Mapping The Gold Bug Variations:The Taxonomy of Intertextuality, the Partial Nature of Metaphoric Structuring, and Postmodernism

William Gillespie

Talking about music is like dancing about architecture.

- Laurie Anderson

The map is not the place. But you need the map to get there.

- Richard Powers

Intertextuality

Postmodern ideas of intertextuality situate all texts within a network of references to, quotations from, and plagiarisms of each other. Every text is an intersection of others. It cannot escape. Language can never refer to anything outside itself. It is not possible for an author to write something new. It is all written. Every word is a reference. Every style is an appropriation. John Barth calls this the "Literature of Exhaustion," which David Cowart describes as "the dread that the precursors have told all the stories, exhausted the genres, `used up' the very language available for artistic creation." In Cowart's description, when writers, to put it politely, "are influenced," or, in the words of Zora Neale Hurston, "borrow," or, in the words of Oscar Wilde, "annex," or, in the words of T.S. Eliot, "steal," or, in the words of Kathy Acker, "plagiarize," or use, as John Gardener puts it, "the Echoic method" (what Joyce calls "stolentelling"), it is a vital and customary aspect of literature. If it is indeed possible to create something new, that is not what great writers do. (1-2)

Types of Intertextuality

There is evidence to support the inevitability of intertextuality: texts refer to other texts in many different ways for many different reasons. Conversely, there is no way to identify something new. It is convenient to assume the death of the author and avoid the possibility of something new or reasons for intertextuality, inevitable or not. But if we keep the author around for awhile, we can ask her if she was attempting to create something identifiably new through intertextuality. Is there old intertextuality and new intertextuality? Can we deconstruct intertextuality that way? How does one distinguish between an author who plagiarizes, an author who quotes, an author who unwittingly quips Ben Franklin or Shakespeare or the Bible, or an author who deliberately mixes sources that would otherwise remain distinct? How does one distinguish between a novel in which the characters listen to a particular piece of music and a novel which is structured according to a piece of music? These are all incidents of intertextuality. We need a taxonomy of intertextuality. Our classification system should be relevant to the author's intentions, as long as she's here. We need to reduce a complexity of intertextual links to something conceptually neat.

The Symbiotic Spectrum

One such lucid taxonomy is David Cowart's Symbiotic Spectrum. It is more a means to classify an entire book than an individual reference. The Symbiotic Spectrum can be read as a scale of precision of representation from nothing represented (right) to representation as close to exact (one-to-one correspondence) as possible (left). Off the right edge of the spectrum is the original or self-begotten text, which, for Cowart, is a dubious notion. Therefore the right end of the symbiotic spectrum is not a point but an ellipsis trailing off into the limit as text approaches originality. Here are the other points on the spectrum advancing from the right to the left:

Ordinary intertextuality

Allusive texts

Texts incorporating extensive parts of other texts

Symbiotic texts

Translations from one genre or medium to another

Translations from one language to another (6)

Because "translation, like signification itself, is never wholly transparent" (meaning that it does not map perfectly -- it is not a perfect one-to-one correspondence), the left end of the spectrum is "a category almost as mythical as the self-begotten text: the `simple' linguistic translation from one language to another." (7) Is this also an ellipsis trailing off into the limit as text approaches an exact replica of itself, or is there an endpoint there? Perhaps that endpoint is cryptography, an exact (one-to-one) translation into code. Does the endpoint have to be a translation between two languages? Otherwise, perhaps the endpoint is plagiarism, an exact translation into itself. Perhaps that endpoint can be found in Jorge Luis Borges' story "Pierre Menard, Author of Don Quixote" where the ninth and thirty-eighth chapters of Part One, and a fragment of the twenty-second chapter of Cervantes' Don Quixote are written again word for word. The intent of the noble Pierre Menard, who does not consult the original, is not to steal or borrow, not to write a contemporary version of Don Quixote, and not to place the character of Don Quixote in a twentieth-century setting:

