I would like to thank members of Glasgow University’s Contemporary Fiction Cluster for feedback on a draft of this article: Maria Dick, Henry Ivry, Emma Lister, Helen Stoddart, and Sophie Vlacos.
- 1 I offer these dates—from the fall of the Berlin Wall to the end of George W. Bush’s second presiden (...)
1There’s a kind of mythology of the American writer as meteoric talent, whose blazing journey arcs above the horizon, casting a new and strange light on everything below, but who burns out spectacularly before any such geriatric terms as “lateness” might be applied. From Poe to Plath, Melville to Wallace, it has become, as Charles B. Harris notes, the “cliched fate” of many American writers (203). Yet, however fitting that mythology may appear, it is, with few exceptions, little help in understanding the various writers gathered under the elastic and capacious label American postmodernism. On a superficial level, what binds together postwar writers as different as Robert Coover, Percival Everett, Toni Morrison, Joyce Carol Oates and Philip Roth is their remarkable longevity and productivity. As such, US postmodern fiction in its late, millennial, phase—perhaps a twenty-year window, opening in 1989 and closing around 2009—1might seem to be the ideal testing ground for the several theories of late writing that emerge in this same period. Yet while art’s confrontation with “unassimilable death” (Barth, 2008 131) is a recurring theme in late postmodern fiction, I argue that reading American postmodernism through the lens of late style both brings the fuzzy borders of the term’s most famous claims into sharper focus, and helps us to see the distinctness of a certain set of postmodern fictions more clearly.
- 2 A distinction is necessary, here, between studies of late authors—which are the focus in this essay (...)
2Most current studies of late writing grow out of Edward Said’s posthumous cluster of essays, published as On Late Style (2006), which, themselves grow out of Theodor Adorno’s short essay, “Late Style in Beethoven.”2 While Said’s formulation of late style varies substantially across his book, and certainly amounts to less than a comprehensive theory, the most frequently invoked aspects of the concept are, by now, too familiar to require detailed elucidation: faced with the inevitability of death, a different accent appears in certain ageing artists’ work, breaking open the aesthetic wholeness of their earlier career. This new late style emerges in confusing final works, marked by deliberate contradictions, a refusal of synthesis, and a fragmentary character that disdains closure and resolution. Although there are significant earlier accounts of late style, Said’s distinctive formulation (and perhaps the eagerness of its critical reception) carries something specific to its cultural moment, a millennial germ that infects and increases the weight of any invocation of last things. Said’s book may be the most well-known, but it is significant that On Late Style is not, after all, the only study of late works from these years. One year after Said, Gordon McMullan published Shakespeare and the Idea of Late Writing (2007), critiquing the concept of late phase work as a “transhistorical, transcultural phenomenon” (7). The same year, Helen Vendler delivered a sequence of lectures (later published as Last Looks, Last Books in 2010), describing a late poetic “binocular style” that attempts to bridge the chasm between “the looming presence of death and the unabated vitality of spirit” (1).
- 3 Think of Coover’s “self-cancelling fictions” (McHale, 1987 20), “The Elevator” and “The Babysitter” (...)
3If there is a sense in which these various theories are prompted and haunted by the fin-de-siècle Zeitgeist, then their application to postmodernism may be more complex than it first appears. Postmodernism’s peculiar fascination with endings and belatedness has, after all, long been recognised: Fredric Jameson even begins his classic study by remarking that the advent of postmodernism is heralded by an inverted millennial phase in which “premonitions of the future” are replaced by announcements of “the end of this or that” (1). But given that so many of Said’s signatures of lateness—refusal of closure, “difficulty,” “unresolved contradiction” (7), disregard for “continuity” (9), construction out of “fissures and fragments” (46), attempts to dismantle “the very notion of identity itself” (82)—already seem to unwittingly describe the field identification marks of early postmodernism,3 it might be argued that not only is Said’s theory itself an example of postmodernism, but also that postmodernism is itself a theory of lateness.
- 4 A similar case might be made for Gaddis’s Agapē Agape, which, as Steven Moore has shown, is an expa (...)
