2. What can literature be?  

    This form of the question may turn out to be more productive than the usual "what is literature?" Like all texts, literary ones are mere artifacts with an indeterminate communicative status until someone does something with them by applying more or less appropriate and relevant conventions. Despite the prevalence of literature in so many cultures, the nature and force of those conventions have not been well explained. The very term "literature” is rather indiscriminately applied --- a collection (or “canon”) of works (e.g. “great books"), a school subject that assigns those works, a topic of criticism, a facet of human culture, and so on. How far all these informal definitions coincide is hard to tell. Little is achieved by defining “literature” as the set of texts society agrees to designate as such. We merely restate the fact of literature, without explaining the preconditions why some texts are admitted over others, why the set changes, and so forth. Evidently, literature can fulfill some functions societies consider worthwhile, though obviously, not all of it can do so for everybody all the time. The question is:  what criteria and activities can be involved in those functions?

     In the past, scholars influenced by linguistics and stylistics would define literature as having "distinctive features" of language that “deviate” from “ordinary language." But this definition involves serious problems. It presupposes an objectified text with independently given "features.” Prose is less adequately handled than poetry. No provision is made for admitting texts from other types as literature, such as “found” poems; or for using ostensibly literary features in other types, such as advertising and political oratory. Anyway, since most texts could be found to deviate in at least a few features from most others, not much is left on the side of the “norm," which seems highly illogical. Ultimately, the “distinctive features" of a text are more a result of the critic's projections and predispositions rather than an original cause. Far from solving our problem, “deviation" merely  rather proposes one more factor to be explained.

    A second possible definition, less fashionable among modern critics, is that "literature" has a special content. It conveys certain ideas or topics, such as we might find in cavalier, pastoral, and sacred poetry. But we encounter much the same problems here. We are objectifying the phenomenon, this time to siphon off its “ideas” instead of its “features.” We are implicitly favoring poetry over prose. And we find the ideas, whatever we decide they are, changing over time or being involved in plenty of texts we wouldn't want to call “literary." Unless we ask why certain ideas might be assigned to particular types of texts, we continue to pursue peripheral effects more than central causes.

    A third definition, in favor among philosophers, is that literature is “fictional,” conveying statements that are not "true" or don't “refer" to anything. But plainly, not all fictional texts qualify as literary; the ordinary “lie" displaces the “truth” but gives no rewards. Nor are all literary texts considered fictional, as we see from documentary art. Moreover, the decision about what is true and in what way is anything but a straightforward yes-or-no matter. "Fictions" refer to entities that must “exist" in some sense, or else we couldn't understand the text. Conversely, “facts” are partly created through our acts of perspective, selection, organization, narration, evaluation, and so on, or else we couldn't talk about them. A novel can be “true to life” although all its characters and events were created by the author's imagination. And a historiographic text has to continually reinvent history by interpreting in hindsight and imposing representational, narrative, or dramatic criteria, often carried over from literature. So historical facts cannot be the opposite pole for literary fictions.

    A fourth definition could be that literature is "rhetorical,” composed with special "figural" techniques of substitution ("tropes”) and arrangement (“schemes"), whereas other texts are “logical,” composed with a “literal" dedication to truth and consistency. But, like “fiction” versus “fact,” this distinction, presupposing “rhetoric” to be optional ornamentation or persuasion, turns out on closer inspection to be another complex gradation. All language is used rhetorically or figurally at least when we substitute words for things, and the arrangement of discourse for the arrangement of the world. All language is also used logically or literally when we give things coherence, identity, and category. At most, these two aspects attain different degrees of prominence and deliberateness in specific cases. Yet even a high degree of rhetoricity doesn't necessarily make a text “literary."

    All these attempts to isolate something specific “in" literature remain unsatisfactory. My general conclusion would be that literature can only be defined with a functional description of what happens when people produce or respond to it. The principle I consider most plausible might be called “alternativity.” Participants in literary communication should be willing to use the text for constituting and contemplating other “worlds” (i.e., configurations of objects and events) besides the accepted “real world.” The text need not appear “fictional” by directly colliding with everyday reality. It may fall anywhere between the extremes of the fantastic and the documentary. Yet the possibility must be left open that whatever world the text is thought to elicit should be related in some interesting and informative way to reality and should show us the latter in perspectives we might otherwise not consider.