Like any man of good taste, Menard detested these useless carnivals, only suitable--he used to say--for evoking plebeian delight in anachronism, or (what is worse) charming us with the primary idea that all epochs are the same, or that they are different. (48)

The story is written in the style of a critical essay, and the narrator compares passages of Cervantes' writing with identical passages by Menard, and concludes that "the second is almost infinitely richer." (53)

Powers' Gold Bug Variations and Bach's Goldberg Variations

Intertextual references are dense in even the title of Richard Powers' The Gold Bug Variations. Powers' novel is an intertext: a well-researched and informative book, a novel in need of an index. Of the artists and scientists referred to, Poe and Bach are merely two that are easy to identify. Of the texts referred to, Bach's Goldberg Variations and the painting Landscape with Fire by Flemish painter Herri met de Bles are two that are not texts. The references are various, and three types emerge: story, metaphorical, and structural. The story references happen as events in the plot: the characters of the novel read, listen to, and discuss Poe and Bach. They study biology. The metaphorical references occur when aspects of Poe, Bach, Shakespeare, biology, cryptography, computer programming, information theory, and history, are used as metaphors for each other or for events in the story. The third type of reference is structural. Unlike the first two types, it is not exactly semantic. What makes the intertextual connections to Bach interesting is that the Goldberg Variations are not text. Without musical notation (which the novel does indeed use), the variations cannot be quoted -- only referred to, or translated. This means that if there is a left endpoint to the Symbiotic Spectrum, The Gold Bug Variations' relationship to the Goldberg Variations can never get there. It falls in the (vast) realm of translation between media.

Translating Music into Prose is Complex

To translate music into prose is not simply to echo the music. It is a problem. It can be solved many ways. A one-to-one mapping is almost out of the question. It is theoretically possible to describe every marking on the score such that the score can be reconstructed perfectly (either in longhand English or code that would read as nonsense in English), but aside from such an attempt, the organization of staff paper cannot be preserved in the nebulous, subjective, disorganized infinity of semantics and all possible ways a sentence can mean. The writer can report her impressions upon a single listening of a piece of music -- many do -- but this is not a translation of, and often barely even a description of, the music. The translator must decide which parameters of music are relevant, which parameters of the text are relevant, and map the two. Parameters of the music can include aspects as diverse as the rhythms, the pitches, and all their smaller and larger-scale structures. Parameters of the text can be acoustic, typographical, or semantic. It is this process, and not the possibility of a one-to-one correspondence, that would make such an attempt worthwhile. To translate music into writing is not to recycle, copy, or mimic the original music. It is to compose something new, in which the original structure is transformed into something identifiable but no longer recognizable. As Cowart puts it: "Translation involves considerable creativity, and the work of the translator is often copyrightable." (7) The structures Bach arrived at through his contortions are not possible, much less plausible, in English. To write them is to construct them. The reference is neither theft nor ingenuity-saving device. Translation across media is one way intertextuality might result in text which is identifiably new.

