surrounding it, suspending their referential function toward reality, and turning statement into pseudo-statement.
Richard Walsh, for his part, interprets literary fiction from the angle of relevance theory. He holds that the reader’s reading needs a context capable of supporting his own hopes, a context constituted by a set of assumptions adopted by the reader, which are part of the cognitive environment. And all assumptions are, to varying degrees, products of inference; this is a relevance-driven pragmatic process and does not depend on the literal truth of what is said. Hence he says: “The problem of fictionality is by no means a problem of truth or falsity, but a problem of relevance. The reader’s motive in seeking a suitable interpretive context is the presupposition of relevance, not the expectation of the literal truth or falsity of the words.” Fictionality “is a contextual assumption; that is, in understanding a fictional utterance, one explicitly assumes that the utterance is fictional. The principal contextual effect of this assumption will place those implicatures that depend on literal truth in a relatively subordinate position, and will give greater weight to those implicatures that establish relevance by means of broad accumulation” (163). When literary discursive activity unfolds the construction of context, the literal implicatures of discourse are placed in a subordinate position, and truth becomes subordinate to relevance. This, too, illustrates from another angle that, in the activity by which a literary work constructs a fictional world, the vertical rules of discourse are temporarily suspended and the horizontal conventions exert the dominant role.
On the basis of the definition above, Searle’s discrimination between literary works and fictional works also becomes problematic. Searle held that some fictional works are literature while others are not; and conversely, some literature is fictional while some is not—for example, the non-fiction novels In Cold Blood and The Armies of the Night. Whether a work is literature is decided by the reader; whether a work is fictional is decided by the author. Let us look at Truman Capote’s In Cold Blood. It begins thus: “The village of Holcomb stands on the high wheat plains of western Kansas—a lonesome area that other Kansans call ‘out there.’ Some seventy miles east of the Colorado border, the countryside, with its hard blue skies and desert-clear air, has an atmosphere that is rather more Far West than Middle West. The local accent is barbed with a prairie twang, a ranch-hand nasalness, and the men, many of them, wear narrow frontier trousers, Stetsons, and high-heeled boots with pointed toes. The land is flat, and the views are awesomely extensive; horses, herds of cattle, a white cluster of grain elevators rising as gracefully as Greek temples are visible long before a traveler reaches them.” From the literal statement, Holcomb is a really existing place, and the environment, background, and local customs described all accord with reality. The author, moreover, clearly declared his intention to write a “non-fiction work,” that all he wrote was fact, and that the whole work was solidly grounded on in-depth investigation. Searle judges fiction and non-fiction precisely on the basis of the author’s intention and the literal statement or theme. Yet, on the other hand, the work’s stylistic features and narrative mode indicate that the author is writing a “novel.” Although the materials are all true and reliable, non-fictional, it nonetheless follows the narrative conventions of the novel. It leads the reader step by step into a vast, lonesome wheat plain, into this imagined world constructed by discourse. That is to say, from the very beginning the work has, in its own narrative posture, conveyed another non-literal message: it demands that the reader cooperate with it and, together with the work’s language, construct a world and enter this world of discourse. When the reader is drawn by the work and immersed in its world, he has in fact already separated from the real world and taken part in the activity of constructing the novel’s world. “Every sentence of a literary work is a constituent of a continuous chain of discursive performances, ceaselessly unfolding the imagined world opened by the first sentence. Words lure the reader into this imagined world” (Miller, “On Literature,” 38). However much truth the statement of literary discourse possesses, and however much the imagined world resembles or coincides with the real world—even if, after reading these lines, the reader is convinced that all the work writes is fact, that it really refers to Holcomb, Kansas—yet when he is immersed in reading, he has already placed himself within “another” imagined world rather than the real world, and the vertical rules connecting word and reality have therefore temporarily lost their force. “The aim of the constitution of an artwork is to draw you in. Your own subjective consciousness warns you to refuse to be captured; but you are pushed into a vortex, as though hijacked or kidnapped” (Waters, 211). The fiction of which Searle speaks refers to a discrepancy between statement and fact, whereas the literary fiction of which we speak refers to the distinctive performativity of literary discourse: it is the act of proclaiming and starting up the construction of a fictional world by discourse itself. From this we may draw a supplementary view: whether the content stated by discourse is fictional is indeed, as Searle says, decided by the author; but the fictional activity of literary utterance performativity is by no means unilaterally decided by the author—it depends chiefly on the cultural conventions contained in the discourse and on the reader’s cooperation.
The root of literary fiction’s special importance in the modern age lies precisely in the fact that discourse has constructed a fictional world. This is an “alien world” distinguished from the real world; the autonomy of literature and aesthetic self-legislation depend upon this independent fictional world. Searle’s “assertion under pretense,” though it has no real referent, is nonetheless a kind of statement and reference of discourse; it differs from the act of discursive construction, for its aim is not to create an independent world of discourse and draw people to take part in it, and the many modern features of literature thereby lose their ground. Nonetheless, this fictional statement also helps us enter the state of literary fictional activity, for by the absence of the real object it cuts off the direct connection between statement and reality, making it more possible to force the reader to turn attention toward the non-verbal acts of discourse and thereby highlight the constructive and illocutionary force of utterance performativity, beginning to enter literary fictional activity. Yet if the reader remains entangled in the problem of the truth or falsity of the statement, this possibility is cancelled. On the other hand, a non-fictional statement, so long as it conveys some literary message in a narrative posture or other non-verbal act, can likewise lead the reader into literary fictional activity. For instance, the line-arrangement of poetry, or paratexts indicating genre, and the like, are all effective strategies that emphasize utterance performativity and influence and lead the reader into literary fictional activity. On this basis, a long-standing puzzle of Western poetics also seems soluble. In Western poetics, fictional poetics and formal poetics have always been at loggerheads, neither able to give a holistic prescription of literature. If we do not define literary fiction by the truth or falsity of the statement, but, from the angle of utterance performativity, regard literary fiction as a kind of capacity for the act of constructing a fictional world, then the contradiction between the two can be dissolved. Even lyric poetry, from the literal angle, truly expresses the poet’s self; yet the form of poetry likewise leads the reader to construct a fictional poetic realm rather than enter the poet’s mind at will. Hence our view: such things as fictional statement, stylistic features, formal features, rhetorical strategies, and various devices of defamiliarization may all implicitly contain cultural conventions, thereby altering the relation between human and discourse, highlighting the performativity and constructive character of literary discourse, and proclaiming the production of a literary fictional world. It is here that the various modern features of literature can be formed.
