Introduction: From Imitation to Intertextuality

This paper offers a perspective on a range of contemporary developments and articulations of the phenomenon of intertextuality in fiction and film. Using as backdrop a brief discussion of different intertextual motifs in Salman Rushdie‘s Haroun and the Sea of Stories (1990), Paul Auster‘s Travels in the Scriptorium (2006) and Pixar‘s animated short film Boundin‟ (2004), it moves on to discuss the highly intertextual relation between the works of Swiss writer Robert Walser and the contemporary American experimentalist Alison Bundy. The paper thus problematizes and qualifies the line of demarcation supposedly existing between texts or works of art and aims to expand and exemplify the scope of reference, citation and paraphrase inherent in the overall concept of intertextuality. This paper springs from a baffled encounter with four postmodern works which all revolve around the theme of ambivalent originality. The four works portray origin and the original as regurgitation and as a result or an end point. And they describe the site of origin as both primary and secondary, and as disturbingly identical to what appears to repeat it. This undermining of the stability and integrity of the point of origin is presented as a thoroughly relational event. Origin is seen to lose its originality in the interaction with its surroundings, echoing Graham Allen speaking of the relationality, interconnectedness and interdependence in modern cultural life‘ (5). In this paper I discuss four different expressions of this interconnectedness‘—expressions which each in their own way portray or testify to patterns of intertextuality. For the nodding, copying, alluding and parroting discussed below are all more or less explicit manifestations of the poststructuralist tenet of the inevitable intertextual dimension of language and text. Several of the key voices of critical thought of the last forty years—all of them representatives of J. Hillis Miller‘s so-called uncanny‘ or Dionysian‘ critics—have addressed this aspect of the deceptively margined and coherent unity of the entity of the Book. As J. Hillis Miller puts it, [a] literary text is not a thing in itself, organically


Introduction: From Imitation to Intertextuality
Neil Forsyth, University of Lausanne The term "Intertextuality" has been used in so many different ways since it was first introduced in the late sixties that it no longer retains any specific meaning, at least without further definition.How, then, are we to account for all the ways in which we move beyond what the authors of various texts (Milton, Blake, Joyce are some pertinent examples) may consciously intend in the way of allusion?What, indeed, has happened to the notion of the multiple discursive contexts of a "text"?
Most of the essays in this book began life as contributions to a seminar at the ESSE conference in Aarhus in August 2008, the goal of which was to review the ways in which the meanings of the original idea have migrated and proliferated.Julia Kristeva (1967) invented the term, deriving it from her knowledge of Mikhail Bakhtin, and in particular his concept of what came to be called, in a classic translation of his work, "the dialogic imagination".The fundamental concept of intertextuality is that no text, much as it might like to appear so, is original; rather it is, because of the nature of language itself, a tissue of inevitable, and to an extent unwitting, references to and quotations from other texts (Allen 2005: 1).In a more recent formulation, however, and in spite of Kristeva"s angry insistence that the word was already being used in 1974 in the banal sense of source-criticism, the respected critic Gérard Genette has returned to "a relation of co-presence between two or several texts", in particular of the "effective presence of one text in another" (8).Where does this leave us?
In the Early Modern period, this latter kind of "intertextuality" was known as "imitation".Originality of the kind that came to be prized by the Romantics, and against which Modernism reacted, as in Eliot"s concepts of "Tradition" or of the "impersonality" of the artist, was not highly valued.Instead, what you learned in school, and what you went on to practice, was the art of imitation-with variation.Whether the models, or what Genette calls the hypotext, were classical-Homer, Theocritus, Virgil, Ovid-or whether they were more recent-Dante, Bocaccio, Petrarch, Tasso, the goal was to work within one of the genres established by those great originals, and go one better if you could.One curious instance of this method is that Milton"s famous claim to be composing "Things unattempted yet in prose or rhyme" (Paradise Lost I.16) is actually a translation of Ariosto"s line in the Orlando Furioso I.2, "Cosa non detta in prosa mai, ni in rima".Both in turn go back to Horace Odes III I 1-2.This is not plagiarism, nor even really allusion: it is simply quotation (to use the three terms that Genette identifies as examples of intertextuality).Milton is not hiding behind a theft, or even parading for a knowledgeable readership an adaptation of his source: he would simply expect his readers to enjoy recognizing the quotation, and perhaps even to feel a slight irony that a claim to originality can be such old hat.He is also announcing the world within which he is placing himself and by which he expects to be read and judged.But conscious quotation of this kind excludes other unconscious similarities, such as the Bastard"s "unattempted yet" in Shakespeare"s King John.That resemblance is certainly "happenstance", as Eric Griffith pointed out in a recent, sceptical essay. 1 It might come under someone"s definition of "intertextuality", but it would no doubt have surprised Milton if anyone had pointed it out-and it adds nothing to our appreciation of the relation between Milton and Ariosto.
