• Notes
• Chapter
• University of Michigan Press
• pp. 201-208
• View Citation
• Additional Information

. In or around March 1905, the literary representation of lives changed. Or at least, that’s when Edwin Slosson, chemist and literary editor of progressive New York weekly newspaper The Independent, noticed it.1 His description of and explanation for this change plots a complex interaction between readerly desire and a cultural surround of pervasive empiricism. Readers no longer seek the “generalized types of humanity as expressed by the artist in painting and sculpture,” Slosson argues, because “romances and poems do not interest them so much as do individuals.” Realistic novels are, therefore, “becoming more nearly a transcript of life” (849). The selecting and synthesizing acts of literary narration are seen as less valuable, less real and, perhaps surprisingly, less interesting to readers than the more mechanical Page 2 →approach of unselectively recording everything that happens within the scope of a life. The preference for and epistemological value attributed to the rigorous recording of observation is also on the ascendant in scientific practice, Slosson continues to point out, which seems increasingly focused on collecting observations for future analysis rather than the proposal of overarching theories. Now, the “candidate for the Ph.D. watches a single amoeba under a microscope and writes his thesis on one day’s doings of its somewhat monotonous life” (Slosson, 849). In literature, as in science, Slosson suggests, the representation of reality is taking on a distinct new form, composed of discrete observations of actual lives, exhaustively collected rather than interpretively selected. In other words, lives are coming to be seen as data collections.

. Provoking critical awareness of the gap between data and narrative is a project that has only become more urgent a century or so after these modernist experiments. Data is9 now constantly being used to narrate our lives. Categorizing algorithms draw from amassed personal data to assign narrative destinies to individuals at crucial junctures, simultaneously predicting and shaping the path of their lives’ unfolding. Data is commonly assumed to bring us closer to objective knowledge of reality, perhaps relieving the need for human interpretation altogether if we can only collect enough. In 2008, Chris Anderson, a writer for Wired magazine, went so far as to assert, “With enough data, the numbers speak for themselves” (108). But the narrative paths categorizing algorithms have assigned to human lives in areas ranging from public policy to personal finance seem, more often than not, to replicate biases about who an individual is and could become.10 It has become increasingly clear that, as John Cheney-Lippold cautions, “Who we are in terms of data depends on how our data is spoken for” (We Are Data, 48). The question of who we are hinges not on what data says, but on who is granted the authority to speak for it and what forms of representation they employ. The data-driven modernists that populate the juncture between life, data, and story model critical data narrative in ways that prepare us to challenge the claims of algorithmic identification and demand alternative modes of narrating our collected lives.

. Bacon’s method thus comes with two mandates: data collection must Page 10 →be exhaustive and it must be exteriorized. Exhaustivity, as conceived by information science, is a measure of the correspondence of an index to the document(s) indexed, a measure of the distance between the secondary model and the complete world of knowledge, however that world is defined (Van Rijsbergen, 24–25). As Mary Poovey describes, the Baconian ideal insists that the “entire globe and all of its inhabitants ought to be subject to empirical observation” (“Limits,” 193). The aspiration to exhaustive collection is essential to data epistemology for two reasons. First, it is only through exhaustive collection that we can be sure to overcome our selective, interpretive nature. Second, exhaustive data is presumed necessary to maximize the material gain that can be leveraged out of an increased ability to predict and control.11

. Laplace’s conjecture assumes that there are a set of natural laws that, working together in complex yet predictable ways, are awaiting only our discovery to reveal seeming chance for the order that it really is. This revealed order will explain reality from the micro to the macro, subsuming “all events, even those which on account of their insignificance do not seem to follow the great laws of nature” (Laplace, 3) in its predictive power. It is only “in ignorance of the ties which unite such events to the entire system of the universe” that “they have been made to depend on final causes or on hazard” (3). Both a priori “final causes” and chance will be abolished by exhaustive data collection. Wendy Chun has named this figure of “an intelligence sufficiently vast to submit this data to analysis,” the “Laplacian subject . . . a sovereign subject capable of ‘knowing all’” (109). The imagination of such a subject entails determinist assumptions about the workings of both animate and inanimate life and nearly unimaginable power for an individual or group that could wield such definitive knowledge.12

. Data epistemology taken to the extreme is what boyd and Crawford have called data mythology, “the widespread belief that large data sets offer a higher form of intelligence and knowledge that can generate insights that were previously impossible, with the aura of truth, objectivity, and accuracy” (663). Data mythology assumes that exhaustive data collection is functionally possible and has the potential to produce atheoretical understanding and universal truth. Because of this assumption, “the apparent empiricism of data-driven research” (Schöch) is often popularly seen to be epistemologically superior to methods, such as those typically employed by humanities scholars, that insist on “context-dependent interpretation and the inevitable ‘situated-ness’ of the researchers and their aims.” Raw data, though an oxymoron,13 is often presented as the closest thing to a complete proxy of reality as we imagine can be objectively attained. An epistemology of data that anticipates seamless translation of data to meaning, though, is fundamentally challenged by data’s core formal features.