In the view of scholars such as Fredric Jameson, the human being can deal with the world only through language and is fundamentally unable to face the in-itself world directly, because human sensation and experience are all determined by language, and the world too is determined by language, ordered and formatted by it. Hence all sensible cognition is a linguistic construction, and all methods of cognition are themselves already various languages. The world we cognize is always partitioned and determined by language, ordered and formatted according to linguistic structure—not the world itself. On the other hand, language is again unable to determine itself. The “difference” on which language depends “is a deferral in the essential sense of time; its structure is wholly process, and can never be made to stop and become a static present; that is, the moment we become aware of it, it flashes past us; therefore its presence is both being-there and not-being-there.” Language is historical; it dimly layers the traces left by history, and it is itself a fleeting trace. Language silts up the dregs deposited by time; “language is a gaping mouth or an aperture onto the not-I,” ceaselessly devouring past, present, and future, devouring self and not-self, so as to fill its own emptiness. Its meaning is always flickering, ungraspable. No one can truly master language; on the contrary, the human being is himself imprisoned in the cage of language. Hence Jameson says: “The very problem of the relation between thought and language exposes the metaphysical view of ‘presence’ … all these notions are, at bottom, various expressions of the metaphysical substance of that crucial doctrine of absolute presence … Derrida’s own problem is precisely that he, too, belongs to this tradition, unable to free himself from its language and conventions, and is bound to fall into a predicament” (144–46). The human being can grasp the world only by using language, and the meaning of language itself is difficult to determine and forever shifting; hence the human being is, after all, utterly incapable of capturing the world itself; even what is captured can only be the so-called “world” filtered, distorted, and even obscured by language—the “world-picture” of which Heidegger speaks, a flickering, indeterminate illusion of the world. The world can only be a linguistic world, and language is forever falsifying and fictionalizing this linguisticized world.
It is precisely the view of language above that dissolves the boundary between the fictional and the non-fictional, and seems also to dissolve the boundary between literature and non-literature. Yet in fact this view of language is still unable to strip literary discourse of its distinctiveness. Although the view above emphasizes that language cannot state and refer to the real world, that what it states and refers to is only the world “reconstructed” by language, a world disguised and fictionalized by language—still, this is, after all, a statemental activity that “points toward” the real world and establishes a close relation with it; it is not the “construction” of a world in the performative sense that literary discourse possesses. The two are situated in different intentional relationships. That is to say, statement does not aim to create another world, but merely uses discourse to let the “same” real world begin and appear in a new way, intending to connect discourse with reality; discourse is only a means and tool for taking part in worldly activity; it differs from literary discourse, whose aim is the creation of “another world.” In statemental activity, the relation among person, language, and world is expressed as: person—language—the world reconstructed by language (or “world”), language being only the human being’s tool for connecting with the world. Whereas in the fictional activity of literary utterance performativity, person and discourse stand in a relation of mutual action and jointly construct a fictional world, while the relation between discourse and the real world is suspended. In Intentionality: An Essay in the Philosophy of Mind, Searle analyzes the model of the speech act. He points out that the assertive speech act may be regarded as matching, in a certain way, an independently existing world, and therefore has the word-to-world direction of fit. The directive speech act, by contrast, is held to bring about a change in the world, so that the world matches the propositional content of the speech act; it has the world-to-word direction of fit (“Intentionality,” 7). Different from these two types of speech act, the act of discursive construction in a literary work suspends its own relation to the world, abandoning, in free imagination, the direction of fit between word and world; its model of the speech act is different. While the relation between the speech act and the real world is suspended, the “tool” status of discourse and words finally sloughs off. In the statemental act and in the constructive act of literary discourse, there is an utterly different intentional relationship among person, word, and world.
To construct an independent fictional world, one cannot simply regard discourse as a tool. As statemental discourse, words often have only an instrumental character; their words are artificially squeezed, regulated, fixed, made into dry, dead abstract concepts, fastened on the chain of logic. Only so are they more convenient for human deployment and use, sparing the human being added encumbrance. And once the referred object (however much it is distorted and fictionalized by language) appears, the word itself vanishes, cast aside—this is the so-called “once the fish is caught the trap is forgotten,” Ingarden’s linguistic “perspectival foreshortening,” or Gadamer’s “self-forgetfulness” of language. While language becomes the human being’s tool, the human being is himself imprisoned by language. The human being is caught in the pale net language weaves and cannot save himself. The performativity of literary discourse is different: it aims to construct a fictional world rather than simply refer to objects, and so it needs to mobilize all the potentialities of words—their sound and rhythm, image, color, meaning, even the linguistic unconscious; from discursive form to content, all the potentials of utterance performativity are fully excavated and mobilized, taking part in the activity of constructing the fictional world, constituting what Ingarden calls the “harmony of metaphysical qualities.” At the same time, the human being’s activity, his imagination, feeling, and whole soul, are also fully stirred and mobilized. Human and discourse exchange, influence, and generate each other, hand in hand building a poetically luxuriant republic of language, a vivid, imagistic, vitality-filled fictional world. As Gadamer says, the language of poetry is not “objectivity” but “the full unfolding of linguisticity, which includes within itself all the forms of language usage and linguistic shaping” (605). Hence, in literary creation, it is not that the author “uses” discourse to reconstruct the world, but that the author “invites” discourse to create the world together. The author seeks out discourse and words, searches painstakingly; discourse and words, in turn, open the author’s heart, kindle the author’s imagination, set ablaze the author’s life, and accompany the author into a strange new world. Author and discourse and words alike fully unfold their own potentials. On the other hand, in literary reading, literary discourse invites the reader to create the world together with it.
