Fictional Archive: The Novel as Information

As my opening remarks on the novel’s archival impulse indicate, writers as well as literary theorists have consistently regarded the novel as a genre with a special stake in information. In this view, the novel’s expansive and detailed accounting of the world produces not just a representation but a record. The notion that the novel acts as an archive, even a reference book, recording information about the time and place it describes, has been persistent enough that critics identify subgenres such as “archival novels” and “encyclopedic novels.” 16 Seen in this light, novels like Ulysses may be unusual in the degree to which they highlight their archival and medial dimensions, but they are symptomatic of a much broader generic investment in information.

There are several related but distinct reasons why the novel has garnered its reputation an informational genre. First, because novels are typically written in prose, they occupy a discursive middle space between writing that is overtly marked as literary and writing that is not. How to define the literary is a question I take up in chapter 2; for now, I note the critical tendency to define it as writing that has qualities beyond the “merely” communicative or informational—writing that, as Arthur Bahr puts it, is defined by its “excess,” its “refusal to submit to the denotative.” 17 Although novels use literary techniques and devices that would be lost in a summary of the plot, paraphrasing a novel seems to be a lesser transgression than paraphrasing poetry is. (Tellingly, Cleanth Brooks’s examples of the heresy of paraphrase are all poems.) If “the fundamental aspect of the novel is its story-telling aspect,” as Forster argues, it follows that novels would be less invested in emphasizing their materiality because “informational texts seek to minimize their perceptual features in the belief that texts calling attention to their vehicular forms interfere with the transmission of their ideas.” 18 The convention (however problematic or reductive) that novels are read primarily for their content aligns them as much with informational texts as with literary ones.

The claim that novels contain information is most often derived from the truism that they capture something real despite the fictional status of their narratives. The novel functions like an archive, this argument goes, because its narrative content bears a close relationship to real life. Watt’s view on this matter has been foundational: “There are important differences in the degree to which different literary forms imitate reality; and the formal realism of the novel allows a more immediate imitation of individual experience set in its temporal and spatial environment than do other literary forms. Consequently the novel’s conventions make much smaller demands on the audience than do most literary conventions.” 19 While realism is far from the only mode in which the novel operates, it has been fundamental to critical definitions of the genre. J. Paul Hunter writes that “realism is a relative matter, but in discussions of the novel, the term has tended to become normative, so that novels tend to be judged qualitatively on the degree or amount of realism to be found in each, as if more is better.” 20 The novel’s realism has taken many forms, from historical realism (the accumulation of details registering the habits of daily life in a certain time and location) to cognitive realism (the realistic representation of thought).21 Scholarship on the novel consistently characterizes it as the genre of the nation, or of modern subjectivity; whether one views a particular novel as realistically capturing the workings of one mind or the quotidian details of life in a particular place, the assumption is that there is a stable referent that the novel represents. This is the paradox of the novel’s archival nature: even when describing a fictional referent—the mind of Leopold Bloom; the social milieu of Casterbridge, Wessex—a novel nonetheless records some real-world truth that this fictional subject models.

A focus on realism, however, does little to explain why novelists began to move away from this style at the same time that information proliferation and its management became major issues. Although we might say that the novel has always incorporated information in some manner, novelists since the modernist era have undertaken an archival agenda more indebted to the forms and genres of information media than to those of realism. Where an early novel reader might have encountered information about the world (in the guise of “the real”) via a set of fictional letters or journal entries, readers of modernist and contemporary novels discover narrative texts that mimic paper file layouts or computer code. The chief questions for the novel in the age of modern information management become: What is information? What are the discursive norms that shape its articulation? How might the novel compare to other information media in recording and preserving information? This last question is pressing because many critical accounts of modern information media cast them as antagonists to literary representation, rehearsing the Kittlerian argument (encapsulated in the epigraph to this chapter) that literature’s once-unique ability to store sensory data was imperiled by the introduction of recording technologies.22 Yet the capacity for direct representation is not the only vector by which literary works may compete with information media. Modernist and contemporary novelists have made the case for the genre’s relevance by grounding their engagement with these questions in an investigation of the forms and affordances of the print book.

Before we proceed further, a few definitions are in order. What is information, and how does it differ from data or knowledge? The smallest, most discrete entity of the three is data, which we might think of as “units or morsels of information.” 23 Where a fact is self-evident and true, a datum is a given, a measurement or a rhetorical statement: “data are representations of observations, objects, or other entities used as evidence of phenomena for the purposes of research or scholarship.” 24 Used in the aggregate, data’s explanatory force accrues collectively. On the other end of the spectrum lies knowledge. From the partial to the total: if data are “units or morsels,” knowledge is “the sum of what is known.” 25 Of data, information, and knowledge, knowledge is the closest to what we would categorize as wisdom or truth. It is also the most subjective category, to the extent that knowledge requires “an individual knower.” 26