That kind of quotation is quite different from what we find Eliot doing at, say, the famous conclusion of The Waste Land where almost every line is a quotation, but from as disparate a set of sources as one could imagine, from popular song or nursery rhyme ("London bridge is falling down") to Gérard de Nerval to Dante to Kyd to the Upanishads.In each of those cases Eliot famously added a footnote (what Genette calls the "paratext") to announce what he was doing.Perhaps the display of learning is similar to Milton"s, and that may, oddly enough, help to account for Eliot"s need to attack Milton so thoroughly.But the quotations are just what he calls them: "these fragments I have shored 1 The reference is to King John, II.i.601,where "the Bastard torrentially reflects that the only reason he is railing against bribery is that nobody has so far troubled to try greasing his palm.We discount this as "static", interference from a shared, but insignificantly shared, atmosphere, unless we impute to Milton a desire to hint with inordinate faintness that we should think of him as a bastard, too" (Griffith).
against my ruin" (l.430).Their very disparity makes the point about what is left to modernism after the war-a wasteland of discarded bits and pieces, jostling each other without any obvious links or coherence.We, the readers, supply the coherence, and all will do so differently, in spite of those notorious footnotes.
Perhaps other forms of textual relationship are also at work in Eliot: appropriations, plagiarisms, parody, pastiche, homage, citation.In fact that list introduces virtually the whole panoply of rewriting strategies.In the era of post-modernism, many texts are rewritings, and are seen to be so, deliberately.Tom Stoppard is sometimes said to have launched this literary practice most thoroughly in 1966 with Rosenkrantz and Guildenstern Are Dead, the same year in which Jean Rhys published The Wide Sargasso Sea, but since then many have stepped in.Angela Carter is a modern favourite for her rewritings of fairy-tales.She spoke of putting "new wine in old bottles, especially if the pressure of the new wine makes the bottles explode" (69).
The similarities among these various ways of describing the relations among texts should not blind us to the differences.They are historically and philosophically distinct.The idea of an explosion of the old bottle shows up the peculiarity of the postmodern concept of rewriting: it is aggressive towards the source-text, but also depends upon it.Eliot, by contrast, spoke in "Tradition and the Individual Talent" of the "existing order" of the monuments of tradition being "ever so slightly altered" by the arrival of the new (5).Even that notion would be alien to the Renaissance doctrine of Imitation (Greene), which was simply a description of what always happens, necessarily, in the writing of good poetry, rather than an ideal to which the writer might aspire.
The Renaissance concept is not to be identified with that now debased term "imitation".In his most outrageous statement about this device, Eliot made up this famous and revealing aphorism (in the Massinger essay, 182): "Immature poets imitate, mature poets steal; bad poets deface what they take, and good poets make it into something better, or at least something different".Eliot was reacting against the Romantic notion of originality, or more exactly, that poetry expresses the author"s personality.He was trying to replace it with a world in which poets work within what he calls "Tradition", in fact within that Renaissance world in which Imitation means not only what immature poets do, but what all do, always.And in which they take pride.But what he succeeded in doing was leaving modernist poetics beached high and dry, broken driftwood that could never be refloated in the sea.
Curiously enough, what Eliot wanted, when he spoke of the simultaneous existence of the whole of western literature, was, one can now see, rather like what that French theory of the late 1960s spelled out: "Any text is an intertext; other texts are present in it, at varying levels, in more or less recognizable forms: the texts of the previous and surrounding culture", as Barthes put it ("Theory" 39) 2 .The text is not an object but a field of activity, or an occasion for it.A challenge to the reader.
In spite of this apparent similarity between what Eliot aspired to and the concept of intertextuality that Kristeva and Barthes invented, the terminology, and indeed the theoretical underpinning, of the two notions are quite different.For one thing, Eliot, in spite of his reaction against the idea of "personality", still thought in terms of authors, and great ones at that (Virgil, Dante), when he described the "Tradition" that could be only "ever so slightly altered".Kristeva and Barthes, however, invented signifiance, a French neologism which proposes that "texts" (another word for whose ubiquity their theory is largely responsible) are potentially infinite in their meaning since readers activate the intertextual meaning of what they read, and each one will be different.The reader, not the author, is the source of meaning (1968).The author"s role in traditional literary criticism has been "to resolve discontinuities of discourse into a harmonious whole" (Young 12) but he was now reduced by Barthes to a mere "scriptor".With the shift to "textuality", and in the wake of Derrida, those discontinuities become the focus of interest.