To put it from another angle: for ordinary statemental discourse, its statemental and referential functions most readily draw attention and occupy the dominant position. When literary discourse uses various means to strengthen and highlight the constructive character of utterance performativity, temporarily suspending the vertical rules that connect word with the real world, this strips discourse’s statement and reference of their original dominant position. And precisely because this dominant function is temporarily interrupted, the human being has to give up his concern with the external objects to which discourse refers and shift his focus toward the fictional world itself, toward the speech act and the word itself; thereupon discourse and word begin to possess what Bakhtin calls “self-consciousness.” “The word is liberated from the names of the objects, phenomena, acts, or features of everyday existence and recovers the primal power of creating its own reality” (Miller, “The Form,” 40). All the force and sensuous features of the word are thereby excavated and displayed; the unconscious content originally sedimented in the word also surfaces; the word displays its geological strata and infinitely deep implications, displays its hazy atmosphere and its allure of ambiguity. The word no longer exists in the manner of a hollow concept; it regains a flesh infused with vital breath, brimming with spirit, feeling, and force. That is to say, when the human being does not use discourse to refer to external objects but concentrates on discourse and its act itself, the word is no longer merely a tool, no longer a hollow concept. Human and word confront each other face to face, respect each other, establish a relation of equal dialogue—“intersubjectivity.” Once the relation between human and language is converted from the original instrumental relation into a relation of intersubjectivity, both human and language are liberated. This is the freest and most fulfilled relation: in the act of mutual cooperation, both parties safeguard their own independence, autonomy, and productivity, while opening themselves most fully and realizing most fully their mutual conjunction and fusion. Thus the “poetic function” of language of which Formalism speaks is fully exercised, and the creativity of human and discourse is most fully unfolded.
There is, besides, another important reason: everyday discourse tends to stress the act of signifying, which inevitably emphasizes the semiotic character of language, whereas literature, having strengthened the constructive character of utterance performativity—and this constructive act often springing from non-verbal cultural conventions—highlights the non-verbal character of discourse. Since it is “non-verbal,” there is no threat of conceptualization; it can only be non-conceptual, non-formatted—a flowing, ever-changing illocutionary force. By means of this flowing, changeable force, literary discourse sweeps the human being into the fictional world, alters the relation between human and language, alters the relation between human and world, and alters human and language themselves. From this angle, “language has no fixed form; it is, rather, energeia—as Humboldt called it; it is not a tool, for it is ceaselessly produced in being used; nor is it a cage; on the contrary, it has in essence an infinite capacity, for it reforms and transforms itself by opening toward difference” (Weinsheimer, 124). In literary fictional activity, discourse is no longer statement in the ordinary sense, no longer a row of word-concepts logically bound and neatly arranged; with its illocutionary force it shatters the logical chains binding it and breaks the conceptual shell of the word, in order to regain freedom and vital force in a brand-new countenance, merging into the eddying flow of force. The performativity of literary discourse brings the dawn of hope for the human being’s breaking free of the cage of language and the liberation of human being and language itself. Hence Heidegger says: “Language is not just one tool among the many a human being possesses; only language affords the very possibility of standing in the midst of the openness of beings … Only where there is language is there world … Language is not a disposable tool, but that event of appropriation (Ereignis) which possesses the highest possibility of human being” (314).
From the analysis above we may draw a view: a literary work, by various means, demotes the prominent position of statement and reference, strengthens and highlights the constructive character and illocutionary force of utterance performativity, and in so doing establishes a new intentional relationship between human and discourse; the word thereby escapes its instrumental status and conceptualized existence, regains a full flesh and an inexhaustible creative force, and the human being secures for himself an idealized free existence.
Having expounded the distinctive performativity of literary discourse, we must, in reverse, make a correction to the view above. We hold: a literary work, while it highlights utterance performativity and proclaims the construction of a fictional world, does not abolish the referential function of discourse; it often delays the referential activity and alters the referential aim and mode. When the reader is intoxicated in the world of the work, the function of discourse referring to real objects is indeed temporarily halted. But once reading is interrupted or ended, or one turns to read with a detached attitude, the reader extricates himself from the literary fictional world and returns to the real world; the vertical rules connecting discourse with reality once more take effect, and the referential function of discourse is restored. Yet the fictional context constituted by the work’s whole cancels discourse’s capacity to refer determinately to real objects; instead, through a certain isomorphic relation between the discursive construct and the real object, it refers in a metaphorical and symbolic way. Because of the indeterminacy of the discursive construct and the relativity of the isomorphic relation, this reference can only be a general reference. Yet it is precisely this “general reference” rather than “determinate reference” that endows the literary work with enormous force, making it possible to mount a broad and deep critique of the real world. The literary type dwells precisely in the synergy of the constructive character of utterance performativity and the distinctive referentiality of literary discourse. The constructive performativity mobilizes all the means of discourse and all the connotations of the word to shape characters, depicting their speech and bearing, looks and expressions, inner worlds with minute care, even penetrating to the marrow; and the general-referential function, because it does not refer to a single determinate real object, makes it possible to refer instead to a certain class of object in the real world, to point to every dimension and every corner of historical and real society, constituting interaction, mutual interpretation, and mutual verification between the two worlds (fictional and real), thereby gaining universality and extraordinary critical force. For example, the Ah Q of Lu Xun’s The True Story of Ah Q does not refer to any one real person, yet lets every reader feel as though the spirit of Ah Q were within himself; the work thus critiques the baseness parasitic in everyone. And literary discourse’s naming and writing of “the spirit of Ah Q” made people, for the first time, clearly aware of this deep-rooted chronic ailment. Again, the Gregor of Kafka’s The Metamorphosis transforms from a company clerk into a great beetle. Even so absurd an image and story can, in a symbolic way, signify the painful experience of human beings’ alienation into non-humans in real society, as well as the distorted relations among people, and with enormous power of generalization realize a profound critique of real society. Miller says: “by leaving reality, one returns to reality” (Miller, “The Form,” 71). Yet this return is a return in a wholly new sense. Literature’s separation from reality precisely provides the opportunity for a deeper, more forceful return to reality and a broader critique of it.