Between data and knowledge lies information, a notoriously difficult word to pin down. Its meaning is often taken as self-evident. Definitions vary from the broad (“any given [datum] of our cognitive experience that can be materially encoded for the purpose of transmission or storage”), to the technical (“a mathematically defined quantity . . . which represents the degree of choice exercised in the selection or formation of one particular symbol . . . out of a number of possible ones, and which is defined logarithmically in terms of the statistical probabilities of occurrence of the symbol”), to the cryptic (“a difference which makes a difference”).27 Here I settle on a working definition: information is data made meaningful by their contextualization. Information, in other words, is meaningful because it takes individual data points and organizes them or otherwise gives them form, setting them in relation to one another. Form lies at the heart of information. This act of giving form to data is what allows for information’s interpretive power. Whereas data are abstract, information is necessarily formed and articulated. There is still an autonomy to this formal organization; while knowledge requires a knower, information does not.28

As a result, information is the term that hews most neatly to the category of fiction. A novel is a fictional work; its epistemological status differs from that of discourses such as scientific writing that purport to objectively describe the factual state of the world. Yet a novel may integrate information about the world, processing this information by embedding it within its narrative. Some novels even include lists of facts or other data structures. For the most part, however, a novel is less likely to contain facts—declarative statements about the true state of reality—than that fuzzier, more interpretive and rhetorical category of information. “The novel,” wrote Forster, “whatever else it may be, is partly a notice board”; he characterized it also as the genre in which “information abounds.” 29 This statement implies not only that novels are archives of the cultures that produce them but also that the novel’s archival drive is specifically informational. It is not simply that a novel contains an impression or representation of reality but that it contains specific details, meaningfully arranged, in forms that are similar to nonfictional, nonliterary informational texts.

This explains why critical descriptions of novelistic form have been so vexed. The genre’s theorists consistently disagree on its defining formal characteristics—or whether it even has any. To cite a few canonical voices: Watt describes “what is often felt as the formlessness of the novel, as compared, say, with tragedy or the ode.” Erich Auerbach speaks of “the broad and elastic form of the novel,” and, for Mikhail Bakhtin, the novel is “plasticity itself,” always “subjecting its established forms for review.” Woolf describes the novel as the “most pliable of all forms.” More recent theorizations make similar claims. To Terry Eagleton, the novel is “the most hybrid of literary forms”; in Michael McKeon’s view, the novel “self-consciously incorporat[es], as part of its form, the problem of its own categorical status.” 30 If theorists of the novel agree on any point, it is that this genre is definable only through its lack of stable formal attributes.

Rather than an absence of form, however, the issue is an overabundance of forms. If the novel is difficult to define, this is a consequence of its expansiveness. Since its inception, it has habitually drawn on a range of genres and discourses, including informational ones. Early novels appropriated different styles of writing, especially nonfictional genres such as travelogue (Daniel Defoe’s Robinson Crusoe and Jonathan Swift’s Gulliver’s Travels) and epistolary correspondence (Samuel Richardson’s Pamela and Clarissa). It was commonplace to use the word history in titles (as in The History of Tom Jones, a Foundling). Early novelists and readers defined the novel in comparison to other discourses that blurred the line between fact and fiction, including journalism.31 Given the practice of imitating and incorporating nonfictional genres, it is not a great stretch for a novel to imitate explicitly informational texts such as encyclopedias or Internet news feeds (examples I discuss in chapters 1 and 2). Formal flexibility is one reason the genre was named for novelty: its push for newness has consistently come through stylistic experimentation with the conventions of other discourses.

At the macro level, then, the novel’s integration of informational discourses is part of its tradition of formal flexibility. At the micro level—the level of the individual page, sentence, and word—novels have incorporated content and textual structures we would not ordinarily consider literary. A double-entry bookkeeping log, a long list of items sitting on a kitchen shelf: such examples go beyond extending the verisimilitude of the narrative world.32 They interrupt the narration with a glut of information, momentarily turning the novel into a reference book. To examine the novel’s informational qualities is to address questions about the genre as a whole: how it is defined (and defies definitions), how it does or does not employ literary language, and how it engages with information as content and as form.

Mediating Scale: Information and the Novel in the Digital Age

In Out of Print, I analyze novels that explore the medial underpinnings of this long-standing generic investment in information. These texts consider how their subjects are doubly mediated, first as narrative representations and second as text embedded in books. While their narratives contend with the task of representing information about places, objects, and people, their unconventional uses of the print book reveal the book’s storage capacities and limitations. This strategy emerged as a literary response to a paradigm shift in information culture. The reason that self-reflexively “bookish” novels (to use Jessica Pressman’s terminology) flourished during the early twentieth century, I am arguing, and the reason they are flourishing again today, is that these are threshold moments in which the scale and mediation of information reached a critical juncture.33 The novelists I study resist modern regimes of mediation, from the systematic management of the modernist period to the corporatization of information under Big Data.