. Data is simultaneously point and collection: discrete notations of perception14 collected in order to exhaustively represent reality. Literally, “data” is both a count noun, “an item of information; a datum” and a mass noun, the plural of datum, “related items of (chiefly numerical) information considered collectively, typically obtained by scientific work and used for reference, analysis, or calculation” (“data,” Oxford English Dictionary). Data becomes legible as data through the specification of a method of collection. As sociologist Roberto Franzosi explains, “Typically it is a specific methodological school that confers the status of datum to specific types of evidence. Data, in other words, are the result of specific types of data collectionPage 14 → techniques” (186). As these definitions highlight, data is functionally and formally inseparable from collection. Data’s grammatical plurality is its implicit collectivity. The colloquial slippage from plural to singular is a telling elision of data’s conceptual complexity and an expression of the pervasive belief (or desire) that data, all by itself, means and means definitively.

. Once put into a collection, each datum holds an equal status, an equal claim to representing a small piece of actually existing reality. Despite the formal effect of commensuration, though, these equally real small pieces are predicated by and represent difference itself. As philosopher of information Luciano Floridi notes, “a datum is ultimately reducible to a lack of uniformity” (Introduction, 23, emphasis in original).15 There would be no data to perceive, record, or represent if difference did not define the “real world” (Floridi, Introduction, 23). Data collections are inevitably plural, heterogenous, and full of conflicting potential information. Without the interventions of probability and statistics, these collections would yield very little in the way of clarity, prediction, or narrative. So, while the popular tendency is to think of data as a solution to uncertainty and ignorance, its proliferating points, when taken seriously as equally important markers of actually existing reality, agitate against stable knowledge, blanket description, and clear trajectories of development.

. Separating a concept of data from the related concepts of information, statistics, and the database is crucial to understanding how it operates as a distinct epistemological concept and representational form. Data does not neatly align with information, which can be roughly defined as data plus meaning (Floridi, Introduction, 21). Information, unlike data, is meant to convey a specific message. Claude Shannon, a foundational theorist of information, proposes that information can be understood as fundamentally made up of two parts: signal and noise.16 Signal is the information intended to be conveyed, and noise is what adds superfluous and misleading information to the signal. It could be said that those who seek, and believe they find, a narrative in collected data make the same distinction. Yet, the distinction between signal and noise is a subjective intervention, an act of selection that overrides the formal equality of the data point to facilitate interpretation. The difference between a data collection and a message (or signal) is that in a data collection, every point is, at some level, a signal as valid as any other, a bearer of potential information about actual reality.

. Most importantly for the goals of this project, data is the formal antithesis of algorithmic determination. “Algorithm” is a term that has taken on many meanings.17 At its most basic level, an algorithm is a series of steps Page 16 →undertaken to perform some operation. Present usage typically assumes that these are analytical steps performed computationally on data to arrive at some kind of predictive determination. Many algorithmic determinations are aimed at identifying people for some purpose—as a good credit risk, as a likely criminal, as a future consumer of a particular product. Such categorizing algorithms are what seem to make the massive amounts of data collected in the course of our daily lives speak for itself. While algorithmic assessments of identity come with probabilistic disclaimers and variable accuracy rates, the consequences they entail for the lives to which they are applied unfold on an all-or-nothing basis. If you are determined to be a likely criminal, you will be treated like one. Recovering data as a representational form prior to and other than the algorithm can help us cultivate a needed skepticism about the authority of mechanical determinations.