Unlike ordinary literary works, in special genres such as the non-fiction novel and biography, discourse’s capacity to refer to real objects can be nearly restored. The time, place, environment, characters, and events recounted in Truman Capote’s In Cold Blood can all clearly refer to real, determinate objects. They often have a double identity: they are at once constituent elements of the work’s discursive world and references to real objects. Unlike non-literary works such as news reports, the novelistic narrative mode adopted by In Cold Blood draws the reader more powerfully, enabling him to follow the narrator’s footsteps deep into every minute detail of the events, deep into the most shadowed inner recesses of the characters, to observe and experience. In the reading process the reader is always drawn by two forces: the artful narration of the novel draws him into the work’s world constructed by discourse; the non-fictional technique strives to pull him back, from time to time, into the real world. The reader thus wanders between two worlds, and the constructive and the referential characters of utterance performativity alternately take the dominant role. Yet on the whole, paratexts marking “non-fiction novel” and the like also play an exceedingly important guiding role in reading.
In expounding the distinctive referentiality of the non-fiction novel and biography, we have already arrived at the border where literature adjoins history, and are very close to historical works. We cannot but discover that historical narrative, so long as it is “narrative,” performs the constructive act of discourse and possesses an illocutionary force that draws the reader into the world of discourse; that is, historical narrative likewise possesses the capacity to construct a fictional world. This is precisely the fictionality of historical narrative emphasized by scholars such as Hayden White, who use it to efface the difference between history and literature. Indeed, if one looks solely at the narrative level, there is no clean boundary between literature and history. To find the difference between the two, one must examine the system of speech acts as a whole. First, we hold, historical discourse pursues the realization of its own referential function; it always strives, by various means, to refer determinately to real time, space, events, persons, and historical circumstances, using archaeological facts, historical records with reliable corroborating material, and the like, to establish the connection between discourse and historical truth—that is, its model of the speech act has the word-to-world direction of fit of which Searle speaks. Once a newly discovered fact breaks an existing historical narrative, it must make adjustment, demanding that discourse fit the historical world. These real, exact spatiotemporal relations and persons, things, and events are like an anchor, letting the otherwise drifting narrative discourse find its place in the long river of history and come to rest. Whereas for ordinary literary works other than the non-fictional, discourse does not pursue determinate reference to real objects; by free imagination it disdains the effort to make discourse directly correspond to and fit the real world. Second, there is the functional structure of the system of speech acts—concretely, the relation between referentiality and constructivity. Although at the narrative level historical discourse, like literary discourse, cannot avoid fictionality, in the two genres the primary-secondary relation between the referential and the constructive characters of discourse is different: literary discourse stresses the constructive act of discourse and adopts various devices to strengthen it; historical discourse stresses the referential act, intending to strengthen the exactness of reference. Both, moreover, rely chiefly on stylistic form and the cultural conventions conferred by genre-marking paratexts to carry out the strengthening. On the other hand, while historical works stress the connection of discourse with the historical world and highlight discourse’s referentiality, they tend to delete as far as possible descriptions unrelated to the main facts and overly formalized matter, demoting the constructive character of discourse, even using extreme forms such as the “chronological table” or “yearly record” to reduce the constructive character of discourse to the lowest level. In fact, the writing strategy of historical works—“taking the year as the main thread, with events following”—already forges the primary-secondary relation between referentiality and constructivity. By contrast, literature often strengthens the constructive character of discourse with imagistic description, or highlights it by means of poetic line-arrangement and rhyme, or simply borrows fantastical subject matter such as “strange tales,” “the supernatural,” and “the demonic” to sever the direct connection between discourse and reality, making constructivity an overwhelmingly dominant discursive force. That is to say, between history and literature there exists, after all, a difference in the model of the speech act and in the functional structure of the system of speech acts.
Besides, equally important is this: “Literary fiction belongs to what Vaihinger calls works of ‘conscious falsity’” (Kermode, 37), whereas history loses the consciousness of the fiction it cannot, after all, escape, or, relying on power, denies this fiction. Any narration is a tailoring and re-creating of fact by a mode of thought and a paradigm of discourse; it is always a kind of fiction. But history, because it intends to fit the world with discourse and believes itself to have done so just right, forgets that it is still fiction, abandons the consciousness of the boundary between reality and fiction, dismantles that boundary, and strengthens the referentiality of discourse—even though in fact the referent is absent, the reference being merely false reference. Literature, on the contrary, is clear about its own fictionality and fully uses this fiction to strengthen the constructive character of discourse, while at the same time securing freedom for itself. Hence we should change Vaihinger’s “conscious falsity” of literature to “consciousness of the boundary between the fictional world and the real world.” We are aware that the world of literary discourse is “another” world, and we willingly and gladly throw ourselves into the construction and experience of this fictional world. History, by contrast, denies this boundary and, wittingly or unwittingly, confuses the two worlds. From this angle, compared with literature, history is closer to myth. Like myth, history is convinced of its own truth and strives, through self-construction, to become an important means by which a people identifies itself, confirms itself, and consolidates itself; literature, too, has this function, and is therefore regarded by Arnold as “national wealth,” yet because it is conscious of its own fictionality it possesses more individual freedom. History clings more to the ready-made collective image of the self in order to gain a sense of stability; literature stresses the individual, opens the self, and pursues a sense of freedom and creativity.