    This account would explain a society's concern for literature with a more compelling motive than "linguistic deviation" or "fictionality” alone. Since, as psychologists and phenomenologists have found, every society's approved version of reality has to omit or deny certain potential aspects or perspectives, an institutionalized forum for presenting and developing excluded alternatives ought to be necessary and useful. The limitations imposed by common sense or official consensus about how the world "really" is can be transcended there without causing widespread disorientation and conflict. Literary authors are not normally reproached for reporting things they never saw happen, or for transforming things they did. Readers are more inclined to tolerate these actions as a means for sampling diverse visions.

     "Poetic" texts would be those during whose use the principle of alternativity is extended to discourse itself. Here too, obvious deviation from ordinary discourse is not required, though often employed to offset the seeming transparency of language. Texts not classified as "poems" can readily be given a poetic function if the organization of their language is regarded as one among several alternatives. Ideally, just as literature as a whole sharpens our sense of the world, poetry sharpens our sense of language. Moreover, the more complex medium of poetry, renegotiating both reality and discourse, can have an especially powerful impact that enables poems to be esteemed as highly significant and enduring expressions.

    The consideration of literature and poetry in numerous forms of human education over the centuries signals some hope that using texts this way could bring far-ranging benefits for the general capacities of the human mind. The danger of unduly stabilizing the appropriation of language and the world might be counteracted. Experiences could be attained that would normally be difficult, hazardous, or impossible. The human range of understanding could be expanded and refined far beyond the exigencies of individual behavior. However, we know such results face serious impediments. The prestige of the literary text makes it particularly prone to being objectified, so that the reader's role is hardly noticed. The writer becomes a conspicuous public figure with a cultural mission of regulating meanings. As professional readers, critics intervene with circumstantial materials and proposed interpretations unavailable to most people. Teachers at all levels of schooling control and restrict literary meanings or distract away from textual experiences by overstressing technicalities of history, biography, genre, or trope. Eventually, the ordinary reader may consider his or her own creative use of the text marginal if not unauthorized.

    In such a complicated situation, I could hardly claim to address what literature "is," because it “is" what people do with it, and those activities are quite diversified. Instead, I might try to sketch what it can be if the principle of “alternativity” is allowed to be a dominant function. Most of the critics examined in Part II would agree that this allowance should be made if literature is to attain its proper valence. But they do not all acknowledge the extent to which aesthetic or critical theories imply a utopian imperative to conceive and experience literature in ways we cannot take for granted in our culture. These factors reflects the comparable imperative, implied by literature itself, to understand our understanding of ourselves and of our reality: the more persuasively and urgently implied, the “greater” a work will seem.

    Scholars have long concurred that literature can elicit a change in how we experience. Aristotle described a renewal of perception, and this notion has returned in various guises in most aesthetic theories ever since. The contention of the Russian Formalists and the Prague Structuralists-that literature "estranges" or “de-automatizes” the processing of a text-sounds plausible in view of recent research on perception and cornprehension.² [2. For surveys of this research, see the sources cited in Note 1. On implications for relating psychology and literature, see Beaugrande (1986); Groeben (1980a, 1980b, 1982); Meutsch (1986); Schmidt (1982).] Everyday reading seems easy, if not automatic, because so many sectors of the complex process are done without attention and thus do not consume much resources or compete with other operations. In return, a high correlation is required between what is expected or predicted and what is perceived or understood. The "alternativity” of literary experiences modifies that correlation. Even if we know in a general manner what the text “is about," because it is famous or already read or fits a genre we know, and so on, the actual experience differs both in quality and quantity from an encounter with comparable materials in non-literary communication. The ostensibly “same” words, things, ideas, and so on, are endowed with distinctive and intensified functions. Aspects we consider special, such as "deviant" features or "fictional” statements, may act as cues and reminders in this process, but they are not indispensable. We could read any text in a literary and aesthetic fashion; but we usually don't when we feel no imperative.

    Potentially at least, the literary experience rewards the rise in effort and complexity by expanding and diversifying possible meanings. The reader's response is governed by the intent to control what ensues and to attain some worthwhile result. While there may be no laws or strict rules for processing a text as literary, we certainly don't proceed at random. As with all texts, the decisive principle is the motive to attain a systematic experience in which particular perceptions, hypotheses, significances, and so on, are assigned a current relevance and value. Failure to do so leads toward disintegration, an unpleasant and disorienting state for the mind, especially if occurring on a large scale or over a long time.