4Yet if postmodernism and late style appear, rendered in the fading glow of their millennial twilight, as the overlapping circles of the same Venn diagram, it is the lateral zones of difference that are particularly revealing. Where Said, for instance, cites Henrik Ibsen as an exemplar of late style for the way When We Dead Awaken returns to origins to reopen “questions of meaning” and “tear apart the career and the artist’s craft” (7), I would suggest Robert Coover as a representative case of late postmodernism. Coover’s late works insistently return to early postmodernism—in Lucky Pierre (2002), for instance, Coover’s ageing porn star betrays “an almost insatiable longing for times past and lost” (395), which surfaces in the author’s persistent invocation of early postmodern texts through his pornographic puns (“Lust in the Funhouse” [297], “Invisible Titties” [390])—but they do not do so in the negative act of rending an earlier career in two. Instead, Coover’s late works characteristically return to origins to extend and continue earlier works rather than to efface them. In the most obvious example, Coover returned in 2014 to present a massive expansion and continuation of his first novel Origin of the Brunists (1966), in his thousand-page extravaganza, The Brunist Day of Wrath. Three years later, he performed a comparable return-and-resurrection act, reaching back beyond postmodernism to Huck Finn, and creating the sequel Huck Out West (2017). While Said presents Ibsen’s career arc as a kind of self-erasing circle, with late works disrupting and undermining earlier wholeness, the accurate visual metaphor for Coover’s exemplary late postmodernism is a spiral: perpetually bending back on itself, but always as a way of reaching forward.4
5More fundamentally, differentiating late postmodern fiction from Said’s formulation begins to unsettle the final term in Said’s label. Whether because of its origins in specifically Germanic linguistic cultural traditions, or from its original singular reference to musical composition, Adorno’s essay—referring variously to style, form, and conventions—bequeaths to Said a certain ambiguity about what the term style precisely indexes that persists through much of his book. This uncertainty becomes more pronounced when the term is transposed into the narrower confines of literary criticism. Although literary style might stretch in different directions depending on context, the “partial list” of stylistic elements that Robert Alter offers in Pen of Iron (2010) is reasonably representative of the term’s general meaning in literature. For Alter, style encompasses “sound (rhythm, alliteration, assonance, and so forth), syntax, idiomatic usage and divergences from it, linguistic register (that is, level of diction), and the cultural and literary associations of language” (26). These are sentence-level effects: style is something that we experience in the fine-grained lexical nuances of word clusters. The style of Said’s late style is something much more amorphous, and so when the late art object in his analysis is literary—Ibsen, Lampedusa—his analysis is never of the way lateness manifests itself in a writer’s sentences.
6With certain specific writers—say Don DeLillo—I think it is possible to demonstrate in Alter’s terms that a late postmodern writer’s style differs from an earlier characteristic style. Consider a characteristic mid-career sentence from DeLillo’s White Noise (1985), describing Jack Gladney falling asleep:
I fell at once into marine oblivion, a deep-dwelling crablike consciousness, silent and dreamless. (155)
At a superficial level, the distinctive stylistic trait, here, is syntactical: DeLillo’s midcareer sentences express a paradoxical mixture of richness of evocation and doubt about language’s ability to precisely describe a thing by clustering clauses that make multiple stabs at definition. These clauses move us from a general “marine oblivion” to the more specific suggestion of a deepwater crabmind, before pulling back for the much looser final description of a mute, dreamless state. There is an epistemology embedded here that parallels the novel’s scepticism toward academic knowledge, in the gradual dilution of apparent precision as we move through the clauses. At the same time, as the sentence begins to describe relationships of similitude—between Jack and crab—so the sentence’s shape alters to formally stress similarity: sonic pairs become important, as DeLillo offers both two alliterative pairs (deep-dwelling crablike consciousness), and an internal rhyme to join two clauses (“-ness” and “-less”). It is possible to go deeper here into DeLillo’s avowed craft as a “sentence-maker” (1991 162): DeLillo likes to track the visual movement of letters across a sentence, and so as Jack describes his fall “into marine oblivion,” so the letters in fall into marine and then gradually dissipate, as if in water themselves, across oblivion.
7It is the consistency of this stylistic tic across much of DeLillo’s middle period that makes it so easy to detect his influence on a writer such as Rachel Kushner, who similarly runs together multiple clauses at the end of her description of a plane landing, leaving “nothing left of the sky but a gunmetal undercarriage, an enveloping screech” (17). Late period DeLillo, by contrast, probably begins with The Body Artist (2001) and brings with it a much sparser style:
This shift from past tense to present had the sound of something overcome, an obstacle or restriction. (49)
- 5 Moving in a different direction—with an eye on the temporality of lateness—Kevin Brazil, challenges (...)