Thus we see: history and literature constitute a gradually changing series of genres, varying with the relation between the referential and the constructive characters of discourse and with changes in the constructive and referential functions themselves. At one extreme are the historical chronicle, the yearly record, and so on; at the other extreme are works with fantastical subject matter or distinctive style (such as poetry); in between are transitional forms, such as non-fiction literature and biography. Because the structural relation between the referentiality and constructivity of discourse is closely bound to cultural convention, when certain conventions, with the lapse of time and the change of context, grow strange and estranged, their influence on the speech act weakens, and the structure of the system of speech acts thereby becomes more unstable. So a historical work—for example, the basic annals, hereditary houses, and biographies in the Records of the Grand Historian—when the real objects it refers to grow ever more distant from the reader and the direct connection ever scarcer, until reference is ruptured, can easily come to be regarded as a literary work. Conversely, a literary work, when with the passing of time its discourse grows stiff to the reader, its capacity for action and constructivity and the illocutionary force by which it draws people gradually exhaust themselves, and the work’s world grows alien to the reader, can scarcely sweep the reader into the activity of discursive construction and the world of discourse; at this point it degenerates into material for historical research, or becomes an object of archaeology. Besides, the reader’s reading habits, reading aims, and state of mind also influence the way he enters the work, his relation to the speech act, and thereby the relation between referentiality and constructivity. Hence he may seek social-historical content within literature, or regard history as a rhymeless “Li Sao.” We hold that the character of a work depends on the model of the speech act and the structure of the system of speech acts, on the relation between the performativity-constructivity and the statemental-referential character of discourse, and on the intentional relationship the reader establishes with discourse—which is bound up both with the features of discourse itself and with the reader, while cultural convention plays an important role between them.
Wittgenstein says: “Use is meaning.” Discourse is many-faceted; or rather, discourse is a system of acts, a multifunctional aggregate: it both states and refers, and performs and constructs. As for reference alone, it has external reference (reference to real objects), internal reference (reference to objects within the text), self-reference (the metalinguistic function), and intertextual reference (intertextuality). As for the performativity and constructivity of discourse, these are more complex still: they are distinct from statement, yet always accompany statement and intertwine with it. In a concrete context and use, discourse may highlight one of these functions while not thereby deleting the rest; in general, these functions are often only in a latent state or background position. The literary work is precisely an experimental field that displays the many-facetedness of the speech act and rehearses the transformation of discourse’s various functions. The writer uses precisely the many-facetedness and changeability of the speech act to display technique and play language games. “When we perform a locutionary act, we use speech: on this occasion, just what method do we use? The speech we use has many methods and many functions” (Austin, 88–89). For example, novelistic metanarration strengthens the self-reference and self-consciousness of narrative discourse and highlights the various features of the act of narration itself. Shklovsky’s “defamiliarization” also stresses the self-reference of the word, but it works chiefly by activating human sensation, thereby letting the word open up and its connotation surface, so that “the stone becomes more like a stone.” Again, literary parody and allusion both strengthen intertextual reference, highlighting intertextuality. But parody, while referring to the original text and discourse, often subverts or mocks them; whereas allusion takes reverence for the classic as its aim, using the allusion to strengthen its own authority, and constituting relations of mutual reflection, mutual dialogue, and mutual elucidation between texts or discourses, achieving a realm of hidden resonance and latent splendor.
The structural transformations among these functions all, at the same time, influence the way and attitude with which the reader engages the work, alter the distance between reader and the work’s world or the reader’s position within it, alter the intentional relationship between reader and text and discourse, and in turn alter the speech act itself, thereby creating all manner of meanings and values. From this we can see: in literary activity, the intentional relationships constructed among person, text, and discourse are exceedingly complex; they intricately interweave, mutually influence, change in an instant, flow without rest—not single, static, stable. The corresponding functional structure of discourse, too, changes ceaselessly, at the same time providing the human being with infinitely varied modes of existence. The intentional relationship between human and discourse, the functional structure of discourse, and the human being’s mode of existence influence one another.
Constructivity is the most characteristic property of literary discourse. On the one hand, the constructive character of literary utterance performativity builds and maintains a literary fictional world independent of the real world—an “alien world” of which Marcuse speaks, an “aesthetic dimension” that opens free modes and infinite possibilities for human existence. When the researcher emphasizes this constructive function of literary discourse, the autonomy, aesthetic self-legislation, and transcendence of literature become its salient features. On the other hand, literary discourse’s capacity to state and refer to reality again reconnects literature with the real world. So, once the researcher fixes his eyes on the statemental and referential character of literary discourse, the activity of constructing the fictional world is interrupted, the vertical rules of discourse regain recognition and take effect, the relation between discourse and reality is displayed, literature recovers its historical dimension, the autonomy of literature is thereby abandoned, and social critique becomes a function of literature not to be ignored. As Frye points out, literary discourse has two tendencies, “centripetal” and “centrifugal,” and the reader’s reading attention moves in both directions at once. But “in all literary verbal structures the final direction of meaning is inward. In literature, the standard of outward meaning is secondary” (64). Hence the conflict of views between aesthetic study and social-historical study of literature has its root precisely in the many-facetedness and multifunctionality of the speech act, in each side of the dispute clinging to one end.