     This tendency to operate systematically and to resist disintegration helps people to have comparable experiences with the same literary text, though divergence is typically greater than for non-literary communication. The proclivity for the systematic is carried over from our dealings with “reality,” but with a major difference. in everyday experience, the advantages of a consistent, reliable environment seem great enough to be worth ignoring or explaining away whatever disrupts coherence. For the world of a literary work, we are willing to be more active in creating such an environment. The main channel of experience passes through the work itself, so that a writer or reader can justly feel that here the coherence of the world is partly his or her own achievement-a sensation much harder to entertain regarding everyday reality. Whereas orientation and management in the real world may be provisionally aided by objectifying it and disclaiming our own role in creating it, no such directive is strategic for a literary world. Thus, our accomplishment there can be a “super-coherent" experience of reinvesting our conventional organizing of reality into a more creative project. The intensity of the engagement is its own greatest reward.

    With that goal in view, an author can anticipate the kinds of organization that should encourage such an experience, for instance, by invoking or combining frameworks of conventions or by posing problems to which the author's hoped-for resolutions are among the more satisfying. Yet total control is impracticable and, no matter what the author might think, undesirable as well, because it would limit the work's potential for continued use. A work that tries to give exclusive, pat answers to its own questions impinges on the reader's role and response. As we will see later, critics can use this ratio between closure and openness as a measure of a work's value.

     It should follow from the argument so far that literature both stabilizes and modifies not only our communicative potential, but also itself. As a system, it elaborates its own peculiar methods of "systemizing" and eventually must transcend its own standards in order to maintain its functionality. Conventions such as generic forms become stabilized as integrative frames for managing the complexity of individual acts of writing or reading. In exchange, authors and readers feel impelled to innovate against these frames, at least over longer periods of time. Ideally, an author never creates the same work twice; and a reader has a new experience with each reading, even of the same text. In practice, innovation must conform to personal dispositions and skills; under favourable conditions, much more could be done, I suspect, than usually is.

     Traditionally, the 'aesthetic" aspect of art has been defined as some interaction between "diversity" and "Unity."³ [3 Compare the formulations in Leibnitz (1720); Hegel (1835); Fechner (1876); Gunzenhäuser (1962); Schmidt (1971). On the derivation of frameworks for organizing perception, see especially Gombrich (1960).] In the usual logic of experience, things are either related and hence significant for each other, or they are not. The 'aesthetic" experience tends to subvert this either/or logic with an imperative to consider further modes of potential relations and significances. Hence, the aesthetic mode is particularly utopian in the sense of Chapter 1: anything may point away to everything.

    The conception of aesthetics as a unifying of diversity has been recently challenged to he a view fit for "classicism,' but not for "modernism. " However, the question is a matter of degree. A work appears "classical' when its unity is considered a necessary entailment of its organization; and "modernist" when that unity seems powerfully resistant. Hence, classicism emphasizes 'harmony, " that is, a deliberate equalizing of parts and features and an adhering to some "canon" of prescribed qualities, typically conceived as 'imitations' of nature if the latter is deemed the epitome of order. "Modernism' strives for a disequilibrium of parts and a defiance of canons. Yet both tendencies are equally "aesthetic' as projects for regulating the interaction between a 'diversity" and a 'unity" not 'in' the work, but in the context a person produces for the work. "Classical" standards are intended to make the applicable contexts immediately available to an initiated public, so that the harmony seems to be inside the work. "Modernist" standards are intended to evade or disrupt available contexts for a largely uninitiated public. No doubt most periods in the history of the arts had both tendencies; but the nature of the process conveys the impression that the only "Modernist" art is whatever appears so right now.

    Both sets of standards entail sufficient disadvantages to prevent either one from becoming an absolute. For its detractors staunchly committed to earlier canons, "modern art' is simply not “art" at all. Yet "classic art" is, if anything, so readily accepted that few people can experience it with its original impact, and its functions dissipate. The 'classic" work seems monumental, inevitable, and the response to it overly rehearsed. The "modernist” work seems calculatedly abrasive, and the response overly opportunistic. Each innovation, if it succeeds in winning an audience, is channelled through the same process whereby "modernism" is relentlessly “classicalized.”

     How to save art theories from this same rigidifying success has become a major issue in criticism. Just as artists may vie in their endless search for some technique that can never seem natural, harmonious, beautiful, and so on, critics may cast about for some theory that dispenses once and for all with such aesthetic fundamentals as “diversity" and "unity.” Yet the "aesthetic”' is well-equipped to absorb its own contraries. Each new critical project highlights new perspectives and shifts the perceived ratios of familiar ones, yet without cancelling the latter. So we can't depend on the standards of any one period to tell us what art “is.“ We can only explore what contexts (canons, conventions, archetypes, aspirations, and so on) people bring to it; and what institutions claim authority to describe or control this process.