Information abundance is a defining feature of contemporary information culture. (I will discuss information proliferation during the modernist era at the end of this chapter.) In the twenty-first century, human communication creates 5 billion gigabytes of information every two days.34 Digital devices capable of producing and storing historically unprecedented volumes of information are omnipresent, and information overload is enough of a cliché to have spawned its own Internet vernacular—“TMI,” “TL;DR.” Contemporary information culture is also distinguished by the mediation of scale through technological management. Whether through a command line, a graphical user interface, or one of the so-called invisible interfaces of ubiquitous computing, the user always interacts with a mediating system.35

More recently, information mediation encompasses the algorithmic analysis of large data sets. This capacity underlies Big Data, the phenomenon that exemplifies how thoroughly scale and its mediation have transformed people’s relationship to information. Although the term entered common usage only in the second decade of the twenty-first century, Big Data quickly became credited with changing the nature of what could be investigated and known.36 “At scale,” writes Christine Borgman, “big data make new questions possible and thinkable. For the first time, scholars can ask questions of datasets where n = all.” 37 According to danah boyd and Kate Crawford, “the widespread belief that large data sets offer a higher form of intelligence and knowledge that can generate insights that were previously impossible” has become akin to “mythology.” 38

Big Data is best understood as a matter of relative rather than absolute scale—“data of a very large size, typically to the extent that its manipulation and management present significant logistical challenges.” 39 Big Data is too much data to be able to manage or use without the intervention of sophisticated systems. Even standard software falls short of the task. Big Data is mediated: without algorithms, sophisticated processing software, and other management systems, the data cannot be usefully analyzed. Big Data, then, is not a set quantity of data so much as a set of attributes or assumptions about what can be done with large, complex data sets. It encompasses the physical forms that data take and the technologies used to analyze them as well as ideas about data: the implications of scale; the ways in which technology mediates our perception of data; the distinction between raw data and meaningful information; and the explanatory power of Big Data. If overload is one side of the information abundance coin, Big Data is the other, for it promises that otherwise-excessive collections of data may be managed. As information is managed by complex systems, however, users cannot directly view either its organizational form or its material form. Information abundance dictates that when we attempt to encounter information at scale, we do so without trying to actually comprehend that scale, without grasping what all of the data mean as a gestalt. If we access individual pieces of a data set, we access them as fragments with no meaningful relation to the whole; more likely, we allow search, analysis, navigation, or visualization programs to analyze the sum of the data and return results and conclusions. This is a useful process, but it is distinct from understanding the information in and of itself.

The conditions I am describing—information scale, the dominance of digital media, and technological mediation—would seem to make obsolete the novel’s ability to record information. Indeed, the idea that the contemporary information ecology is antagonistic to the novel surfaces frequently in discussions regarding the transition from print to digital media. Digital media, one argument goes, alter human cognition by privileging hyper attention over deep attention—the latter being necessary for novels, whose length and complexity necessitate prolonged, intensive reading.40 Without wishing to reify this view, I note how commonly scholarship on the attentional impact of digital media has used the novel as a reference point. Nicholas Carr and Sven Birkerts cite novels by authors including Charles Dickens and Henry James as exemplars of texts that require the sustained, contemplative reading they fear the Internet is destroying. Hayles similarly associates deep attention with this genre: her examples of activities that require deep attention include reading “a novel by Dickens” and Jane Austen’s Pride and Prejudice.41 While accounts of digital media’s effects on cognition range from measured to alarmist, such concerns speak to a belief that the novel is linked to a print-centric reading culture and that it is at odds with digital media as a result.42

Another line of argument views the novel as inherently opposed to digital information. The most obvious contrast is one of scale. The novel is the literary genre known for, and typically defined by, size. In Forster’s formulation, for example, a novel is “any fictitious prose work over 50,000 words.” Naomi Baron notes that “the novel as a literary genre entered English with a weighty presence”; she cites Penguin Classics editions of Richardson’s Pamela and Clarissa as being 544 pages and 1,536 pages, respectively.43 The tendency to conflate the novel with the print book likely has to do with the fact that the former could not have flourished without the latter because of the genre’s noticeable length. As Walter Benjamin wrote, the novel is “distinguishe[d] . . . from the story (and from the epic . . .) in its essential dependence on the book” because “the dissemination of the novel became possible only with the invention of printing.” 44 To produce and circulate numerous copies of a long text, a publisher needed the mass-production technology of movable type and preferably the codex format to organize the pages. But a single novel that is very long—or even serialized, such as Danielewski’s multivolume The Familiar (whose first five of twenty-seven planned volumes run to more than 4,000 pages)—is dwarfed in scale by digital information storage. The amount of data needed for Big Data analysis is at an even greater degree of remove. When novels attempt to integrate information into their narratives, that information must necessarily be of a far lesser quantity than can be stored by newer media.