Furthermore, the self-referential act of discourse strengthens the self-consciousness of discourse and highlights the features of discourse itself. If the researcher turns to attend to the formal features of discourse itself, Formalist poetics and Structuralist poetics find their growing point here. Or if he goes further to attend to the cultural-political features of discourse—the relation of discourse to power and ideology, the modes of production, operation, and mechanism of discourse, the influence of discursive rhetoric and illocutionary force on reading, and so on—then postmodern narrative theory and the new rhetoric again have their stage. Literary discourse is itself a system of acts; it links performativity and constructivity with statement and reference, and with self-reference, joining them through; it gathers all these functions in one body and endows literature with many-facetedness. As Antoine Compagnon points out, “literature is a passage that runs through the high wall and links the two (the fictional world and the real world)” (130). Hence we hold: the multifunctionality of the literary speech act destines the many-facetedness of literature, destines that literature must bear multiple characters within it, characters that on the surface are mutually contradictory, even mutually canceling.
For works of different literary schools and genres, the constructivity and referentiality of discourse are in different structural relations and in different modes and states of operation. For example, in documentary literature, naturalist and realist literature, surrealist literature, postmodernist literature, and so on, the relation between the constructivity and referentiality of discourse, and the mode of discourse’s reference, differ utterly, thereby displaying distinct stylistic features. Accordingly, we hold: in the face of the many-facetedness, multifunctionality, and indeterminacy of the literary speech act, an either-or, simple, absolute research attitude is undesirable. Conclusions drawn from isolated analysis from a single angle cannot correctly reveal the actual state of literary activity; they are not only one-sided but also distort the features the work, as a unity of contradictions, actually possesses in that aspect. The task before us is how, between mutually opposed research lines, to seek a coordinated, compromising new line; how, with a flowing and changing vision, from the relation of mutual action among the various speech acts and functions, to synthesize past research, so as to grasp literary activity more deeply and holistically.
In Mind: A Brief Introduction, Searle points out: “Linguistic intentionality must be explained in terms of the discursive mode of the mind’s intentionality, and not the reverse … the meaning of language is derived from intentionality, and it must come from the primal intentionality of the mind” (“Mind,” 144). In Searle’s view, discursive intentionality springs from the author’s intention; the reader must, following discursive intentionality, grope for the primal intentionality of the author’s mind in order truly to grasp the work and accurately read it. That is to say, the intentional relationship the reader establishes with discourse must be identical with the primal intentional relationship between author and discourse. The author’s intention is indeed closely bound up with discursive intentionality, and also connected with the reader’s reading. So just how do author, discourse, and reader relate concretely? How are these relations established? This still awaits deeper analysis. We hold: from a certain angle, the author mainly, through rhetorical strategy, endows literary discourse with a particular intention and illocutionary force, thereby regulating the relation between reader and text and discourse, influencing the mode of the speech act, and so indirectly influencing the reader’s reception and understanding.
For narrative literature, the choice and determination of the narrator is exceedingly crucial. In a literary work, the narrator generally possesses authority; it implies or influences the person, perspective, and so on adopted by the narration—including narrative style and narrative voice (direct speech, indirect speech, tense, dialogic, monologic, and so on)—and influences, even determines, the relation between the narrator and the work’s characters and events. The literary work uses the narrator and its various narrative means to influence the reader’s reading attitude, regulate the relation between reader and text and discourse, regulate the distance and mode of association between the reader and the work’s world and the characters and events it contains, and regulate the way the reader experiences events and regards the work’s world. For example, Goethe’s The Sorrows of Young Werther, with the vividness and intimacy peculiar to the first-person narrator, effectively narrows the psychical distance between reader and the protagonist Werther, strengthens the work’s infectiousness, gives the reader a soul-shaking shock, and even became one of the causes influencing some young readers to “mistakenly identify.” By comparison, the third-person narrator tends to keep a distance from the narrated object, facing the narrated characters and events with a relatively detached attitude; this is more favorable to works that stress real-world critique. And the narrator’s view and evaluation of the world will, to some degree, unwittingly influence the reader’s view and evaluation. That is to say, the intentionality possessed by, and the force and degree of force produced by, the act of doing things with words are closely tied to rhetorical strategy; the author can thereby endow discourse with illocutionary force and indirectly embody his own writing intention and cultural power.
If the narrator is unreliable—“a narrator whose norms and conduct are at odds with the norms of the implied author; whose values (taste, judgment, moral sense) deviate from those of the implied author; a narrator whose reliability of narration is undermined by various of his narrative features” (Prince, 239)—the narrator’s authority may be entirely stripped away. The reader no longer closely follows the narrator’s footsteps or follows his guidance, but, with a skeptical eye, constantly questions the narrator and all he narrates. This situation endows the reader with a relatively detached reading attitude; his relation to discourse, and his way of engaging the work’s fictional world, change. Yet the narrator’s unreliability is, after all, often something the author does on purpose. The author sets up the narrator while unwilling to grant him trustworthiness, restricting or even stripping away the narrator’s reliability; hence, in the general case, this is likewise an embodiment of authorial intention. Brecht’s strategy has its own distinctive features. He uses asides, captions, sets, defamiliarizing performance, and various other devices to “estrange” the audience, from time to time, out of the dramatic plot, letting them watch with a detached attitude from outside the staged fictional world, thereby realizing drama’s critical and educational functions. In novels and other literary works, the author can likewise use frequent shifts of perspective, metanarration, and parody to interrupt the narrative thread and disturb the narrative process, or use devices of the absurd to create distance between reader and the work’s fictional world, letting the reader transcend the fictional world, return to reality, and critique reality. For example, Tolstoy’s Kholstomer lets the reader view the world through the eyes of a little colt, shaking off habitual notions and ideological constraints, while this absurd device does not dissolve the psychical distance between reader and colt, effectively regulating the reader’s relation to the work’s fictional world, influencing his reading attitude, and letting the reader transcend the work’s world from time to time, thereby maintaining critical consciousness.
The process by which the author’s intention is realized through rhetorical strategy is precisely the process by which the author exercises power; it is the process by which the author, borrowing rhetorical strategy, manipulates discursive intention and illocutionary force and thereby indirectly influences the reader. Both author and reader live in social reality, and neither can act at will apart from “convention.” It is precisely this convention—“the totality of social, historical, and cultural norms regulating the function and mode of communication of fictional prose” (Lothe, 19)—that guarantees the smooth progress of reading communication and seems also to guarantee the realization of authorial intention.