     History undeniably reveals an enormous range among the texts accepted as literature and poetry. This diffusion is to be predicted if alternativity operates both against everyday reality and against any one alternative frame whose total predominance would dissolve the main function of the literary experience. Every trend eventually undermines itself. Romanticism was followed by Realism; Realism by Surrealism; Surrealism by Documentary Art; and so on. Hence, the fit between a work and some class of “genre,” “style," "movement,” and so on to which it may be assigned is inherently tenuous, making the class itself far more problematic than literary studies has traditionally admitted (Ch. 3). In a functional definition, distinctions between literary subtypes are just as much gradations as those between literature and other text types.

     Though a functional definition does not decide which texts must be the literary ones, it suggests that literature stands apart from neighbouring communicative domains in terms of its uses. History, philosophy, and theology can all overlap with literature, but in each case, the dominant function is distinctive. I shall briefly sketch some lines for describing those distinctions.

    Historical texts are mainly rated for their accuracy respecting specific “facts,” whereas literature is rated for its insights into what is in principle revealing about the human situation. (Few critics would insist that Hamlet and Lear are bad plays because their historical foundations are distorted or doubtful.) The use or discovery of quasiliterary representational and narrative techniques in historiography enables some history texts to survive long after their status as factual accounts has been undermined. Their “past" becomes another alternative world to the present, no longer purporting to be the latter's ancestor or cause.

    Philosophy differs from literature by emphasizing explanation over representation. Any represented aspects of reality illustrate explicitly presented conceptions, usually within a single schematic. Literary authors who follow this trend seem to be "philosophizing," and as readers we may feet free to believe their realities mean something quite different (cf. Ch. 6, 8-9, 12, 14, 17-18). The philosophizing so profuse in literary works of the eighteenth and early nineteenth centuries became unreliable, ironic, and finally vestigial, without impairing the literariness of the domain at all. Today, we can esteem works like Candide or La nouvelle Héloise as literature without valuing, or even taking very seriously, the author’s philosophical ambitions. And we can prize authors like Shakespeare, whose philosophical outlook we might well despair of reconstructing. A functional differentiation between the literary and the philosophical is therefore feasible, though often evasive.

     The situation appears similar in theology. Here, the representation of reality is still more predominantly illustrative, though its explanation proceeds less by logical argument than by assertion of dogma. And that dogma has to be accepted as a matter of orthodox belief. A refusal to do so is a dangerous heresy, which is why wars are much more likely to be fought over religion than over literature or philosophy. To read a sacred text like The Bible as literature is to situate its other-worldly reality within a spectrum of alternatives, a heretical move for the true believer. Reciprocally, literary works whose theological groundings were conceived as alternatives to orthodox faith, such as created by Milton, Blake, or Yeats, get higher literary ratings than a work of the conventional piety we expect from a church hymnal; but they seldom become central texts for an institutionalized religion.

    A functional approach might also help to clarify the traditional demarcation between "good” and “bad" literature, or “high” and "low." To create a literary work is to promise an insightful experience. This promise by itself is ambitious in a medium replete with “classics,” and the possibilities of failure are manifold. The reader may feel the experience wasn't rewarding, or that the author didn't manage to convey what was proposed. If, as stressed by critics like Iser or Bloom, the reader re-creates the work, then such lapses are readily imagined (“that's not what I would have done").

    Also, the engagement of the reader's personality and ideology may deter the enterprise. Proposing alternative realities and questioning the prevailing one can offend those whose investment in that reality is heavy yet inadequately reasoned. Conversely, conventions may be put in question that the reader barely subscribes to anyhow, for example, after the liberalization of Western society passed beyond the ideologizing initiatives of nineteenth-century “reformist" novelists. The original function no longer seems vital, though others may be found, such as a sharpening of social awareness in unfamiliar contexts. Specific values, even very widespread ones, can undergo radical changes. For instance, "sentimentality" was highly esteemed in eighteenth-century literature, especially in the novel, as the disquisitions of Fiedler and Jauss expound. Current critics routinely denounce it with equal emphasis. If we adopt a modified version of Schiller's concept of “sentimental" art reverting to earlier forms, sentimentality could be defined as a regressive, usually emotive contemplation dissociated from convincing motivations. To be sentimental is to regress toward childhood or adolescence, when emotions were less stable and integrated; and to displace emotions from occasions when they were genuinely felt over to those where they are artificially or gratuitously indulged. The once-revolutionary gesture that uplifted emotion for bringing insight by challenging the supremacy of reason (e.g., in Rousseau) becomes a reactionary blurring of insight by refusing to reason beyond sensation.