Other theorists have argued that the novel is (or appears to be) obsolete in the age of digital information because computational methods of storing and processing information differ profoundly from literary narrative in their ability to represent and make sense of the world. Lev Manovich has described narrative and database as opposing cultural forms, arguing that database has usurped narrative’s cultural position. While his characterization of these two forms as antithetical (as “natural enemies”) has detractors, even accounts that characterize narrative and database as “natural symbionts” confirm their vastly different modes of operation.45 Manovich’s argument is symptomatic of the view that the culture of digital information is at odds with the novel because of how each approaches explaining the world. Michael Wutz writes that “the novel . . . can yet insist on a notion of information, and on the processes of commuting such information into such . . . terms as knowledge, insight, and wisdom, that are qualitatively different from the binary bits of computer processing,” so that novels can “offer a system of information and knowledge self-consciously different from the computable databases of mainframes and networks.” 46 In contrast, the logic of Big Data—and of the information scale and mediation practices for which it often stands as synecdoche—insists that data analysis explains the world by finding patterns not previously discernable. This is not knowledge or insight but pattern recognition through statistical analysis. If narrative used to be the dominant system for making sense of the world, Big Data, rather than database, is now its clearest challenger. In all of these critical narratives, scholars invoke the novel’s tie to the print book, implicitly or directly: they assume that, as the book becomes a residual medium, the novel, in turn, becomes a residual form.

Metamedia: The Novel and the Print Book

The novels I study contest these intertwined crises of modern mediation: the obscuration of information’s meaningful forms and the apparent displacement of novels and books. As we have seen with the example of Manhattan Transfer, these texts theorize mediation via their own materiality, disrupting the conventions of page and print. Medially engaged literary experimentation—described variously as “technotext,” “multimodal,” and “metamedial”—has become a prevalent feature of the contemporary novel (although I argue later in this chapter that its origins lie in the modernist period).47 The best-known example is Danielewski’s House of Leaves (2000), a novel whose innovative typographical layouts mimic the disorienting architecture of the eponymous house. Other examples include novels featuring blacked-out sections of text, typographical collages culminating in a flipbook, and the simulation of an old, annotated library book.48 These novels could not be translated to a digital screen without losing significant aspects of their composition. “If,” writes Leah Price, “the book has been invisible (or intangible) to most twentieth-century literary critics, it isn’t simply because we aren’t trained to analyze material culture; it’s also because a commonsense Cartesianism teaches us to filter out the look, the feel, the smell of the printed page.” Novels that foreground their textual materiality push back against this invisibility, asserting what Kiene Brillenburg Wurth calls “book presence.” Wurth defines book presence as “the effect of an ongoing process of the becoming obsolescent of the book,” wherein “the book precisely materialized when it became immaterial as an information medium.” 49 Such novels highlight how books function as information media and as literary media.

To refer to this subset of novels, I follow Alexander Starre in using the term metamedial. In the tradition of metafiction, metamedial novels self-reflexively call attention to the artifice of narrative, representation, and literary language. The distinction is this: in metamedial novels, linguistic artificiality is conditioned and compounded by the artifactuality of the print book. Drawing on Patricia Waugh’s definition of metafiction, Starre describes metamediality as “a form of artistic self-reference that systematically mirrors, addresses, or interrogates the material properties of its medium. Literary metamediality therefore draws attention to the status of texts as medial artefacts and examines the relationship between text and book.” 50 Metamedial works constitute a specific subcategory of metafiction, in which play with the work’s material properties forces the reader to consider the work’s fictionality. Studies of metafiction have long been concerned with literary representation. The metamedial novels I examine in Out of Print continue this interrogation into the nature of what, exactly, fiction may describe and encode, but they do so by considering how the novel’s representative qualities dovetail with or diverge from other methods of recording information, as well as how such recording is shaped by media. My use of the term metamedial also evokes the desire evident in these novels to turn books into multimedia meta-media—media capable of subsuming all other media.

Metamedial novels reveal what mediation obscures. We might think of novels as alternative media, leveraging the print book to critique information management’s dominant assumptions about form and mediation.51 In contrast to storage media such as microfilm and computer hard drives, books contain text that is humanly readable. In technological terms, print books operate very differently from interface-based media. At its broadest level of meaning, an interface is any “surface lying between two portions of matter or space, and forming their common boundary.” My use of the term aligns most closely to the concept of a user interface.52 I am particularly concerned with interfaces—whether physical objects or processes—that intercede between users seeking information and the information as it is inscribed in the storage medium. Reading screens fall in this category; so do the protocols that govern information retrieval from complex systems, whether these be algorithms or guidelines for information professionals.

Such interfaces are what chiefly distinguish what I am calling modern information mediation from earlier systems of information management. Like the messenger god Hermes, a medium is both an information vessel and an emissary: it carries its content across space (radio, telephone), time (writing, phonography), or both. Mediation entails a rupture between the creation of the message and the moment or place of its reception. Modern information culture is marked by an even greater degree of removal because of the expectation that information has reached a threshold of scale that can only be managed indirectly. The Internet seems nearly infinite, so we locate what we need through search interfaces; a century ago, when there seemed to be too many books, information professionals began to store data on microform media, which they accessed with reading screens. During the past century, the use of such interfaces has encouraged the perception that information is immaterial and formless, an idea to be consumed rather than data instantiated in a medium.