Of course, the workings of power in the process of literary activity may be more complex. In Fictions of Authority, Lanser advanced the concept of “narrative voice,” re-endowing narrative discourse with cultural-political content and interpreting narrative discourse with particular emphasis on the feminine standpoint. She says: “In the Western literary tradition of the past two centuries, discursive authority has by and large naturally been ascribed to educated white males within the dominant ideology, differing only in degree of closeness. Hence the degree to which the narrator’s position approaches this dominant social power became the principal factor constituting the authorial authority of discourse. At the same time, the constituent factors of discursive authority also include text-writing strategies that change with the historical process” (6). That is to say, in a patriarchal society, even if a woman author keeps to her own feminine standpoint and strives to utter the woman’s own voice in literary writing, the discourse she uses and the writing strategies she adopts still belong to tradition, to a patriarchal tradition forged by the dominant ideology, and to carry out a thoroughgoing remaking is an extremely arduous process. To utter her own voice, the woman author should possess a clear gender consciousness, vigilantly keep a distance from “dominant social power,” wash away the residual gender bias in discourse, and recast all manner of writing strategies. To resist “dominant social power,” the woman author even has to borrow the voice of the female collective to speak. While writing, the woman author opens up another battlefield, a war declared upon discourse and writing strategy.
The analysis above lets us see: among author, text and discourse, and reader, there seems to exist a single-directional relation of power operation. Even Lanser’s interpretation only makes this power operation more complex. Some research that interprets the power relations in literary activity from the cultural-political angle often follows just this line of thought. Yet within it lies an important defect: such analysis ignores the distinctiveness of literary discursive activity, equates literary discursive activity with everyday discursive activity, and regards literary discourse as a simple tool of communication. That is to say, the analysis above still rests on the intentional relationships “person—language—world (reconstructed)” and “person—language—world,” disregarding that literary activity, besides the general form of discursive communication, also contains another special form of communication, and disregarding that literary discourse, besides its semiotic character, is also a distinctive speech act. In the literary speech act, between human and discourse there is established a most distinctive intentional relationship: the relationship of intersubjectivity; at the same time, literary discourse displays its other, non-instrumentalized countenance. In literary activity, the various intentional relationships interweave most complexly, and discourse and word, too, present the most manifold countenance and color.
In the activity by which literary discourse constructs the fictional world, between human and discourse there is established a new intentional relationship: the relationship of intersubjectivity. It is precisely this intersubjective relationship that endows literary discourse with an autonomy resistant to authorial intention, and endows the reader with the independence of autonomous reading. Language is historically formed; in language, layer upon layer, is silted the mud deposited by history; hence any language, like rock strata, layers historical faults. And language, as a tool of communication, while it connects person to person, group to group, region to region, nation to nation, itself undergoes exchange and fusion. From this angle, language can only be what Bakhtin calls “heteroglossia”: it is fused from languages of different systems, between which exist crisscrossing boundaries and fissures. But in the everyday use of language these faults and fissures are overlooked. As an instrumental language, people have neither leisure nor intent to attend to these ruptures; they live in language as in transparent air and water, and out of familiarity grow numb, no longer seeing the flaws of language or feeling its features. Yet literary activity’s emphasis on the performativity of discourse alters the intentional relationship between human and discourse, endows discourse with independent autonomy, activates all the connotations and life-force of the word, and at the same time opens the ruptures that originally existed. From another angle, instrumental discourse’s emphasis on linguistic logic and real context strictly constrains the expression of discursive meaning and represses all discursive intentions that deviate from authorial intent; whereas literary discourse’s disdain for linguistic logic and its separation from the real context allow discourse itself to harbor countless ambiguities and intentions, and at the same time let these latent meanings and intentions surface unrepressed. “Words are their tradition of use: through concrete utterance they preserve usage and theme. The historical world leaves indelible traces upon the word; hence language cannot be understood apart from its utterance” (Weinsheimer, 119). If instrumentalized, conceptualized everyday discourse can fairly obediently follow the intention of the author (the speaker) and embody the power-will of the author (the speaker), then literary discourse, as non-instrumentalized, non-conceptualized discourse, by virtue of the relative independence, vitality, and existential ruptures it possesses, has the force to resist the author’s manipulation. While it reluctantly accepts the duty rhetoric assigns it, it also stubbornly resists these tasks, resists the author’s power-will.
In Discourse in the Novel, Bakhtin gave a deep account of novelistic discourse as “latent dialogue” and “heteroglossia.” In Bakhtin’s view, the novel is “a microcosm of heteroglossia.” This is so not only because the novel embeds widely different kinds of discourse—letters, diaries, proclamations, philosophical discourse, the imitation of folk sayings and everyday speech—but also because discourse itself is a polyglot phenomenon and a uniquely individual polyphonic phenomenon. Even within a unified national language, it splits into various social dialects, the expressive habits of various groups, professional jargons, various stylistic languages, the languages of various generations and ages, the languages of authority figures or of the lower orders, the modish language, the vulgar language, and so on; and these, moreover, in the process of communication among different kinds of discourse, mutually distinguish and at the same time mutually hybridize and blend, constituting a discourse full of “latent dialogue.” From this angle, any discourse is heteroglot; an absolutely pure, unified discourse hardly exists. “The word is wholly carved up, permeated with all manner of intentions and tones. For the consciousness of the human being who lives in language, language is not an abstract system organized in normative forms, but a concrete view of the world expressed in heteroglossia. All words bear the scent of profession, genre, school, party, particular work, particular character, a certain generation, a certain age, a certain day and hour, and so on. Every word bears the scent of the context of the tense social life in which it stands; all words and forms are full of intentions. Words inevitably carry the resonance (of genre, school, individual) acquired in their context” (Bakhtin, 74). The novel uses precisely these social heteroglot phenomena and the uniquely individual polyphonic phenomena grounded on them to master its subject matter, tell stories, depict characters and the work’s world. Although the author strives to forge them into a perfect artistic whole, making heteroglossia subordinate to the “highest rhetorical whole,” he does not expel the centrifugal tendency of heteroglossia or smooth away its divergences; on the contrary, he lets heteroglossia mutually resonate and converse. Literary discourse, as what Bakhtin calls a “higher form of linguistic life” possessing “self-consciousness,” makes the dispersive features within discourse, and the divergent and even mutually conflicting intentions among words, become activated, highlighted, and brought into play.