. By considering the data point and data collection alongside and among the roster of modernist aesthetic forms, we recover an archive of resistance to the conflation of data and narrative in the context of lives. The period of this study marks the convergence of data as an epistemological concept with the rise of modernist aesthetic projects founded in various experiences of modernity. A full text search for the keyword “data” in the Modernist Journals Project returns 538 hits,18 in both literarily oriented and more general interest publications, and in editorials, essays, reviews, and advertisements. A sampling of these usages conveys a sense of the period’s data discourse. Most substantively, forty-three of these hits are from The Crisis, founded and edited by Du Bois, dedicated from its opening issue to “to set forth those facts and arguments which show the danger of race prejudice, particularly as manifested to-day toward colored people” (“Editorial,” 10). The presentation of data, and discussions of the need for data on topics relevant to African American life, are a central feature of this journal. Roughly another dozen hits are from The Little Review, and roughly half of these usages come from the pen of Ezra Pound,19 some deployed in editorials printed alongside the serialized books of Joyce’s Ulysses. Although these hits include one poem,20 usages in Little Review tend to refer to data in a fairly typical manner, as something gathered and consulted by experts. However, these more or less pedestrian usages sometimes take place in the context of fairly contentious debates over ideas that ostensibly arise from data, such as an editorial on recent scientific investigations into gender.21 Finally, advertisements position data as a product feature that will advantage savvy users. Harpers of January 1911, for example, includes an advertisement for the Mahin Advertising Company’s Data Book, which claims to apply data’s Page 17 →powers of prediction to supply ten tests a salesperson might use to judge written sales copy “so that you can pretty nearly tell a selling advertisement before you must pay for it” (“Salesman”). Aside from Du Bois and the Crisis, most usages of the term do not imply the type of critical attention to data itself that I argue characterizes the life writing of data-driven modernists. They do, however, demonstrate that data was in cultural circulation as a concept, imagined as collections of facts being accumulated by scientific observers out there, somewhere, slowly revealing the answers to certain questions.

. By considering the data point and data collection alongside and among the roster of modernist aesthetic forms, we recover an archive of resistance to the conflation of data and narrative in the context of lives. The period of this study marks the convergence of data as an epistemological concept with the rise of modernist aesthetic projects founded in various experiences of modernity. A full text search for the keyword “data” in the Modernist Journals Project returns 538 hits,18 in both literarily oriented and more general interest publications, and in editorials, essays, reviews, and advertisements. A sampling of these usages conveys a sense of the period’s data discourse. Most substantively, forty-three of these hits are from The Crisis, founded and edited by Du Bois, dedicated from its opening issue to “to set forth those facts and arguments which show the danger of race prejudice, particularly as manifested to-day toward colored people” (“Editorial,” 10). The presentation of data, and discussions of the need for data on topics relevant to African American life, are a central feature of this journal. Roughly another dozen hits are from The Little Review, and roughly half of these usages come from the pen of Ezra Pound,19 some deployed in editorials printed alongside the serialized books of Joyce’s Ulysses. Although these hits include one poem,20 usages in Little Review tend to refer to data in a fairly typical manner, as something gathered and consulted by experts. However, these more or less pedestrian usages sometimes take place in the context of fairly contentious debates over ideas that ostensibly arise from data, such as an editorial on recent scientific investigations into gender.21 Finally, advertisements position data as a product feature that will advantage savvy users. Harpers of January 1911, for example, includes an advertisement for the Mahin Advertising Company’s Data Book, which claims to apply data’s Page 17 →powers of prediction to supply ten tests a salesperson might use to judge written sales copy “so that you can pretty nearly tell a selling advertisement before you must pay for it” (“Salesman”). Aside from Du Bois and the Crisis, most usages of the term do not imply the type of critical attention to data itself that I argue characterizes the life writing of data-driven modernists. They do, however, demonstrate that data was in cultural circulation as a concept, imagined as collections of facts being accumulated by scientific observers out there, somewhere, slowly revealing the answers to certain questions.

. Collecting Lives examines U.S. modernist life writing forms as sites of critical engagement with the data episteme, a cultural surround in which data and its collection are presumed to offer unprecedented access to reality, truth, and power. Data collection, as fundamental to both empiricism and imperialism,2 is one of modernity’s foundational representational forms. Further, it is a form that has an especially prominent role in U.S. attempts to represent and control the minoritized and economically precarious social groups created by modernity. The application of data collection to human selfhood3 in the late nineteenth- and early twentieth-century United States provides an instructive prehistory to the underlying question of the relationship between data, life, and narrative that animates contemporary debate over algorithmic modes of identification. This historical frame centers U.S. modernist life writers W. E. B. Du Bois, Henry Adams, Gertrude Stein, and Ida B. Wells-Barnett. Working in human-oriented empiricist disciplines in the midst of methodological reconfiguration around the work of data collection, these writers experiment with data as a form of representing lives. Using a comparative approach that situates works from a range of modernist canons in the data episteme, I seek to recognize the long conceptual history of data as a technology of selfhood and its intersection with material histories of power—who has the ability to collect data about whom and whose narration of that data is seen as authoritative. I argue that these writers draw from their work in sociology, history, psychology, and journalism to formulate critical data aesthetics as they confront questions of identity around race, gender, and nation both in their research and their life writing.