Structural Reference in The Gold Bug Variations

Powers translates certain aspects of the structure of the Goldberg Variations into certain aspects of the structure of The Gold Bug Variations. There are 30 numbered chapters for the thirty variations, and chapters entitled Aria at the beginning and end of the novel. As the first 29 Goldberg Variations cycle through three types of variation (Dance, Arabesque, and Canon) almost ten times, there are three narrative voices the novel switches between. Powers attempts to translate diminution and augmentation -- formal devices in music in which a voice in a canon harmonizes with a sped-up or slowed-down version of itself. While most chapters contain a section for each of the three narrative voices, four chapters near the center of the book, XIII, XIV, XV, XVI, are each devoted to one of the three, demonstrating a deceleration in tempo. There is a chapter (XXII) which switches narrative voice midsection demonstrating an acceleration in tempo. Jay Labinger describes this part of the book as a "stretto, a fugal device where the musical voices follow one another at shorter temporal intervals than the previously established pattern." (87-88) The structure of the simultaneous narratives refers to formal aspects of the canon. In the Goldberg canons, a melody is harmonized with an exact replica of itself which enters a measure later, usually transposed up or down the scale, and (twice) in inversion. Two of the three narrative threads have a similar shape, or sequence of events. As the novel switches between them (as a novel usually does not present two simultaneous texts) the events in one of the stories closely follow the events in the other. The recognizable dramaturgy of the love story make this counterpoint easy to follow. Labinger, in suggesting a means for mapping a canon into fiction, proposes a solution for overcoming the problem of how to have simultaneous voices in text -- a medium restricted to a single voice at a time: "one could imagine a quite literal canonic form with the three lines printed one underneath another as in a musical score; but that would merely be annoying: readers are not equipped to take in several lines of text simultaneously, as musical listeners can." Julio Cortazar's Hopscotch has a chapter which demonstrates almost exactly this, and it is not annoying (191). Other stories by Samuel Delany, Chriz Mazza, and others find a means to present simultaneous texts. To solve a difficult problem, however, Labinger is casually suggesting an innovative form of writing in order to represent a traditional (centuries old) form of music. Labinger does not address what seems to be the next question. Both musical notation and text advance through time from the left of the line to the right. However, in musical writing, the vertical axis is pitch, and both the shape and positioning of the individual voices reflect this. In fiction, the only vertical axis that is conventionally understood is in Freitag's plot triangle, and even then the conventional understanding is not nearly as precise as in music. The dramatic tension in a story cannot be measured or perfectly understood whereas the frequency of a note can be measured and explained by a musician or physicist. If we take up Labinger's suggestion (and I intend to someday), do we assume that the voice lower on the page is a bass clef: subdued dramatic intensity in counterpoint to the more shrill and urgent text running parallel above it? To describe what Powers does here as "augmentation," "diminution," or "stretto," reveals some aspects of these chapters, is misleading about others, and conceals virtually everything about them. The Goldberg Variations' structures are not recognizable in the novel -- if one were to compose a piece of music for solo keyboard based on the novel one would probably not write the Goldberg Variations. The odds of it are as small as the odds of Powers arriving at this structure any other way. The structure of the novel is not the same as the structure it is a map of -- it is new.

Metaphor

In Metaphors We Live By, George Lakoff and Mark Johnson argue that a metaphor is not a literary device, it is a fundamental mechanism of human cognition (3). The metaphors in everyday speech reveal the way we think. There are metaphors in virtually every utterance. Many metaphors are invoked through a single word (invoked: metaphor as evil spirit). In the book More Than Cool Reason, George Lakoff and Mark Turner introduce the terms "source domain," "target domain," and "mapping" to describe the process of metaphor (63). In a metaphor, the "target domain" is structured by the "source domain." This structuring is called "mapping." The source domain is typically something more clearly understood, often through physical experience. The target domain is typically more abstract, thus the necessity to structure it in terms of something more clearly understood. The mapping is a correspondence between features of the source and target domain which can never be perfect. Each attempt to structure a metaphor will reveal aspects of the target domain while concealing others. A metaphor is a translation, a correspondence which is not one-to-one. An awareness of metaphor, and the ability to choose new ones, can allow us to construct, reveal, amplify, prod, taste, and get a whiff of different aspects of a target domain.

Mapping

This, the partial nature of metaphoric structuring, is similar to the partial nature of translating music into text. Similarly, the failure of different languages to map perfectly prevents Cowart's "transparent" translation. Similar imperfections may be found in the mapping of language onto reality. Whether or not language can even refer outside itself, it does not have a one-to-one correspondence with reality. There is not a different word for every pebble, a different species for every sparrow, or a different Latin name for every dandelion. Language is reductive. Even science may be only a translucent translation of the phenomena it describes. A map cannot show every aspect of a place:

Metaphors for Text

What is text, anyway, and how does one discuss it? Lakoff and Johnson explore two common metaphors: ideas as food (108), and arguments as buildings (98). O'Donnell and Davis describe the text as a woven fabric (x). Ideas as food is a metaphor used more often to describe the consumption than the production of ideas. For example, I often have trouble digesting postmodernism. To stretch the metaphor to include idea production (author as cook), is intertextuality cooking a meal entirely from leftovers, or are all texts evidence of a random play of ingredients? Ideas as buildings is a metaphor used to describe how structurally sound an idea is: whether it can stand on its foundation without buttressing or support. Text as woven fabric is a metaphor constructed specifically to reveal the intertextual nature of text. In this metaphor, however, every weaving is a weaving of other weavings: there is no cloth or yarn, no text that is original, no ingredients.