A fundamental hesitation over language remains here signalled this time by the way the sentence stalls over its single comma. But the efforts at clarification of the vague something are briefer now—a single clause—and much less playful than in the earlier text.5
8But while style might calibrate relative lateness in DeLillo, it is less effective as a measure of other late postmodernists, and so the precision of the label late style falls into question. The sprawling verbal landscape in Gaddis’s Agapē Agape (2002), for example, is not strongly different in a given sentence from those that fill his masterpiece, J R (1975); the difference lies in the larger form of the work: its contraction from dia- to monologue. The punning verbal exuberance of Coover’s sentences in Brunist Day of Wrath differs from those in the well-behaved Origin of the Brunists, but are close to those that make up the larger body of his output. What differentiates this late work from all of Coover’s earlier writing, is the massive formal expansion of the book itself. In this respect, Gaddis and Coover are, I think, more representative of late postmodernism than DeLillo. The challenge these writers faced was not to invent a new sentence-level style to register lateness, but rather to find a form adequate to convey it. In their late works, many postmodern American novelists seek a narrative shape that allows them to register and reflect upon lateness, whilst also returning to their earlier texts in a fashion that allows them (like Coover) to spiral forward into their late period.
- 6 Since 2016, when the landmark John Barth: A Body of Words was published, there are full-length arti (...)
9In this essay, I take John Barth’s late work, Once Upon a Time (1994), as an exemplary case of how a postmodern writer wrestled with form, rather than style, to explore, and at times, resist lateness. The book presents itself as a kind of memoir, in which a stalled sailing trip and an encounter with Barth’s double—J. W. Scribner/Schreiber—takes us back in time to review the origin and writing of Barth’s first six books, as a way of carrying his career forward. As such, Once Upon a Time is particularly revelatory in this context in part because it is—with Gaddis’s A Frolic of His Own (1994)—arguably the earliest late postmodern text. But it is also worth studying in this context because while a sizeable critical industry quickly formed around Barth’s early publications, his late works have been almost completely overlooked by critics, with even recent scholarship tending to treat Barth purely in terms of his early fictions.6 I want to demonstrate that while Once Upon a Time is an instructive example of late form, it is also—on its own terms—an unusually sophisticated artistic achievement that merits and rewards as much consideration as those early texts. But to understand how Once Upon a Time challenges Said’s conception of lateness, it is helpful to begin by considering larger objections to lateness studies.
10Michael and Linda Hutcheon critique studies of late style for purporting to reveal a universal phase in artists’ lives, when they actually unmask “the aesthetic values of the critic” (5). Thus Georg Simmel’s study of late Goethe tells us more about Simmel’s own preference for “wholeness, coherence, [and] synthesis” than it does about Goethe’s evolution, while Adorno’s and Said’s accounts project their taste for “fragmentation, dissonance, [and] lack (or impossibility) of reconciliation” over the art’s reality (5). Given this tendency, the Hutcheons ask, why should we even bother to persist with a critical approach and vocabulary that potentially does little more than index a critic’s poorly hidden agenda? The answer to this question, and the important distinction that I think makes lateness a valid framework for approaching postmodernism, is that in the case of many American postmodern novelists—and especially ageing, white, male novelists—it is not the critic who proposes lateness as a frame for thinking about these works, but rather the novelists themselves. This tendency is even evident in the portentous-sounding titles of many of these millennial novels: Gaddis originally planned to call A Frolic of His Own (1994), The Last Act (478n); David Markson’s final works propose a Vanishing Point (2003), before The Last Novel (2007); Lee Siegel’s sequence of Love-novels climaxes with Love and the Incredibly Old Man (2008); while DeLillo’s slim, late works move toward the Point Omega (2010), before leaving us with The Silence (2020).
11On the surface, the title of John Barth’s Once Upon a Time: A Floating Opera indulges in none of this grandstanding play with last things. Yet while the title is ultimately more subtle than it first appears, Barth saves his most explicit appeal to lateness until very near the end of the book, when his author surrogate describes the text we have been reading as “what I couldn’t and can’t resist thinking of as my Last Book” (382). Barth immediately qualifies this claim as “a touch […] self-dramatizing,” but the distinctive part of the quote is less the familiar postmodern appeal to lastness, and more the generic haziness of the word book, which looks so pale next to his contemporaries’ choices. As a proposed title, Gaddis’s The Last Act is at least a relatively precise designation for a text that includes long extracts from Oscar Crease’s play, Once at Antietam. The Last Novel similarly speaks quite specifically to an ongoing thread that winds through Markson’s final books, probing the relationship between authorial intent and generic classification. Last Book, by contrast, is a deliberately ambiguous label, and its strategic choice becomes much more evident when we set Once Upon a Time next to his second collection of essays, Further Fridays (1995).