Viewed in this light, descriptions of the page (or book, or narrative) as an interface gloss over the major differences between words printed on a page and bound in a codex and words stored on a hard drive and read on a screen.53 The former are immediately accessible, combining the site of storage with the site of inscription. The latter, as Lori Emerson describes, by design involve “an interface that recedes from view, ideally to the point of invisibility.” 54 Additionally, the volumetric space of the book, like the two-dimensional space of the page, becomes a framework for understanding the organization of its contents. In these ways, the affordances of the print book work against the principles of transparent mediation so integral to modern information management. My research shows that the value and function of media that feature interfaces has consistently been articulated in direct opposition to that of the book.

As metamedial novels reflect on the print book as an object—material, aesthetic, tactile, structured—they expose and challenge the practices and ideas that have underpinned modern mediation. They also prompt a reexamination of the long-standing relationship between the novel and the book. These texts are thus useful for addressing the interoperation of genre and medium. When scholars have discussed media from the perspective of genre theory, they have tended to view the novel in very general media terms (“writing,” “written narrative”) or to passingly mention that novels are a “print genre” in the sense of being artifacts of predigital or post-oral culture. Even monographs in this field that study the novel’s relationship to media rarely discuss metamedial novels or the print book’s influence on novelistic form.55 Conversely, scholars who take a media studies approach to literature have tended to frame their analysis in broad terms—“book fictions” or “experimental literature,” to give two examples—treating genre as largely incidental to the works’ textual strategies.56 As a result, while Marshall, Wutz, John Lurz, and others have advanced the theorization of the impact that print and the book have had on the novel, this area remains understudied.57

Because metamedial novels foreground this impact, they reveal the usefulness of format as a critical category for the study of the novel and the study of the book. As Jonathan Sterne describes, the category of format is complementary to but distinct from that of medium: “format denotes a whole range of decisions that affect the look, feel, experience, and workings of a medium. It also names a set of rules according to which a technology can operate.” 58 The format Sterne focuses on, the MP3, is a much more compact field of study than is the print book; the former is tightly delimited, technologically and historically. Moreover, we might devote attention to different formats of print books—hardcover, mass-market paperback, trade paperback, and so on. Yet the category of the print book already offers a strategic condensation and combination of media: for instance, print, type, paper, codex, text. A print book is a mini-media ecology. Its format encompasses physical characteristics (its tactile and bound form), organizational systems (the arrangement of pages within a codex and text within those pages), and phenomenological registers (including aesthetics, sensory attributes, and perceptions of aura). Metamedial novels emphasize these aspects, which are attributes of all print novels and which contribute to the processes by which novels make visible the forms of information.

Format is also significant in its semantic adjacency to genre. Both “novel” and “print book” suggest defaults and norms. Formats provide a valuable framework for theorizing defaults because they are not reducible to either their nondiscursive technical properties or the phenomenological domain of their perception. As Amaranth Borsuk writes, “the thing we picture when someone says ‘book’ is an idea as much as an object.” 59 A print book codifies a series of default forms, arrangements, and uses, but its affordances also entail potentiality. Defaults are less interesting in and of themselves than as the standards from which works may deviate. So, too, in the case of novels. Novels, more than poetry, correspond to Johanna Drucker’s definition of the “unmarked” text, “the single grey block of undisturbed text” that allows readers to read words on the page without seeing them as visual or material artifacts: “Literary works . . . essentially adopted the unmarked mode of Gutenberg’s biblical setting as their norm. . . . The aspirations of typographers serving the literary muse are to make the text as uniform, as neutral, as accessible and seamless as possible.” 60 Metamedial novels work against this default aesthetic, disrupting expectations for genre and format alike as they adopt the forms of information and information management. Through these novels, we better understand the aesthetics of information scale.

The issue of defaults and deviations brings me to a final note regarding my rationale for focusing on metamedial novels. To what degree can one generalize about the genre from these idiosyncratic texts? Because metamedial novels are highly experimental, they are necessarily limit cases that test the boundaries of the genre. Yet this very quality makes them exemplars of a genre that has been formed through limit-case experiments. The novel is always already a genre of exception. I will have more to say about exemplarity and novelty in later chapters; here, I propose that, while the self-conscious bookishness of metamedial novels is unusual, it is indicative of the ways in which the novel’s encounters with information have been shaped by the book throughout its history. Metamedial novels not only reveal the conventions of novels and of books; they also remind us that the relationship between the two constitutes its own default.