. By considering the data point and data collection alongside and among the roster of modernist aesthetic forms, we recover an archive of resistance to the conflation of data and narrative in the context of lives. The period of this study marks the convergence of data as an epistemological concept with the rise of modernist aesthetic projects founded in various experiences of modernity. A full text search for the keyword “data” in the Modernist Journals Project returns 538 hits,18 in both literarily oriented and more general interest publications, and in editorials, essays, reviews, and advertisements. A sampling of these usages conveys a sense of the period’s data discourse. Most substantively, forty-three of these hits are from The Crisis, founded and edited by Du Bois, dedicated from its opening issue to “to set forth those facts and arguments which show the danger of race prejudice, particularly as manifested to-day toward colored people” (“Editorial,” 10). The presentation of data, and discussions of the need for data on topics relevant to African American life, are a central feature of this journal. Roughly another dozen hits are from The Little Review, and roughly half of these usages come from the pen of Ezra Pound,19 some deployed in editorials printed alongside the serialized books of Joyce’s Ulysses. Although these hits include one poem,20 usages in Little Review tend to refer to data in a fairly typical manner, as something gathered and consulted by experts. However, these more or less pedestrian usages sometimes take place in the context of fairly contentious debates over ideas that ostensibly arise from data, such as an editorial on recent scientific investigations into gender.21 Finally, advertisements position data as a product feature that will advantage savvy users. Harpers of January 1911, for example, includes an advertisement for the Mahin Advertising Company’s Data Book, which claims to apply data’s Page 17 →powers of prediction to supply ten tests a salesperson might use to judge written sales copy “so that you can pretty nearly tell a selling advertisement before you must pay for it” (“Salesman”). Aside from Du Bois and the Crisis, most usages of the term do not imply the type of critical attention to data itself that I argue characterizes the life writing of data-driven modernists. They do, however, demonstrate that data was in cultural circulation as a concept, imagined as collections of facts being accumulated by scientific observers out there, somewhere, slowly revealing the answers to certain questions.

. By considering the data point and data collection alongside and among the roster of modernist aesthetic forms, we recover an archive of resistance to the conflation of data and narrative in the context of lives. The period of this study marks the convergence of data as an epistemological concept with the rise of modernist aesthetic projects founded in various experiences of modernity. A full text search for the keyword “data” in the Modernist Journals Project returns 538 hits,18 in both literarily oriented and more general interest publications, and in editorials, essays, reviews, and advertisements. A sampling of these usages conveys a sense of the period’s data discourse. Most substantively, forty-three of these hits are from The Crisis, founded and edited by Du Bois, dedicated from its opening issue to “to set forth those facts and arguments which show the danger of race prejudice, particularly as manifested to-day toward colored people” (“Editorial,” 10). The presentation of data, and discussions of the need for data on topics relevant to African American life, are a central feature of this journal. Roughly another dozen hits are from The Little Review, and roughly half of these usages come from the pen of Ezra Pound,19 some deployed in editorials printed alongside the serialized books of Joyce’s Ulysses. Although these hits include one poem,20 usages in Little Review tend to refer to data in a fairly typical manner, as something gathered and consulted by experts. However, these more or less pedestrian usages sometimes take place in the context of fairly contentious debates over ideas that ostensibly arise from data, such as an editorial on recent scientific investigations into gender.21 Finally, advertisements position data as a product feature that will advantage savvy users. Harpers of January 1911, for example, includes an advertisement for the Mahin Advertising Company’s Data Book, which claims to apply data’s Page 17 →powers of prediction to supply ten tests a salesperson might use to judge written sales copy “so that you can pretty nearly tell a selling advertisement before you must pay for it” (“Salesman”). Aside from Du Bois and the Crisis, most usages of the term do not imply the type of critical attention to data itself that I argue characterizes the life writing of data-driven modernists. They do, however, demonstrate that data was in cultural circulation as a concept, imagined as collections of facts being accumulated by scientific observers out there, somewhere, slowly revealing the answers to certain questions.

. Montage, or the juxtaposition of disparate images, texts, and other representational fragments, is in many regards similar to the mode of assemblagePage 18 → that I argue characterizes narratives constructed in relationship to data collections. The practitioner of montage, like the data-driven modernist, seeks to call attention to the fact that observation, whether in the form of a single image or collection of data points, does not speak for itself but requires interpretive intervention. As Sally Stein has said of montage in U.S. photography, “The principle of active mediation underlying photomontage casts doubt on the adequacy of the autonomous photograph by suggesting that meaning required more than the selection of subject matter in the viewfinder” (132–33). Yet the practice of montage is fundamentally different than data collection, because its collecting aesthetics are more directed, toward a predetermined disruptive effect22 or the telling of a particular story,23 rather than primarily the result of a commitment to gather exhaustively within a set of chronological or physical parameters.