12Published just a year apart, the two books were clearly parallel endeavours, and in some ways share overlapping purposes: just as Once Upon a Time revisits and reimagines his earlier writings, so Barth describes some of Further Fridays’s essays as exercises in “surveying my own backtrail” (x). More suggestively, Further Fridays does not simply share topical interests with Once Upon a Time, it actively repeats parts of the earlier book. In some cases, this is simply a matter of recycled phrases: “solipsism à deux” (1995 20; 1994 5, 163, 223); “coaxial esemplasy” (1995 282, 334; 1994 20 & passim); reveries over the binder “bought during freshman orientation week in September 1947” (1995 191; 1994 21). In other cases, whole sections are included in both texts: most readers will spot that both texts include variants of Barth’s analysis of his exegetical aria on “Jack and Jill” (1995 209-221; 1994 210-220), but might not spot that all but the first paragraph of the essay “Kenosis” first appeared in the earlier book (1995 158-63; 1994 9-15); a long paragraph from “Teacher” has a similar origin (1995 15; 1994 371); while Barth’s short sketch of modern literature’s split into “James Joyces on the one hand and James Micheners on the other” makes the cut twice (1995 357; 1994 24-25), as does a supporting article from the New York Times (1995 358; 1994 24n).
13Such doubling between books might seem careless or frustrating, although ultimately repetition and rearrangement are very much at the complex heart of Once Upon a Time. What these overlaps do illustrate, however, is the extent to which we are witnessing generic border crossings across these two books: what Barth, in 1994, calls fiction, can, in 1995, equally pass as essay. That Once Upon a Time—like the autofiction and autocriticism whose stylistic blurrings it anticipates—evades easy generic classification, even for Barth, is further underlined by the sheer variety of labels he reaches for to describe this book across his essays. Further Fridays variously presents Once Upon a Time as a “novel-cum-memoir” (131n); an “arabesque” (335); a “semi-novel/ semi-reminiscence” (187); a “memoir wrapped in a novel” (239); a “complex narrative argyle-sock” (191); and a “sort of novel” (333).
- 7 Once is, on one level, a repetition of LETTERS. Both are “stock-taking novel[s]” (1994 120), and bo (...)
14Each of these descriptions deserves an explicatory essay in itself. The analogy of narrative as sock, for instance, might be traced back to Walter Scott’s meditation on narrative coherence in his preface to The Abbot (1820), where the “ingenuity which brings to a general combination all the loose threads of a narrative” is likened to “the knitter at the finishing of her stocking” (xxvi). Moving forward it would surely pick up Virginia Woolf’s To the Lighthouse (1927), where Mrs Ramsay knits her stocking while her husband reads “old Sir Walter’s” The Antiquary (1816) (128), as Woolf, herself, knits together multiple moments, minds, and memories. Finally, it would reach LETTERS, where Barth likens the design and construction of the book that weaves together his six preceding novels to “the knitting of argyle socks” (xvii).7 There is something to both this ancestry and the idea of interwoven, repeating fabric designs that speaks to Once Upon a Time’s late form, but Barth’s uncertainty about how to label the text in his essays carried over into his composition of Once Upon a Time.
15In 1990, Barth was at work on two books. The first, The New Science, had a title whose dual resonances stretched in opposite directions, reaching back to the past via Giambattista Vico, whilst also looking forward, by echoing the subtitle to James Gleick’s Chaos: Making a New Science (1987). Vico’s importance is partly structural—just as Vico’s New Science was split into five books, so Barth’s proposed novel was planned to have five parts. Gleick’s significance rests more in the content’s investment in Chaos theory, though Barth’s book’s purview is broad, as illustrated by the way the dense, allusive prose runs from Flaubert, through Joyce, to Beckett, while it simultaneously weaves its way through quantum physics, Chaos theory, Thomas Kuhn, and C. P. Snow:
What John would really to write (heaven forbeth!) is Al’s maxinovel about nothing; that’s Mary’s theory of everything, sensitive to Eve and Adam’s riverrun effect. That good book is a quarky attractor of anomalies; yet our search for a new paradigm meets resistance from both cultures, and there’s a whole third world out there; we can’t go on like this, we’ll go on, unnameable. (Barth, The New Science, undated)
The book is not obviously autobiographical in the manner of Once Upon a Time, though it does rehearse versions of characters who will appear in the later published text: J.W. Scribner narrates the opening section, while his wife, Beth Duer, also features.