The novel has never been entirely bound to the book. Historically, many novels were serialized in periodicals, and the connection between novel and book is tenuous in the age of digital e-readers. Novelists are adapting their genre for new platforms such as smartphones, and even print novels are never really “born print” in the manner that digital works are born digital. Today, novels are likely to be born as digital documents.61 In the past, they would have been born as manuscript or typescript. The relationship between the novel and the media forms it takes is not reducible to a print-digital dichotomy. Yet the association between novels and books persists. This association is present in the slippage when a colleague says, “I just read a new book on my Kindle.” It is present in Robert Coover’s argument in his 1992 essay “The End of Books” that hypertext literature would kill the novel by killing the book.62 It is present when we perform a Google Image search for “novel” and the algorithm returns pictures of books. The connection between novel and book may not be absolute, but it is a tenacious and important convention. I map a range of literary responses to information and its mediation, from canonically informational novels such as Ulysses to novels whose investment in modern information culture is less obvious. This analysis explores a wide range of potentiality—what the book may do for the novel and for information storage, rather than what it must do or what it habitually does.

Information Culture Circa 1900: Modern Mediation and Metamedial Modernism

The preceding sections have primarily discussed the media conditions and literary responses of the twenty-first century: algorithmic analysis, digital information storage, recent novelistic experimentation, and so forth. In this volume, I put these twenty-first-century literary and informational landscapes into dialogue with those of the modernist era. Contemporary information mediation did not spring fully formed from digital media. It represents one stage in a history of ideas about how information should be managed and what role the book should play in that management. The excavation of digital media’s predigital roots is a widespread practice within media studies, exemplified by work such as Sterne’s and Lisa Gitelman’s, and many media theorists have identified aspects of pre-1900 media culture that anticipate digital culture. I focus on the modernist period because the first decades of the twentieth century ushered in information management techniques, and novelistic responses to the assumptions entailed by those techniques, that remain active a century later. Information mediation and the future of the book were central issues for both modernism and modernity.

The information culture and media-centered literary experiments of the modernist period share commonalities with those of the twenty-first century that make their comparison especially fruitful. That media were essential to modernism is by now well established. Scholars in modernist studies have examined everything from telegraphs and phonographs to corridors and vulcanized rubber under the rubric of media.63 Media theorists, too, have cited the importance of the decades circa 1900, building on Kittler’s description of the epistemic break ushered in with the discourse network of 1900.64 (That this media transition was intertwined with modernist literature is evident in Kittler’s inclusion of Gertrude Stein as an emblematic figure of that discourse network.) The new modernist studies have asserted that aspects of modernity and modernism extended beyond the temporal and geographical boundaries traditionally associated with these terms.65 Although I do not claim the contemporary period as precisely modernist, my research contributes to the vertical expansion of modernist studies in its discussion of how and why elements of modernism’s information culture have been remobilized a century later.

Additionally, the idea that the links between the early twentieth and twenty-first centuries are worthy of critical attention has been gaining momentum. Manovich, for instance, describes new media’s formal and logical indebtedness to the avant-garde cinema of the 1920s. Alan Liu cites Frederick Winslow Taylor’s scientific management as an origin point for the structured and encoded discourse that Liu argues is integral to “discourse network 2000.” Paul Stephens’s history of poetic interventions into the cultural conditions of information overload moves from Stein to contemporary conceptual writing. Jessica Pressman has made the most sustained analysis of the connections between these periods: she argues that the modernist imperative to “make it new” was influenced by media and that it provided a model of experimentation taken up a century later by writers of electronic literature.66 I uncover further significant connections by demonstrating how, in both eras, concerns about the scale of information, and consequent uncertainty about the print book’s future, have been met by renewed novelistic interest in textual materiality as a way of rethinking emerging practices of mediation.

Key features of information proliferation and management that have reached their apotheosis in the digital age originated in the early twentieth century. As chapter 1 describes in detail, while anxiety about information overload predates the early twentieth century, what was new were the forms information took as it proliferated across new media and as it was managed with new techniques. The volume of printed matter exploded. Book production quadrupled in the United Kingdom between the 1840s and 1916. In the United States, the number of new titles being published in 1910 was six times greater than it had been three decades earlier.67 The claim that “readers [were being] overwhelmed with the avalanche of books,” as the Athenaeum declared in 1912, became a common enough trope for a columnist in the Literary Digest to wearily write in 1923, “There is ‘an avalanche of books,’ of course.” 68 Similar complaints surfaced about other media, as periodical circulation increased dramatically, alongside advertisements, documents, and files. Benjamin worried that children were so besieged by “locust swarms of print” from newspapers, advertisements, and other texts that they lacked the ability to focus on a print book: “Before a child of our time finds his way clear to opening a book, his eyes have been exposed to such a blizzard of changing, colorful, conflicting letters that the chances of his penetrating the archaic stillness of the book are slight.” 69 Concerns about information proliferation, and the perception that its scale was historically unprecedented, led to the development of systematic management. The growing cadre of information professionals devised standardized processes for storing and accessing information, positioning organizational systems as interfaces that mediated between users and data. Some information professionals also advocated for the use of microform, which necessitated magnification screens or other reading interfaces. As information became more heavily mediated in these ways, the print book seemed a less efficient storage medium by comparison. Ideologically if not technologically, the information culture of the early twentieth century set into motion practices and assumptions about information and its mediation that continue to operate in the age of Big Data.