Metaphors for Intertextuality

What does one call a single intertextual connection? How does one refer to a reference? Metaphors used to describe intertextuality include: the reference as intertextual echo and the author as locus of intertextuality. Is the music of Bach "echoed" in the Gold Bug Variations? Given the impossibility of reproducing music within text, the metaphor of the echo has a misleading literal interpretation. Music is a sound, and can actually echo. Is echo the best metaphor? In the Third College Edition of Webster's New World Dictionary, the first two definitions of the word "locus" are

While the metaphor is constructed with the first of these definitions -- an author is not a person but a place where texts intersect -- the metaphor implied by the second definition is useful. We will map it momentarily.

Where Lakoff ends and Powers Begins

The book Metaphors we Live By is a convincing argument that everyday metaphors are coherent, culturally bound cognitive structures that shape our thought and speech. While encouraging an awareness of how our thought and language are structured by metaphor, the book only hints at the possibility of consciously creating our own metaphors to live by. It is no wonder the book stops there. The implied next step is a leap off the edge of understanding. Everyday metaphors systematically order and clarify our world. We are so familiar with their mapping we seldom notice what has been omitted. New metaphors challenge our basic understanding of things. They confuse before they clarify. Their failure to map perfectly is noticeable and disorienting. They are not the maps we are used to using. As Lakoff and Johnson illustrate, a common metaphor for time (in our culture) is time is money (7). The phrase "time is money" is familiar to us and we live by it, whether in earnest or in irony. A vague understanding of time is clarified by a more immediate understanding of money. Time can be saved, invested, bought, spent, or wasted. This is a way to structure an understanding of time that reveals some aspects of time, is misleading about others, and conceals the rest. The metaphor reveals that an individual can run out of time. It structures a false understanding of time as personal property. Finally, it conceals aspects of time: you can't save up time. The metaphor conceals aspects of its source domain as well. It doesn't acknowledge that one can run out of money and continue living, or die rich. Here is an example of a new metaphor for time: time is a dog. This metaphor seems odd, given the dynamic (river) and mathematical (money) models we normally use to describe time. Time is not a furry quadruped with sharp teeth. We can, however, easily imagine taking the time for a walk. Time will eventually destroy your shoes. We can feed our time, and take care of it, so that we may grow old together. Time is a dog is an example of a new metaphor which is largely unmapped. We don't know what we can say about time with it. An effort to find out, though, may allow us to talk and think about time in a way were previously unable to. Where Lakoff ends and Powers begins are the new metaphors mapped in the Gold Bug Variations.