16The second, clearly related project, seems always to have been titled Once Upon a Time, but its subtitle varies in a way that reflects Further Fridays’s drifting descriptions of the work. In Barth’s notebook 4 (1990-2007), the work was conceived as “A Floating Lifestory.” In the first draft, dated 1990, the subtitle had become the more explicitly autobiographical “Floating Memoir”; sometime later, Memoir is scored out, in favour of the abstract “Arabesque.” Finally, Arabesque is itself scored out, and replaced by “Opera,” a designation that is securely in place by later typed drafts, dated 1993.
17The point here is not that these labels suggest unexplored forking paths—invocations “splintered off from the main thrust of the compositions,” to be left, “fallen away and abandoned” (Said 11). Instead it is that Barth ultimately finds a way to synthesise what might seem entirely incongruous impulses—to write a lifestory, a memoir, an arabesque, a work modelled on both an argyle sock and the new science of Chaology. Where Said’s late style is exemplified by Beethoven’s negative achievement—works that are “about lost totality,” that “remain unreconciled, uncoopted by a higher synthesis” because “their irresolution and unsynthesised fragmentariness are constitutive” (13, 12)—the art of Barth’s late form invokes polarities primarily to absorb and contain them. Again, Barth’s title is revelatory: on the most basic level, Once Upon a Time points, of course, to beginnings, whether to the stock opening of fairy tales, or, more formally, to the origins of storytelling in oral folk tales. Yet as Barth renders the title within his book, it points toward endings: abbreviated to “OUT,” it signals that the ageing author is making his way toward the exit (1994 23). Even more confusingly, within the total body of Barth’s work, the title echoes Lost in the Funhouse’s (1968) “Frame-Tale” (“Once upon a time there was a story that began” 1-2), in whose looping, infinite repetition the phrase is neither beginning nor end, but endless middle. It is (like The New Science) a deceptively simple title whose multiple resonances stretch both back to origins and forward to the end, whilst nevertheless circling on the edges of infinity.
- 8 Earlier studies, such as Gordon Slethaug’s Beautiful Chaos (2000), discuss recursive symmetry in su (...)
18The book’s formal design repeats this same forward and backward motion, with the forward energies reflecting Barth’s readings in Chaos Theory during the late 1980s and early 1990s.8 Chaos theory is, on one level, the part of Once Upon a Time that most clearly catches the intellectual overspill from the unfinished manuscript of The New Science. Once Upon a Time contains a sequence of explicit references to Chaos Theory, positing History as “a Mandelbrot Set” (180), emerging novel ideas as a “large Attractor” (73), and updating his earlier publication to produce “lost in the fractal funhouse” (226). At a more structural level, Barth proposes the concepts of “Weak Chaos” and “self-organized criticality” (17, 20) early in the book to account for several facets of his narrative, which are informed by his source text, Per Bak and Kan Chen’s “Self-Organized Criticality” (1991). Focusing on the way “large interactive systems perpetually organize themselves to a critical state in which a minor event starts a chain reaction that can lead to a catastrophe” (46), Bak and Chen offer the example of “the dynamics of sandpiles.” In this model, grains of sand are dropped, one at a time, onto a flat surface, and when the subsequent heap of sand has stopped growing (because the amount of sand falling off the heap is the same as the amount being added), the system is said to have reached a critical state. At this point, any grain of sand has the potential to start an avalanche. An observer of a localised point could predict the likelihood of an imminent avalanche, but could not predict whether the avalanche would be small or catastrophic, because the scale of such events is “a consequence of the total history of the entire pile” (48). In a fully chaotic system, uncertainty about initial conditions exponentially expands to make predictions impossible; self-organized critical systems (and Bak and Chen expand outward from grains of sand to include earthquakes, traffic flow, weather systems in this category) are “weakly chaotic” (52), which means that they “evolve on the border of chaos,” and can be subject to longer-term forecasts: in this case, predicting the likelihood, if not the scale, of an avalanche.
19Bak and Chen propose that weak chaos is “very common in nature” (52), and such iterative processes—aggregating small repetitions toward the threshold of a critical state—pulse throughout Once Upon a Time, underwriting the book’s character interactions, structure, and larger epistemology. At the level of character relationships, some of these moments are explicitly flagged. Act One’s climactic “brawlbrat brawl,” for instance, is a consequence of a sequence of Barth and Jay’s interactions in which “Criticality self-organizes; differences accumulate like vapors from a basement gas leak […] until any spark blows the roof” (204). Other conflicts do not name the concept, but clearly follow the same logic, as in the early blowout between Barth and his partner, where the uniform accumulation of grievances over time that leads to the threshold are underlined by Barth’s anaphoric construction: she is angered by “his characteristic occasional recklessness, his absentmindedness […] his hyposensitivity, his emotional shoal draft” (83).