. Montage, or the juxtaposition of disparate images, texts, and other representational fragments, is in many regards similar to the mode of assemblagePage 18 → that I argue characterizes narratives constructed in relationship to data collections. The practitioner of montage, like the data-driven modernist, seeks to call attention to the fact that observation, whether in the form of a single image or collection of data points, does not speak for itself but requires interpretive intervention. As Sally Stein has said of montage in U.S. photography, “The principle of active mediation underlying photomontage casts doubt on the adequacy of the autonomous photograph by suggesting that meaning required more than the selection of subject matter in the viewfinder” (132–33). Yet the practice of montage is fundamentally different than data collection, because its collecting aesthetics are more directed, toward a predetermined disruptive effect22 or the telling of a particular story,23 rather than primarily the result of a commitment to gather exhaustively within a set of chronological or physical parameters.

. Like montage and the impression, data as a modernist form crosses the boundary of textual and visual media. Because my focus is primarily textual, I have limited my analysis to the ways in which these writings analogize important aspects of visualization, such as the effect of parallelism generated by parataxis and descriptive accumulation, but I do not deal at length with visualizations themselves. There are undoubtedly more provocative connections between modernist visual arts and data visualization than this study focused on life writing can encompass.24

. Many of the formal features I claim for the data aesthetic—such as the abundance of detailed description, repetition, lack of narrative structure, a focus on social others—have also been attributed to naturalism, realism, and postmodernism.25 Given that these, too, are major aesthetic movementsPage 19 → of the data-oriented nineteenth and twentieth centuries, some level of response to and engagement with the concept of data is to be expected across aesthetic movements and some formal overlap as well. I see data collection as one of the methods and forms relevant to David Harvey’s assessment that “modernism, in short, took on multiple perspectivism and relativism as its epistemology for revealing what it still undertook to be the true nature of a unified, though complex, underlying reality” (27). What I think differentiates the writers and texts of this study is their explicit engagement with data epistemology as a means of seeking that reality. Data-driven modernists take the thought experiment of data collection further than others as they both question and embrace data’s potential for revelation. They are obdurately receptive and committed to continuing collection, which makes them willing to encounter and proclaim the deep strangeness that a data collection approaching exhaustivity reveals as real. The type of modernism I examine arrives at its break with narrative convention through a critical yet committed search for what is real via data collection.

. As Laplace articulated and contemporary claims on behalf of data often imply,26 the envisioned ends of data collection are that an underlying narrative will be revealed, allowing knowledge and prediction to coincide. Yet, data’s formal and epistemological dependence on the form of collection challenges narrativity rather than confirming it. Data begins as a data point—whether that point is a measurement, description, test result, case study, or life story—but the data point is understood never to stand alone but always to be part of a heterogeneous collection of formally equal points. Prior to analysis, data points exist in a flat order, in parallel, each one as real as the next. These collected points hold sway as more real than any interpretation that arises from them because they are understood as further removed from human intervention, our problematic tendency to rush them into sense-making frameworks. Implicit in the goal of exhaustive representation is the belief that each existing “particular” bears some amount of invaluable potential information. To overlook anything, or to assume one has found the paradigmatic case that will explain all others, is to commit the error of mental projection. The data collector must be willing to record and represent all that she finds. Exhaustivity therefore requires, at least in theory, a subjective shift toward radical receptivity to reality as it is encountered. Bacon chides the squeamish or unduly proper empiricist, “With regard to the meanness, or even the filthiness of particulars, for which (as Pliny observes), an apology is requisite, such subjects are no less worthy of admission into natural history than the most magnificent and costly. . . . For that which is deserving of existence is deserving of knowledge, the image of existence” (95). If we desire knowledge of the real, preexisting assumptions about worthy objects of knowledge cannot circumscribe the collection of data. All that exists must be observed, recorded, and represented.