Metaphor Reference and the Goldbug Variations

In the Gold Bug Variations, descriptions of the DNA molecule are intertwined with descriptions of the Goldberg Variations. The language of genetics and the language of music mix like ammonia and methane in the lightning of Powers' metaphorical acumen. The use of new metaphors to better understand different aspects of things is critical in the story. Stuart Ressler is initially out to crack the genetic code. With story references to Poe's "The Gold Bug," cryptography is established as a metaphor for the genetic code. This metaphor is misleading: it implies that a simple key will reveal a perfect meaning to the four-molecule sequence and all the complexities of life will be spelled out, letter for letter. Ressler is transfixed when he first hears the Goldberg Variations, which cannot be translated into English through any key, but can be inspected successfully for patterns on larger and smaller scales. He begins to structure his understanding of the DNA molecule by his understanding of the Goldberg Variations. The new map is a good one. The Goldberg Variations share many numerical parameters with the DNA molecule: "Bach had a habit of imbedding mystic numbers in his compositions; these ones happen to correspond to the ones nature imbeds in its own. But this coincidence was the least of the qualities that made this music Ressler's best metaphor for the living gene." (579) It takes three nucleotide Bases to code for one of twenty amino acids, and the variations occur in groups of three. Furthermore, "Superimposed over those first four triplet rungs, a diversionary tune that, with grace notes, contains twenty tones. Two halves of the aria, each sixteen, bars, both scored to repeat, totalling sixty-four measures." (192) There are sixty-four possible combinations of the three nucleotide bases. Lastly, and most importantly, both the Variations and the DNA molecule can be reduced to four elements -- four scale steps descending from Do, or four nucleotides -- from which all their complexities come. The DNA molecule is used as a source domain for the Variations as well. A Canon is described as if "two copies twist around one another with helical precision." (580) A drawing of a double-helix looks like a drawing of two identical curves slightly out of sync, as if a melody following itself a measure later. There are more than enough superficial similarities to justify the mapping. But what does the new map reveal that the old one does not? For one thing, while the DNA molecule and the variations can both be reduced to their simplest terms -- four bases, four notes -- in both cases the simplicity of the code is only interesting because of the complexity and variety of the possible messages. "I listened to these miniatures for a year, pulled out of them the most marvelous genetic analogies. But at the end, the music refused to reduce, and it hurt worse than before." (193)

Intertextual Variations

The Gold Bug Variations is based on, and structured by, Bach's Goldberg Variations. Another connection is drawn between DNA and the Goldberg Variations through a pun on the word "Base." (191) Based on, according to the novel, is another way of saying structured by. Can this be a new metaphor for intertextuality? Cervantes' Don Quixote is the subject and object of writing by Borges, Barthes, Foucault, Shklovsky, Auster, Acker, Cliff's notes, and music by Strauss. I have even discovered that Don Quixote is frequently referred to in 20 consonant poetry (an attempt to map 12-tone musical structure onto the alphabet) because words with both Q and X are rare and force the probability of the English alphabet. Don Quixote is a high-scoring reference in Scrabble. To say that these secondary texts are based on Don Quixote, to use the metaphor from The Gold Bug Variations, firstly implies that they are variations on the theme of Don Quixote. The purpose of a theme and variation structure in music is less to repeat the theme than to explore its possible permutations. It is the differences that can be wrung from a mostly identical melody that are ultimately the subject, not the melody itself. Julia Kristeva ('s translator), citing Bakhtin, uses the term "voices" to refer to the texts which intersect in a literary work. In music, "voice" refers to a melody when it is heard simultaneously with other voices in counterpoint. (O'Donnell, 281) Bach takes a simple melody and, by repeating it in simultaneity with other melodies, changes the way the original melody is heard. This is a fine metaphor for intertextuality: a harmony of different voices. The literal definition of "voice," when we speak of texts as voices, implies a speaker, or author. Bach's Goldberg Variations are merely one of many voices in counterpoint in The Gold Bug Variations.

Intertextual Offspring

The metaphor of the Base secondly implies that The Gold Bug Variations is a genetic copy of the Goldberg Variations. Earlier we examined the metaphor of the locus of intertextuality. There is an immediate danger in using genetics as a metaphor for intertextuality in an intertextual realm dominated by men. Unless we are to risk discussing Powers as someone who was ultimately sired by Bach -- or one of Bach's sons -- with the implied assistance of females whose genotypes and phenotypes are irrelevant to history, we need to remove human beings from our source domain altogether. Are the Goldberg Variations a locus -- a trait which The Gold Bug Variations has, like blue eyes, claws, fur, a beak, or fins? Are music and fiction different species? Do we preserve the historical sequence of Bach, then Powers, or assume that they coexist on the same or on different continents? Is the 32-variation structure a gene they both inherited from an ahistorical intertextual gene pool? The metaphor can be constructed to answer some of these questions. The metaphor reveals that intertextuality is an attempt to encircle an infinity of variations. As every dandelion is unique, so is Pierre Menard's Don Quixote different from Cervantes'. The metaphor can also be constructed to reveal the inevitability of intertextuality. Text can be thought of as a biosphere, which encirles a potentially infinite play of amino acids. While there is room for infinite diversity, no form of life can exist outside of earth. Text is continually evolving and recombining its elements. Every book is unique, even a transparent translation between languages, or Pierre Menard's Don Quixote (an invisible translation between cultures). A rattlesnake will function differently in the desert than it will in the bayou. Intertextuality as an ecosystem conceals any possibility of an autonomous text while it conceals the significance of an intertext. Within this metaphor, the autonomous text is impossible, and the intertext mundane. A text cannot exist without its influences, but neither can its influences be perfectly mapped onto it. The text's mutations, the surfacing of its recessive traits, and its adaptation to new circumstances, are more important than its persistence. It is the interdependence between wildly varying texts, rather than the genealogy of any one of them, that is significant.