20The same concept is repeated at a more overarching level that bends back to reflect on the text’s architecture and existence. “The fact is,” Barth explains of the point at which his apprentice writings gave way to something more substantial:
that something was happening, changing, clarifying, though still not quite crystallizing. A mostly undramatic but nonetheless seismic ground-shift was reordering my sensibility: a shift compounded of accumulated narrative experience and reading, plus that really rather scarifying moral wring-out with the Scribner/ Duers […] plus a consequent more or less harrowing re-examination of myself and my “belief environment”—and the serendipitous extension of my Dorchester Tales research to nineteenth-century blackface minstrel shows in general, and, in particular, to a Chesapeake showboat. (282-283)
The word seismic is the first clue that Barth wants us to read this iterative process in terms of a natural system approaching a critical state, but in a way that reflects the book’s holistic totality, the sentence’s structure, also imitates the process it describes. The first sentence offers an accumulation of verbs—happening, changing, clarifying—that do not particularly modify each other, as much as they clump together in a small linguistic sandpile. The second sentence’s logic also points toward accumulation, aggregating largely unrelated items in a long chain sentence, strung around four ands and two pluses. Building on this, self-organised criticality becomes, for Barth, not simply a description of how he crossed the threshold to become a writer, but the principle that underwrites his longevity. Reaching, again, for the scientific vocabulary of natural processes, Barth applies self-criticality to his career, in a sentence that culminates at the point where sheer repetition is more important than punctuation. Barth explains that word by word, “the sentences get written […] and like molecules into complex organic compounds, they aggregate into narrative passages, pages, chapters, books books books” (323).
- 9 Retirement; turning 60; the end of a spiral notebook begun in 1967.
- 10 Zack Bowen persuasively argues that while opera’s have three complete acts, Barth’s book actually f (...)
21The same iterative aesthetic can be applied to the book’s origin, which Barth attributes to the critical accumulation of a series “close-outs” (17),9 and even to the book’s larger structure. Ostensibly organised as an “opera,”10 and split into eight parts, the narrative, regardless of these divisions, is consistently relayed as a series of short takes, mostly headed by an italicised phrase, that range in length from a short paragraph up to thirteen pages, aggregating into the text’s total form. Viewed in the light of self-organized criticality, and running counter to the book’s imagined time travels, the rhythmic beat created by accumulative processes operating across multiple scales—whether of small narrative units accumulating, petty aggravations between personalities, or phrases clotting together in a sentence—acts not just as a reminder of time’s arrow, but also of the welter of natural processes that ground the narrative: from the lunar rhythm of the tides, to the ebb and flow of life’s passage.
22Yet while self-organised criticality is one aspect of Once Upon a Time’s Chaos-informed structure, the book’s complex, layered form is also designed to offer a literary analogue of the so-called Koch Snowflake. First described in 1904 by the Swedish mathematician, Helge von Koch, the Koch Snowflake is created by taking an equilateral triangle, and progressively adding triangles one third as large, to the middle third of the initial triangle’s sides. By repeatedly following this simple rule an unusually complex shape is generated, which, in an image Barth reproduces in Further Fridays (316), James Gleick illustrates as follows (Fig. 1):
Fig. 1. The Koch Snowflake
James Gleick, Chaos: Making a New Science. New York: Viking, 1987 (99).
23The intriguing aspect of this process is that, while the total area covered by the snowflake will never exceed “the area of a circle drawn around the original triangle” (Gleick 99), the perimeter of the shape increases with every iteration, and theoretically stretches toward infinity. The Koch Snowflake therefore creates the paradoxical combination of “infinite length in a finite space.” The result “disturbed many of the turn-of-the-century mathematicians” who thought about it (Gleick 100) but provided Benoît Mandelbrot with a “preliminary model of a coastline” (Mandelbrot 40), inasmuch as it demonstrated recursive symmetry: identical patterns repeating at different levels of magnification.
24To parallel this form, Barth builds a recursive structure, where triangular shapes repeat at different levels of magnification, into his book. At the largest scale, Barth reminds us that the entire book follows a triangular pattern, and includes a short sketch of Freytag’s classic “nonequilateral triangle” (78) (Fig. 2). Zooming in from the overall shape of the book to an individual narrative unit, Barth titles one section “Trianglish aria” (75). The setting is provided by an area that Barth names “the Chesapeake Triangle” (16), after the larger and more famous “Bermuda Triangle” (76). At a smaller scale still, the waves in this area are triangular—“frothing pyramids” and “water pyramids” (97)—while the boat atop them has a sail that’s a “full foretriangle” (85), and a bed “in the mini-triangle” of the ship’s “forward V” (101). The boat navigates in relation to “triangular red […] dayboards” that mark channels (90), while the characters aboard the triangular boat themselves experience anger that is “three-sided like” the “triangle” shaped space they occupy (83).