De Man says: “The speech act inherent in language is independent of intention, or of any motive and wish we might have” (de Man, “The Resistance,” 96). In a literary work, the various discourses and words are always unwilling to wipe away the imprint of their “identity,” unwilling to suppress the manifold intentions clinging to them; they constitute a brilliantly colored, polyphonous heteroglossia. In this heteroglossia gather all manner of distinctive viewpoints for observing the world, all manner of ways of understanding the world and special attitudes reflecting the meaning and value of things; all these resist the highest rhetorical whole the author imposes on the work, deconstruct the unified vision the author strives to construct, resist the author’s power to enforce his intention, and at the same time corrode the ideological character of the work. And if we further consider the transcendence implicit in the constructive act of literary discourse, then the problems of power and ideology become more complex still.
From the angle of the relation between reader and discourse, the literary fictional world grants the reader the autonomy of reading, letting him, according to his own will and even sudden inspiration and whim, select from discourse what most attracts him, compose his own independent vision, and thereby subvert the author’s intention and the mainstream ideology’s control over the work. In particular, the cultural conventions contained in discourse are only an “invitation,” a “guidance,” a “should,” not a “must”; they are not a coercive “law,” and the reader may altogether disregard convention. The practice of disregarding cultural convention and interpreting the text purely according to one’s own taste is precisely the strategy of deconstruction. Commenting on Derrida, Abrams pointed out: “He plays a double game: when reading another’s text he introduces his own interpretive strategies, but when he wants to communicate his interpretive method and results to his own readers, he tacitly relies on the norms of the community” (272). Reading a work, he may violate cultural convention to make his interpretation startling; but to convey his own views, he cannot but honor cultural convention, to ensure that others can understand according to his intention. And the tangled, manifold inner intentions of literary discourse, together with its disdain for logical relations, precisely provide the best site and opportunity for deconstruction to play its “double game” using “double rules.” So when the reader does not submit to the cultural conventions contained in literary discourse, does not follow the dictates of rhetorical strategy and the illocutionary force of discourse, but indulges his own will, entering the text from a side, secondary, or marginal direction, and from time to time altering his relation to discourse, altering the relation between the performativity-constructivity and the statemental-referential character of discourse, then he may dismantle the work’s fictional world into pieces, and the work’s meaning and its ideological character meet with deconstruction. This is also why we can always read overtones from literature, read meanings the author never anticipated, break through habitual interpretation, and gain extra rewards. We hold: literary discourse is a prism; it provides the reader with various different paths into the text, and thereby provides the reader with various different ways of interpreting the work and regarding the world. This also means that literature is at once subject to cultural power and always escaping its grip; literature is at once enveloped by the dominant ideology and corroding it. Literature provides for human beings a free road out of totalitarianism.
In the whole process of literary writing and reading we discover all manner of conflicts: on the one hand, the author’s writing is a struggle, attempting by various means to assert his own power and push his own intention; literary discourse, however, with its dispersed, divergent intentions and faults, disturbs and resists the realization of authorial intention. In fact, the author cannot but open battlefield after battlefield in writing—a war against tradition and social ideology, against literary discourse, words, and writing strategy—and this is, moreover, almost a hopeless, desperate battle with little chance of victory. On the other hand, the reader must in turn declare war on the tangled, manifold discursive intentions and faults; he strives to comb and forge discourse into a whole conforming to his own will and desire; literary discourse, however, flickers, flees, and obstinately holds to its self-divergent intentions and ruptures, opposing the reader’s effort, so that any reading can only be “misreading.” Reading is, as de Man says, “an act of understanding that can never be obeyed or confirmed” (de Man, “Blindness and Insight,” 107).
At the same time, we discover all manner of cooperations: author and literary discourse converse with, stimulate, and generate each other. “Writing is initiatory, in the fresh sense I give the word, so that it is at once dangerous and disquieting. It does not know where it is going; no wisdom can keep it from rushing headlong toward the meaning it constructs, toward its future. Its willfulness can be gained only in the relaxation of this impetus. Therefore it has no insurance. To the writer, writing is a first navigation—even if he is no atheist but merely a writer—and this navigation has no grace” (Derrida, 17–18). In literary writing, these dangerous, disquieting words, not knowing where they end, after all bear the author along and begin together a “first navigation” without grace and without harbor of rest. The reader, too, is dragged onto the “ship” constructed by discourse; his fate is tied not to the author but to the text and discourse. While discourse invites the reader to construct the work’s world together, it also reconstructs the reader, leaving an indelible imprint upon his personal fate.
The speech act is the act of the human being. From this angle, language is both “masterless” and, in the concrete operation of discourse in literary writing and reading, at the same time “mastered”; in the moment of its free dissemination it is always continually stamped with the personal seal. At this point we further discover: language can forever be only human language; it keeps its own independence and at the same time is at every moment entangled with the human being. It is both collective, bound to the touchstone of the collective, and yet personal, its relation to the human being in ceaseless change. The changeability of the system of literary speech acts provides the human being with manifold paths into the text, and thereby forges the many-facetedness of literature. To discard the human being, to drive the human being out of the world of discourse, and to discuss literature by language alone, can only be a hopeless battle, for the human being has, in advance, already exiled himself.