Reductionism/Holism

The Goldberg Variations are described as variations on a simple four-note theme. All of life on earth is described as variations on a four-Base theme. The two are described in terms of one another. Due to the partial nature of metaphorical structuring, every map is an oversimplification. Oversimplifications are necessary to understand complicated things, but there is always the risk that the map will become a substitute for the landscape it represents. There are detectable patterns in the Goldberg Variations, and it can be oversimplified surprisingly effectively. Yet, "To try to locate, in that thematic germ, what Ressler spent a life listening to would be to search in those schematics -- line drawings showing every subassembly of every carburetor part -- for a semblance of the functioning car." (585) The later Stuart Ressler looks back on his reductionist mindframe as insanity: "As you can imagine, I fast approached the conviction that either everything in the universe fit into a regular pattern or I was, at my tender age, perilously close to a weekend at what people in the late fifties were fond of calling the Funny Farm." (192) The novel is not about the failure of the map's oversimplification nor the victory of the landscape's complexity. It is about the relationship between the two. It is about the necessity of metaphor.

Richard Powers

I visited Richard Powers the first time hoping for the key. I wanted him to roll out a genetic blueprint of the novel. I had tried to read Bach's scores in hope of finding the Base for the novel. This line of inquiry faltered, and momentarily I felt as if the younger Ressler had visited the older Ressler to find the code, and received instead a frown and a sad shake of the head. At first I was afraid that the structural reference I was after was not there. It is there, though. Strangely, the novel can be reduced in that fashion. Aspects of it can be found in Bach's score. As Labinger has demonstrated, the plot can be drawn as a simple graph. (88) It is not every novel that possesses such structural beauty that it can be analyzed reductively. Furthermore, the techniques Powers used are not easy to do. Ultimately, however, the novel is about the futility of such logic. Ressler is distracted from his quest to find the correct key by the beauty and complexity of the door. There is nothing in the simplicity of the canon, the number four, or the four scale steps descending from do to describe measure 20 of the canon at the fourth and why it makes me laugh.

Intertextuality and Amateur Theory

In writing, ones words, if not carefully regulated, quickly fall into familiar clusters and orders. In writing an academic paper for example, the recognizable formal aspects of theory exert a powerful gravity. The limits of intertextuality imprison one. My sentences quickly slip into the passive voice as I neutralize the subject "I" to simulate the objective voice of a disembodied omniscience. I use the verb "to be" frequently, and always in the present tense so there will be no doubt I am telling it like it is, always has been, and always will be. I am full of doubt, uncertainty, drastic last minute revisions, but the reader must not realize this.