Fig. 2. Sketch of Freytag's “nonequilateral triangle"
John Barth, Once Upon a Time: A Floating Opera. Boston: Little, 1994 (78).
- 11 The book tends to pose key conversations between two characters in presence of a largely silent thi (...)
- 12 Further Fridays was, of course, published after OUT, but most of the essays precede it.
25There are plenty more threes in the book that might be catalogued—not least the “three fading calls” that open the overture (1)—as well as a consistent, repeating pattern of triangular relationships.11 But aside from demonstrating Barth’s mastery of his text’s architecture, this series of repeated shapes across multiple scales serves three clear purposes. First, in terms of content, it provides (via Mandelbrot) a natural model that justifies the principle of repetition with difference that accounts for three kinds of repetition in Once Upon a Time: (1) the book’s synoptic return to his earlier novels; (2) its general repetition of “riffs that [he’s] been noodling for forty years […] floating operas, water-messages and night-sea journeys, lost paths and last voyages” (382); and (3) its specific repetition of passages that appear in his essays.12
- 13 McLaughlin acutely reads such juxtapositions as emphasising the difference between the real world ( (...)
26Second, at the level of narrative viewpoint, it incorporates the idea of scalar shifts to explain the way the book’s supple narrative voice elastically stretches between different perspectives. This covers, in part, those Barthian metaleptic jumps between narrating 1992 and reflecting on the time of writing in 1990. But it also encompasses the sequence of strategic narrative zoom outs, some of which suddenly and massively enlarge the scale of the narrative’s temporal lens, taking us out of the narrative present (“leisurely once-upon-a-timelessness” [1995 67]) and into a perspective inflected by “geological time” (36), by “eons of evolution” (45). These radical temporal expansions are inevitably counterbalanced by the viewpoint’s corresponding tendency to periodically zero in on the present moment, laying narrative time next to historical moments: “ I foot this note shortly after Labor Day ’92, the season’s maiden Atlantic hurricane having lately devasted first Bahamian Eleuthera […] then south Florida […] and finally part of Louisiana” (343n). These latter shifts, which serve to reframe and reconnect art and history, become, as Robert L. McLaughlin has shown, one of the recognisable fingerprints of Barth’s late works.13
27Finally, the Koch Snowflake’s paradoxical blend of finite scope with infinite boundary stands as a formal figure for Barth lateness, a way that the book’s structure seeks to resist the relentless ways that “Time’s arrow streaks forward” (142). While the book’s final line works in recursive style to rearrange the title’s specific identification of a single moment (“once upon a time”), and acknowledge the irresistible linearity of our single passage through a lifetime (“the time is once upon” [398]), the Koch Snowflake structure works to artfully resist that movement. That is, in a millennial update and reversal of Laurence Sterne’s Tristra-pœdia, in which Tristram’s father’s efforts to write a “system of education” for his son are “rendered entirely useless” because the boy grows faster than his father can write (Sterne 336, 338), Once Upon a Time writes moments in the form of a Koch Snowflake to deepen and elongate time, stretching the narrative present to infinite length to postpone the finite end.
- 14 This combination stems from The New Science, in which Schreiber’s “son,” Tom, plays a computer game (...)
- 15 Columbus Day marks the explorer reaching the Bahamas on 12 October 1492, and Barth places that date (...)