Graham Allen insists on the distinction: intertextuality, at least as Kristeva and Barthes were using the concept, is not to be confused with influence, allusion and all the other intentional ways in which one writer refers to or quotes from another.Influence remains within a vision of 2 Barthes, "Theory of the Text", was actually an entry in the Encyclopaedia universalis, entitled "Texte (théorie du)".It thus made the term official.Barthes writes that "tout texte est un intertexte; d"autres textes sont présents en lui à des niveaux variables, sous des formes plus ou moins reconnaissables: les textes de la culture antérieure et ceux de la culture environnante; tout texte est un tissu nouveau de citations révolues."literary works that believes meaning to stem from the intention of an author.Intertextuality involves a recognition that meaning lies between texts in networks which are ultimately only partially recoverable, only partially readable (or traceable).
Barthes"s textual analysis was particularly influenced by Julia Kristeva"s work on the notion of text and intertextuality and by Jacques Derrida"s deconstructive account of the sign.Kristeva had recently come to Paris to work with Barthes, and she brought with her a sophisticated understanding of Mikhail Bakhtin, until then virtually unknown in France.Bakhtin (1989Bakhtin ( [1929] ] 131) had argued that No member of a verbal community can ever find words in the language that are neutral, exempt from the aspirations and evaluations of the other, uninhabited by the other"s voice.On the contrary, he receives the word by the other"s voice and it remains filled with that voice.He intervenes in his own context from another context, already penetrated by the other"s intentions.His own intention finds a word already lived in.
The impact of these collective influences led Barthes to develop an approach to the reading of narrative texts that marked the decisive step in the shift from structuralism to post-structuralism.Instead of seeking to relate texts to a structuralist notion of the abstract system of narrative, he now developed a method that foregrounds the involvement of texts in the vast intertextual arena of cultural codes and meanings out of which they are woven.Textual analysis, based on this intertextual notion of meaning, replaces the apparently scientific and objective approach of structuralism with an emphasis on the openness of the text (its meaning can never be fully captured or resolved) and the productive role of the reader of the text (each individual reader brings with them a specific and distinct if in no way unique relation to the "cultural text").In "Theory of the Text", Barthes argued that a text has meaning only when a reader activates the potential meanings intertextually "present" within it.A text, viewed intertextually, only exists in the act of reading.
Since then other theorists have developed the concept, and the strict definition proposed by Graham Allen has not been adhered to, as he ruefully acknowledges.In particular, Gérard Genette has reined in the potential anarchy of Barthes"s approach to reading by breaking up the original idea into sub-categories.He proposed the term "transtextuality" as a more inclusive term than "intertextuality", and listed five subtypes: 1 intertextuality: quotation, plagiarism, allusion (as in the Milton instance above); 2 paratextuality: the relation between a text and its "paratext"-that which surrounds the main body of the text-such as titles, headings, prefaces, epigraphs, dedications, acknowledgements, footnotes, illustrations, dust jackets, etc; 3 architextuality: designation of a text as belonging to a genre or genres; 4 metatextuality: explicit or implicit critical commentary of one text on another text (metatextuality can be hard to distinguish from the following category, but Nabokov"s Pale Fire would be an obvious and parodic instance); 5 hypotextuality (Genette"s term was hypertextuality): the relation between a text and a preceding "hypotext" -a text on which it is based but which it transforms, modifies, elaborates or extends (including parody, spoof, sequel, translation).Genette uses Virgil"s, and especially Joyce"s, relation to Homer as a standard instance, but develops many elaborate sub-categories, of which Joyce himself uses quite a few.
To that list, computer-based hypertextuality (6) should be added: i.e., text which can take the reader directly to other texts (regardless of authorship or location).This kind of intertextuality disrupts the conventional "linearity" of texts.I would add (7) intratextuality for the kind of repetition or echo within a text which also breaks up the linear reading and requires the reader to fold the text in a different way.For a long, complex text like Paradise Lost or Ulysses, it is indispensable to describe one of the ways in which the poem or prose text means.
Genette"s books will seem to some the production of an obsession for categorisation gone mad.Yet, like F.K. Stanzel"s Theory of Fiction for example, the classifications allow us to perceive similarities among works that might otherwise seem very distant from each other.And above all it may help us as we try to rethink the various concepts of intertextuality that have proved fruitful in the writing of the essays and conference papers collected here.In Barthes and Kristeva, intertextuality is a feature of all literature when opened to the reader"s imagination.But Genette teaches us to distinguish kinds, and thus to get closer to the particular literary relations we want to study.And unlike Kristeva he explicitly does not exclude the conscious imitation that had been so integral a part of how Early Modern literature demanded to be read.