Wrestling with the question of whether the novel could meaningfully act as an archive for societies whose informational output had reached watershed levels, novelists looked to the book as a microcosm of information management. Most literary criticism treats metamedial experimentation as a product of the digital era. Hayles and others have noted that this experimentation has intensified in the contemporary novel partly in reaction to the perceived obsolescence of the book in the age of digital information and e-readers and partly because its creation is enabled by the use of digital platforms that allow novelists to more easily design visually complex works.70 Roughly a century ago, however, metamedial experimentation in the novel also emerged as a trend, in a similar effort to reevaluate the genre’s relevance. Marshall has documented the novel’s awareness of its function as a medium during this period: arguing that novels like Manhattan Transfer and Native Son figure corridors as a metaphor for the novel’s mediating role in the representation of modern interiority, she studies how they represent technologies of “corridoricity” such as “ducts, pipes, [and] urban infrastructures” as a way to “refer explicitly to the inscription technologies of the book” and to “thematiz[e] the network of relations between the material printed artifact and its fictionality.” 71 Marshall’s attention to the book is unusual: while modernist studies has benefited from a wealth of critical attention to print culture, the book has tended to be eclipsed by newspapers, little magazines, and other periodicals. The major exception is Lurz’s The Death of the Book, which argues for the importance of the book as an object to the modernist novel. As modernist studies begins to contend with the embeddedness of novels in books, I want to suggest that metamedial novels’ emphasis on this embeddedness resulted from a merger of modernist writers’ preoccupation with form and their awareness of the shift from books to modern information media.

The modernist period was a time when visual artists turned their attention to their media and when poets felt, to quote William Carlos Williams, that “all the world was going crazy about typographical form.” 72 We should add the book to the list of media on which modernist writers focused their formal investigations. Literary experimentation with the visual and tactile dimensions of media during the modernist era tends to be discussed with reference to poetry—the disorienting typographical layouts of the Vorticists, for instance. But novelists, too, played with the printed page. Ulysses disrupts conventional textual arrangement with devices such as question-and-answer sets and mock newspaper headlines. William Faulkner attempted to publish The Sound and the Fury using several colors of ink to mark narrative temporalities. Photographs became more commonplace; we see them in novels such as André Breton’s Nadja and Woolf’s Orlando. Novelists were also more involved in the production of their novels as physical artifacts. Like the artists’ books that also commenced as a practice in the early twentieth century, these novels are “self-referential and self-aware objects” that “interrogate the codex, calling into question how books communicate and how we read, using every aspect of their structure, form, and content to make meaning.” 73 This novelistic experimentation was not limited to exploring form for its own sake; it also reimagined and revised the relationship between the novel’s representative abilities and its medium. As writers reflected on the novel’s place in a society dominated by information, their works showed the novel’s representation of information culture to be predicated on the book’s mediation of information.

From the Modernist Era to the Twenty-First Century: Information as the Formation of Data

I put the modernist and contemporary periods into dialogue to draw out the rich connections between them. I also do so in order to think broadly about the comparative work of media history. To establish my reading of the early twentieth and twenty-first centuries as parallel moments of media transition, the next four chapters alternate between these eras, with the final chapter serving as a synthesis. A productive alchemy arises from the juxtaposition. This parallax reading brings to light how modernist ideas about literary form chime with ideas about form from information culture, and it clarifies how the novel’s archival project has been reimagined through the ways in which books and other information media have been theorized in critical and popular accounts. When we view the contemporary moment as continuing the legacy of modernism’s engagement with information, mediation, and form, we better understand our own information culture. Out of Print models how we may take into account the way in which one era’s media “anticipates” another’s (the word is frequently used to locate similarities), without either reinstating a teleological view of history or simply describing the two as uncannily alike without delineating any trajectory between them. It is not quite that modernist information culture gave rise directly to today’s information culture, nor is it quite as fuzzy as the statement that the one merely anticipated the other. To identify continuities as well as divergences between these periods is to formulate a more robust theorization of novelists’ responses to the scale of information, tracking the persistence of forms across time.

The chapters that follow consider several related but distinct aspects of form central to information, the novel, and the book: form as organization; form as embodiment; and form’s role in innovation, preservation, and obsolescence. Taken together, these chapters develop a concept I call the formation of data. As I have stated, I define information as data made meaningful by the ways in which it is contextualized and situated. Information is constituted by the forms it takes. The formation of data encompasses how data are organized when they are collected and stored, how they are embodied in media, and how these ways of giving form to data create meaning. Metamedial novels highlight all these facets of the formation of data.