. Narrative therefore looks different through the lens of data. It looks more contingent, more provisional, and less like what we would usually Page 22 →recognize as narrative because the epistemological commitment to collection complicates the process of selection, or the exclusion of points deemed insignificant in light of a predetermined ending. Instead, through its conceptual privileging of collection over selection, data generates aesthetic forms that emphasize parallelism and co-presence of heterogenous elements. Timothy Lenoir offers the following definition of parallelism in formal terms by way of contrasting it with its conceptual foil, seriality: “Seriality is exemplified in narratives, routines, algorithms, melodies, timelines; parallelism is exemplified in scenes, episodes, harmonies, contexts, atmospheres, and images. Parallelism foregrounds presence, simultaneity, co-occurrence” (xxvi–vii). The parallel/serial duo underlies every representational form, but one or the other dynamic can predominate. Data collection as a form emphasizes parallelism. Each data point exists in a formally parallel state, representing an actually existing reality and exerting the conceptual force of equal importance and potential meaning. Beeswarm plots and dot maps, forms of data visualization in which every data point is visible as a discrete presence, emphasize this formal feature inherent to data.27 As Manovich observes of new media objects built on database structures, many “do not tell stories; they do not have a beginning or end; in fact, they do not have any development. . . . Instead, they are collections of individual items” (218).28 The conceptual force of commensuration and co-presence inherent to data collection agitates against the finality of any act of selection, complicating the methodological procedures through which we move from data to meaning.

. Narrative therefore looks different through the lens of data. It looks more contingent, more provisional, and less like what we would usually Page 22 →recognize as narrative because the epistemological commitment to collection complicates the process of selection, or the exclusion of points deemed insignificant in light of a predetermined ending. Instead, through its conceptual privileging of collection over selection, data generates aesthetic forms that emphasize parallelism and co-presence of heterogenous elements. Timothy Lenoir offers the following definition of parallelism in formal terms by way of contrasting it with its conceptual foil, seriality: “Seriality is exemplified in narratives, routines, algorithms, melodies, timelines; parallelism is exemplified in scenes, episodes, harmonies, contexts, atmospheres, and images. Parallelism foregrounds presence, simultaneity, co-occurrence” (xxvi–vii). The parallel/serial duo underlies every representational form, but one or the other dynamic can predominate. Data collection as a form emphasizes parallelism. Each data point exists in a formally parallel state, representing an actually existing reality and exerting the conceptual force of equal importance and potential meaning. Beeswarm plots and dot maps, forms of data visualization in which every data point is visible as a discrete presence, emphasize this formal feature inherent to data.27 As Manovich observes of new media objects built on database structures, many “do not tell stories; they do not have a beginning or end; in fact, they do not have any development. . . . Instead, they are collections of individual items” (218).28 The conceptual force of commensuration and co-presence inherent to data collection agitates against the finality of any act of selection, complicating the methodological procedures through which we move from data to meaning.

. Chapter one, “‘Such A Body of Information’: W. E. B. Du Bois, Data, and the Re-assemblage of Race and Self,” demonstrates how Du Bois uses data collection—as a conceptual framework for empirical reality, as a method of sociological inquiry, and as a representational form—to intervene in fixed narratives of African American life and selfhood. I locate his methodological innovations in sociology in The Philadelphia Negro (1899) as arising from an insistence on more exhaustive practices of data collection. I trace how his embodied collection of and interaction with data leads to ways of seeing the self and the world that surface and develop in his multi-formal autobiographical works. Specifically, the data of The Philadelphia Negro creates a parallel aesthetic of inclusion that surfaces complexity where a deadly imposed coherence threatens to stifle the narrative trajectories of African American lives. The aesthetic of complexity continues to inform The Souls of Black Folk’s construction of Black collectivity through a multiplicity of rhetorical and formal modes, displacing any single historical narrative, social trajectory, or empirical assessment of African American life.29 The epistemology of data collection then drives the crowd-sourced politics of Darkwater, which translates the ideal of exhaustive representation into an aspirational democratic imaginary. Finally, Dusk of Dawn’s recollection of a life ricocheted between disciplinary, geographical, and social spaces creates an assemblage-driven Black self that is not essentially other, but experientially othered by repeated encounter, both mediated and direct, with the physical and psychological violence of racialization. Taken together, these texts also constitute a persistent critique of empiricism as a tool for progressive social change, both for what it cannot represent and what it cannot do.