28Merging Weak Chaos and the Koch Snowflake would provide a sufficiently complex narrative form for most novels and most writers, but, for Barth, this pattern accounts for only one half of the book’s design. Counterbalancing the forward-looking embrace of emerging Chaos theories, Barth reaches back in time to make many of the same formal features echo the ancient Islamic “arabesque” (1994 324).14 Drawing variously on Friedrich von Schlegel, Michael Craig Hillmann, and G. R. Thompson, Barth outlines the different ways that the fractal patterns that we’ve located in his book are paralleled by Persian carpet designs, which depend upon “highly intricate patterns of geometric designs […] involving structurally repeated […] interpenetrating […] triangular” structures (Hillmann quoted in 1995 312). But aside from doubling the resonance of the book’s form by adding a historical dimension to its design, the arabesque serves another key structural function. Hillmann explains that Persian carpet makers sought to reflect the world’s imperfections by deliberately incorporating errors and inconsistencies that fracture the illusion of a perfect pattern (1995 313). Echoing this intention, Barth works a “calendrical-error” into his plot (1994 388), suggesting at the book’s end that, in 1990, he miscalculated on which date Columbus Day would fall in 1992, when he imagines it at the start.15 At the same time, Barth strategically breaches the perfect symmetry of his book’s formal design, by defaulting on his own plan to present “three acts […] twenty narrative yearsworth each” (226), and more or less abandoning that structure after two full acts, and relying on discrete narrative shortcuts to bring about a resolution. These calculated imperfections allow Barth to pay homage to an Eastern tradition (1001 Nights, The Ocean of Story) that was clearly seminal for his work, but they also allow him to correct the “error” that had been levelled against his earlier stock-taking novel, LETTERS. That book, Tom LeClair argued, suffered from an “excess” of “mastery,” in which the “author’s godlike creative control of his text” casts too domineering a shadow over the reader’s experience (176). Carefully constructed to reveal the author’s fallibility as well as his craftmanship, Once Upon a Time evades a repetition of that charge.
29Layering the resonances of its recursive structure with references to ancient carpets, coastline calculations, and theoretical shapes, Once Upon a Time represents Barth’s efforts to forge a form capable of reflecting and recording the opening of the late phase of his long writing life. While Barth’s style characteristically parallels his formal choices, the defining units for addressing lateness in this book are larger than the sentence, and require a distinction between form and style that is not normally made in readings of late works. The neat union of polarised movements—forward and back—introduced by the title, and iterated through the book’s structure, demonstrates little of the refusal of synthesis Said ascribes to late works, but does explain Barth’s ultimate choice of subtitle. Arabesque would not accurately serve as a label, because arabesque describes only one half of the book’s form. Equally, Barth was probably wise to reject “lifestory” and “memoir” as subtitles, since, even in their absence, critics have tended to read the book as “more fact […] than pure fiction” (Bowen 201), and so appealing “more to the committed than the casual reader of Barth’s fiction” because of its “treasure trove of autobiographical facts and details” (Harris 203). Floating Opera, on the other hand, suggests generative repetition, and tries to point toward the books rather than the biography. Yet when even such significant and sympathetic Barth scholars as Zack Bowen and Charles B. Harris seem to allow the stuff of memoir to risk overshadowing the clarity of Barth’s art, the case for the book’s sophisticated form needs to be made.
- 16 Given the proximity of publication dates, no line of influence is implied, or even possible here. T (...)
30If the phrase once upon a time is a beginning that Barth turns into an ending, then the book, Once Upon a Time addresses ending in a way that also makes it a beginning. As the first book in what we might think of as Barth’s late phase, Once Upon a Time begins the critically neglected sequence of often subtle and innovative fictions that close out Barth’s career over the next seventeen years. Several of these works reprise or expand upon devices and concepts introduced in Once Upon a Time, such as “Ad Infinitum” (1996), which similarly poises infinite divisibility against time’s arrow. Others parallel its tendency to return to earlier works to spiral forward to something new, such as Every Third Thought’s (2011) return to The Development’s (2008) George I. Newett. But I would like to conclude this essay by suggesting that Once Upon a Time’s complex form might reach beyond the generation of late postmodernists, to be the unexpected ancestor of several works by younger writers. In this sense, then, Barth is so late that he becomes early. His decision to design his heavily footnoted book around the Koch Snowflake has numerous affinities with David Foster Wallace’s notorious plan to pattern his later Infinite Jest (1996) after a Sierpiński Gasket (Gleick, in fact, summarises both forms on the same page of Chaos [100]). While it is probably a coincidence that both Barth’s and Wallace’s books were edited by Michael Pietsch, Barth’s early announcement of a shift in metaleptic levels in his mock autobiography—“Author speaking” (1994 8)—seems to be clearly echoed by Wallace’s invocation at the start of the fake autobiographical sections of The Pale King (2011): “Author here” (66). While, as Zack Bowen notes, Barth’s survey through his early works in Once Upon a Time is a prose analogue of W. B. Yeats’s “The Circus Animals’ Desertion” (1939) (196), a similar form is employed by Richard Powers’s Galatea 2.2 (1995)16 and Bret Easton Ellis’s Lunar Park (2005). If Barth’s synthetic artistry challenges the sense of late work as incoherent catastrophe, its prescient qualities might nevertheless support Said’s conception of the late writer as someone “beyond their own time, ahead of it in terms of daring and startling newness” (135).
BARTH, John. The New Science. Unpublished. Undated.
Flux RSS