The first two chapters focus on information’s arrangement—its organizational form. Chapter 1 describes information shock, modernism’s version of information overload, a condition produced by textual proliferation too vast and disorganized to manage. I argue that modernist novelists contributed to debates about information shock by representing the cognitive experiences it created and merging an informational aesthetic with a renewed attention to the form of the book. Focusing on Ulysses, I argue that Joyce highlights the novel’s textual materiality in order to critique mediation, which I show to be central to the emerging field of information management. I take Paul Otlet’s Mundaneum archive as a paradigmatic case study, documenting how the systematic management of information, as it moved away from the book to meet new economies of information scale, entailed methods of organization, processing, and navigation whose details could be hidden from lay users. Against such mediation, Ulysses demonstrates how the metamedial novel could reestablish form’s centrality to information.

Chapter 2 moves from early twentieth-century information management’s use of classification systems to the algorithmic analysis and search systems that dominate mediation in the era of Big Data. In my reading, contemporary novelists’ response has been not to represent this scale in and of itself but to explore how data are made meaningful through the limits that contain them. I show how, in texts including Danielewski’s Only Revolutions and Matthew McIntosh’s theMystery.doc, narrative, book, and page all become systems that impose a comprehensible order on information. These works challenge conventional definitions of the literary by producing literary text through the distillation of large quantities of information. The value of the book as a novelistic medium, these texts suggest, is its ability to make conceptual and spatial limits visible as frameworks, as the novel itself becomes a meaningful framework for contextualizing information.

The second set of chapters examines the implications of information’s embodiment in media—its material form. Chapter 3 argues that the book’s tactility and consequent ability to preserve the physical traces of readers took on a new significance during the modernist period. This emphasis on what I call haptic storage occurred in response to new regimes of personal information collection that sought to quantify people via data and to the rise of microform as a competitor to the print book for information storage. In this pivotal moment, the use of microform reading interfaces encouraged the rhetorical stance that medial embodiment was inconsequential to information, a principle epitomized by Bob Brown’s microfilm-inspired Readies project. I read Woolf’s Orlando as a critique of this disregard of embodiment and material form. Orlando problematizes the idea that any quantity of information could substitute for the presence of a living subject. Instead, it suggests that Woolf’s project of capturing the life of her estranged lover becomes realized only when narrative descriptions and the accumulation of biographical detail are combined with the haptic “information” created when readers interact with books.

Chapter 4 analyzes how novelists have used the book’s hypermediacy and affordance of haptic storage to challenge the posthumanist principle that a sufficiently large amount of digital data can represent or replace a living subject. I examine this principle in the context of several digital immortality projects, arguing that the shift from print books to digital media has influenced the metaphors and ideologies of digital immortality, in addition to making this research possible. In response to the hope that people could live forever as informational representations, transcending their mortal human bodies, I study several works of contemporary literature that dramatize the potential horror of such existence. I analyze two novels and a memoir-elegy—Steven Hall’s The Raw Shark Texts, Foer’s Extremely Loud and Incredibly Close, and Anne Carson’s Nox, respectively—arguing that they demonstrate how an emphasis on the book’s materiality can be useful to the work of mourning. My analysis of these works clarifies how books and digital media have inspired distinct rhetorical positions regarding representation and death because of how they mediate information.

Chapter 5 synthesizes the preceding chapters, turning from the early twentieth and twenty-first centuries to the future. I examine the close links between discourse on the death of the novel and discourse on the death of the book, particularly as both have been influenced by media transition and issues of platform stability. I argue that the book, in its fragility and its persistence, has served as a model for thinking through futurity. I read the novel as a literary form especially suited to theorizing media transition, and I contrast the celebration of innovation in the technology industry with skepticism about innovation with literary form. The chapter analyzes three visions of how the relationships among print books, digital media, and literary reading may manifest in the future: Katie Paterson’s Future Library Project, Stephen King’s Kindle novella UR, and Sebastian Schmieg and Silvio Lorusso’s artist’s book 56 Broken Kindle Screens. I conclude by discussing the ongoing relevance of modernist literature and modernist studies for thinking about media, temporality, and futurity.

The novel has been transformed over the last century, and this transformation has intensified with the growing scale of information. Novelists have explored the porous boundaries between what constitutes literary narrative and what constitutes information, and they have leveraged the book’s hypermediacy to incorporate the aesthetics of information and critique the assumptions that underlie mediation. Novelists’ experiments with the book and literary form reassert form’s importance to information, even if the dominant conditions of contemporary information culture work against a conscious awareness of, or interaction with, these forms. To construct a medium-specific approach to the novel is thus not only to consider how the print book has impacted the novel but also to consider how the novel circulates within, and interacts with, its media ecology. The print novel is not a nostalgic form. It continues to shape popular understandings of information even as it adapts to engage with new media, new practices of mediating information, and new ideas about the book.