. Collecting Lives examines U.S. modernist life writing forms as sites of critical engagement with the data episteme, a cultural surround in which data and its collection are presumed to offer unprecedented access to reality, truth, and power. Data collection, as fundamental to both empiricism and imperialism,2 is one of modernity’s foundational representational forms. Further, it is a form that has an especially prominent role in U.S. attempts to represent and control the minoritized and economically precarious social groups created by modernity. The application of data collection to human selfhood3 in the late nineteenth- and early twentieth-century United States provides an instructive prehistory to the underlying question of the relationship between data, life, and narrative that animates contemporary debate over algorithmic modes of identification. This historical frame centers U.S. modernist life writers W. E. B. Du Bois, Henry Adams, Gertrude Stein, and Ida B. Wells-Barnett. Working in human-oriented empiricist disciplines in the midst of methodological reconfiguration around the work of data collection, these writers experiment with data as a form of representing lives. Using a comparative approach that situates works from a range of modernist canons in the data episteme, I seek to recognize the long conceptual history of data as a technology of selfhood and its intersection with material histories of power—who has the ability to collect data about whom and whose narration of that data is seen as authoritative. I argue that these writers draw from their work in sociology, history, psychology, and journalism to formulate critical data aesthetics as they confront questions of identity around race, gender, and nation both in their research and their life writing.

. Chapter four, “To Reproduce a Record: Ida B. Wells-Barnett and the Labor of Data Collection,” returns to a collector of the data of Black life, and death, who defied threats of physical violence to claim the status of data narrator and intervene in such projections. In her investigative reporting on lynching and in her autobiography, Wells-Barnett’s critical data aesthetic refracts the dynamics of race, class, and gender that have defined the prior chapters through the lens of Black womanhood. Recognizing that the received narrative about lynching as an aberrational event spurred by Black criminality serves to forestall public outrage and response, she undertakes “an investigation of every lynching I read about” (Crusade, 64). She seeks to compile an exhaustive collection of lynching data in order to disrupt this narrative, and she does so by rigorously collecting newspaper coverage of lynchings in the white press. She finds, however, that data collection does not guarantee data circulation. She must go one step further than Du Bois in embodying data, traveling across the United States and the United Kingdom to share it through direct public address. The methods and experiences of this campaign ultimately shape her autobiographical practice, as she assembles a large portion of her life narrative by reproducing newspaper clippings reporting on her work. While Wells-Barnett has recently become more widely known as a forerunner of the contemporary turn toward data journalism,30 her innovative use of data aesthetics has not been closely considered or connected to her autobiographical work. Her work models data collection as a mode of grassroots activism rather than an Page 32 →elite professional practice, making the creative labor of data collection visible through her attention to the material conditions of publication, preservation, and circulation of data.

. Between the popularity of self tracking devices and the “collect it all”31 ethos of both state and corporate surveillance, nearly all of our lives are now either collecting or collected lives. Often both. But the uses to which our collected life data can be put, the stories that are told with our data, vary widely and are applied differentially. In the coda, “Data-Driven Modernism Against Algorithmic Identity,” I consider how the critical stances toward data offered by these writers’ modernist data aesthetics provide insight for contemporary resistance to algorithmic identification, the increasingly pervasive process of “identity formation that works through mathematical algorithms to infer categories of identity on otherwise anonymous beings” (Cheney-Lippold, “New,” 165). Algorithmic identification is, in many ways, only the most recent attempt to assign identity narratives on the basis of seemingly empirical evidence, performing the same kinds of narrative condemnation or valuation of lives on the basis of race, gender, and class that these writers challenged. Connecting these pre-digital practices of life data collection with contemporary modes of assembling the self as data demonstrates the continuing necessity of considering the relationship of narrative form to data, especially as data is used to represent lives.

. Data collection as a method of empirical inquiry, though, was emerging more as alibi than antidote to racist theories and practices. The 1890s marked both the beginning of Du Bois’s professional life and the emergence of what Khalil Muhammed has termed “a new social scientific discourse on the Negro Problem . . . set in motion by a racial data revolution” (33). Recapitulating the classic dynamics of data discourse in white supremacist form, studies drawing on data’s presumed objectivity and inarguability to make quantitative cases for Black inferiority proliferated. Nathaniel Shaler’s “Science and the African Problem” (1890), for example, Page 39 →published in the Atlantic Monthly, sidesteps overtly racist argumentation by calling for massive data collection about African Americans. The data he calls for includes anthropological research into African heritage and anthropometry. Clearly, while these data may be construed as more objective because they are constituted of measurement and historical event, they are conceptualized from a deeply racist theory of human difference, one which assumes whiteness embodies the epitome of development and blackness obvious inferiority. Frederick Hoffman’s Race Traits and Tendencies (1896) is the ensuing landmark of racial data collection. Hoffman, a German-born insurance actuary, assembled previously collected data on African American crime and mortality to “prove” that African Americans could not only be justifiably charged higher insurance premiums but were also, as a race, destined for extinction and therefore undeserving of organized, sustained social assistance (let alone compensation for generations of forced labor). As Muhammad describes, “Hoffman combined crime statistics with a well-crafted white supremacist narrative to shape the reading of Black criminality while trying to minimize the appearance of doing so” (51). Data collectors like Shaler and Hoffman conjoin data discourse to unreconstructed racial essentialism. Their work was influential not because of its rigor but because its appearance of objectivity provided adequate cover to legitimize blatant white supremacy. The use of data provided only a veneer of empiricism that was used to support arguments for racial inferiority that would appease Northern suspicions of Southern racism and offer a new pseudoscientific cover for revamped racist practices in insurance sales and policing.1