    This example suggests that values might be restated in terms of how readers feel encouraged to respond. Yet every inquiry into values runs up against the fundamental ambivalence in the potential uses of literature. An audience may prefer to seek a diversion not to sharpen their sense of the world, but to dull it with escapist entertainments. Artists may know well enough which of these ends they are serving, though the public's response can be unpredictable. Writers and readers who attempt the harder, more constructive task understandably misprize those who prefer the easier, less consequential one. The latter task tends to reduce the functionality of art by stripping the response process of any reflection on the principles of constituting worlds. Complexity is not integrated, but traded for simplicity. And the utopian imperative is undercut by suggesting that utopia is either already here or else too purely imaginary as to be worth attempting. Such uses make literature act as a regressive or conservative force substituting contemplation for action.⁴ [4 Compare Enzensberger (1977)].

     On the other side, literature can make life appear more meaningful or coherent than does everyday experience, especially in societies with sharply rising complexity and fragmentation. After experiencing one's world as an arbitrary heap of incidents and accidents, we might turn with relief to building supercoherent literary worlds. The turn might resemble escapism, an alibi for leaving the world as it is. However, the very fact that a literary world seems to reconcile the contradictions of our reality encourages us to formulate them and to see that they can, in principle, be reconciled. Of course, this response, like any other, cannot be compelled. But if we do not respond this way, the fault is not that of literature as a whole, but only of specific works and forces that shape response, or of our own fixity or helplessness as readers. Literature, as Ernst Bloch says, preserves the hope for utopia but doesn't undertake to deliver it. We alone can attempt to do that, and then only as a open-ended project with no final victory or happy ending. In an actual utopia, the intensified and free investment of the subject in constituting its object-world might, as Hegel surmised, overcome the alienation between subject and object enforced by the versions of reality in previous societies like our own.

    My brief outline in this chapter of a functional approach and some of the aspects of literature it might address will be filled in during the surveys of later on. I hope that until then, it might not be rejected out of hand, though several challenges are frequently voiced. One objection is that its does not fit people's common sense impressions of what they do with literature. Readers feel they are experiencing not their own understanding, but the world of the work, identifying with its characters and their goals, undergoing powerful emotions, experiencing sensations of pleasure and pain, beauty and repulsion, tranquillity and terror, and so forth. The higher-level framework I propose seems to abstract away from all those activities, which are undeniably factors of the total process, by viewing them all as constitutive realizations of significance. Still, higher-level descriptions of cognition and performance need not appear intuitively obvious to people who function contextually most of the time. Recursion irritates the practical mind; and self-reference has always been the stumbling block for logic, as logicians from the sophists all the way to Gödel and Hofstadter well know. Hence, whatever it means to think about thinking, to understand understanding, to write about writing, to read about reading, and so on, will probably not be found to fit commonsensical intuitions.

    At this stage, high-level abstractions deserve a fair chance, now that descriptions of literature have so long tended in the opposite direction by hypostasizing specific projects and demands: that literature be “beautiful” and “harmonious," "instructing and delighting" us, "purging the passions," “illustrating the ideal in nature," and so on. In the long run, such hypostases have -proven unhelpfully restrictive, the more so if construed as binding values for authors and critics.  We might now work from the other end by seeking a sufficiently high-level framework to encompass the central aspects of literature, and then fitting available projects into appropriate sectors. For this purpose, my outline might, on the contrary, be attacked for not being abstract enough, for downplaying its own historical and ideological context and thereby hypostasizing an aesthetics peculiar to classical or modernise art in the West. That art is admittedly my primary contact with art in general and supplies most of the illustrations for our sample critics; in Asian art, for example, the philosophical and theological functions were apparently less clearly distinguished from literary ones.⁵ [5 On Chinese theories of literature, see Liu 1973.] Yet here again I would uphold the thesis that seen from a sufficiently high level, some mental processes are fundamental for humans across all cultures. Which level we would need is of course still to be decided.

    In any case, my modest outline is offered not to define literature itself, but to provide a heuristic for engaging contemporary literary theorists. Headway on such a project seems worthwhile enough that judgment of the outline itself should reflect its usefulness in that context. Moving up and down among levels of abstraction is surely a necessary dialectic for pursuing the understanding of understanding. I would simply assert that the nature of literature, however it be defined, renders critical discourse a useful domain for that pursuit.

Click here to go to Chapter Three

Click here to go to Critical Discourse Main Page