Epiphany

I am afraid I am writing something that will never do any good to a reader. In constructing the paper to gain acceptance in a graduate writing program I am relying heavily on intertextuality to obscure my personal variations from the phenotypical college student. I inch along intertextuality's scaffolding. I serve leftovers with fresh garnish. I advance arguments by citing examples from fantastical Borges stories and never with my experience. I mummify myself in intertextuality's fabric. In a novel that celebrates technical aspects of love, loss, life, and complexity, it is convenient to order it along a single aspect. I can do that with this book. But the nagging gloom is that the book is still unread for me. Like a scientist or composer I constantly reduce as I investigate. I do not surrender to beauty or mystery. I refuse to be awestruck or foolish. With the Bach however, my eyes can no longer follow the rapid unfolding of score. I close them and listen and in my confusion am taken to tears and led back, shaken, soothed, and am left incoherent with no convenient description or memory. Novels and music are difficult for most of us to listen to. After two readings, multiple scannings, two interviews with the author, extensive perusal of primary and secondary sources, I am ready to forget everything I know and hear this novel for the first time.

Is it Postmodern or After?

Is The Gold Bug Variations postmodern? Despite its references, it seems to want to tell a story rather than draw attention to its own textual artifice. Is it intertextual? Yes. It is a novel packed with frequent and technical references, all deliberate, most academic. The novel manages to ignore or sidestep references to 50s or 80s commercial culture (with the exception of 50s rock). There is little trace of parody, but ample evidence of a sense of humor. The project explores the contradictions between art and science as well as the similarities. In short, the multiplicity of interpretations incorporated into the text are not contradictory or pastiche. None dominate, but they add up to something. Does it draw attention to its own construction? Yes. However, the novel neither flaunts its construction, nor appears to have grown organically. The characters are not flat, but a number of coincidences suggest they are not real (Stuart Ressler's birthdate corresponding to Glenn Gould's, as a tiny example). Is it about the failure of language? Yes. But the novel is also about language's necessity. The novel celebrates diversity and is an affectionate rebuttal to a more modernist desire to collapse the world into a tangible order. Richard Powers told me the novel is not there to resolve ancient dualities, like holism/reductionism (or post/modernism -- a recent ancient duality). He said "a map is not the place, but you need the map to get there." This is an admission that the map is necessary in all its insufficiency. The novel is balanced in this duality. It celebrates the system of map and place, source and target, language and reality. Postmodernism invites a multiplicity of interpretations with none dominant. Does postmodernism situate its own interpretation within this network? Does postmodernism come after modernism to supplant it, or to balance it? Is it coda or counterpoint? The Gold Bug Variations is not postmodern. It is after. We understand the limitations of understanding. Now what?

Conclusion?

While the existence of The Gold Bug Variations is hardly an argument against the inevitability of intertextuality, it demonstrates text's endless adaptability and ability to create new from the old. Through references to Bach in the story, metaphor, and structure, the novel becomes a mutation of a novel, and a counterpoint to the music of Bach.

WORKS WOVEN IN

Bach. Goldberg Variations. New York: CBS, 1956.

Baker, John F. "PW Interviews." Publishers Weekly 16 8 1991: 37-38.

Borges, Jorge Luis. Ficciones. New York: Grove Weidenfield, 1962.

Cortazar, Julio. Hopscotch. New York: Pantheon, 1966.

Cowart, David. Literary Symbiosis. Athens: The University of Georgia Press, 1993.

Hofstadter, Douglas R. Goedel, Escher, Bach. New York: Vintage, 1989.

Labinger, Jay A. "Encoding an Infinite Message." Configurations 1 (1995): 79-93.

Lakoff, George, and Mark Johnson. Metaphors We Live By. Chicago: University of Chicago Press, 1980.

Lakoff, George, and Mark Turner. More Than Cool Reason. Chicago: The University of Chicago Press, 1989.

Neufeldt, Victoria, Editor in Chief. Webster's New World Dictionary. New York: Simon and Schuster, 1988.

O'Donnell, Patrick, and Robert Con Davis. Intertextuality and Contemporary American Fiction. Baltimore: Johns Hopkins University Press, 1989.

Poe, Edgar Allen. The Gold Bug and Other Tales. New York: Dover Publications, Inc., 1991.

Powers, Richard. The Gold Bug Variations. New York: Harper Collins, 1991.

Stites, Janet. "Bordercrossings: a Conversation in Cyberspace." Omni Nov. 1993: 39-113.