. Du Bois sees data differently. He conceptualizes empirical study as his opportunity to intervene in a world “thinking wrong about race, because it did not know” (Dusk of Dawn, 58). The narrow scope and reductive forms of data available to build this knowledge are part of the problem. When the data of Black life is constricted to records of criminality and mortality, a simplistic single narrative can be formed and repeated in part because evidence outside of the death-defined plot has been ignored or not collected at all. Further, data used in solely statistical ways performs a dehumanizing abstraction upon its subjects, isolating events and measurements that should be contextualized in circumstances, histories, and locations. Du Bois hopes to disrupt these paths of interpretation through more comprehensive data representation of African American reality. When he receives the commission of the University of Pennsylvania, Du Bois determines to “ignore the pitiful stipend” and build a sociological method built on “facts, any and all facts, concerning the American Negro and his plight” (Dusk of Dawn, 51). The value he places on the collection of these “facts, any and all facts” suggests the power that he believes data holds. He is sponsored to collect data because it is assumed that it will confirm preconception, but he conceptualizes data as disruptive and conjectures that it has the potential to tell a different story.2 He builds his career on data, but he wagers no less than his humanity on the belief that starting with the act of collecting data, rather than the act of proposing a theory, will disrupt predetermined assumptions about racial destiny and potential.

. Du Bois thus describes a prototype of the middle class white citizen demonstrating what social theorists of race today term implicit stereotyping, having absorbed circulated images of a few as the reality of all in another social group.3 This observer also has a racially inflected version of the anesthetized subjectivity that modernist aesthetics of defamiliarization attempt to disrupt: he is so entrenched in familiar narratives that he is unable to perceive the very world he moves through, his thought short-circuited. As Eduardo Bonilla-Silva has noted, “The central component of any dominant racial ideology is its frames or set paths for interpretation” (26). The epistemological imperative of exhaustive data collection does not deny that crime and poverty exist but does insist on documenting them in comprehensive material context so as to intervene in the next step of the white observer’s thought process: “All this is true—all these problems are there and of threatening intricacy; unfortunately, however, the interest of the ordinary man of affairs is apt to stop here. Crime, poverty and idleness affect his interests unfavorably and he would have them stopped; he looks upon these slums and slum characters as unpleasant things which should in some way be removed for the best interests of all” (The Philadelphia Negro, 3–4). The “ordinary man of affairs” cannot perceive particularity; his potential engagement with reality is scuttled by superficial conclusions.

. The continual co-presence of racializing discourses in circulation around and publicly applied to one’s self creates a form of selfhood that Du Bois famously describes as double consciousness. Double consciousness as he describes it here is “this sense of always looking at one’s self through the eyes of others, of measuring one’s soul by the tape of a world that looks on in amused contempt and pity” (The Souls of Black Folk, 8). While many critical readings4 have elaborated the concept of double consciousness and its social and historical production, contextualizing the language of its formulation in the epistemology of data suggests that we must also think about how the cultural force of empiricist thought figures into the production of such consciousness. Du Bois refers specifically to the idea of quantitative measurement—the measuring tape of the soul devised by another world—being used to define the dimensions of African American life. The spectatorial stance of the sociological observer also plays a role in generating this consciousness. The “sense of always looking at one’s self through the eyes of others” is the central dynamic of empirical inquiry turned upon the self. To know the self is to see the self as another, for in empiricist terms, only an observer outside the self can truly see. Observation of African American lives is further estranged from the self because the terms of legitimate observation are defined by white cultures of empiricism. And of course, the privileged legitimacy of what this observer sees is underwritten by the idea that such a distanced, objective stance is not only possible but also synonymous with the revelation of the real. While there are elements of this formulation that can be seen as unique to Du Bois’s experience and particularly relevant for African American experience at this time, it also enunciates a facet of the modern self: the accumulation and narrativization of data can constitute an externally circulating form of selfhood that disrupts an individual’s observation and narrativization of his own experience. As Wynter and Weheliye have argued, Black life is ground zero for the definition of the human, and in this case, ground zero for the constitution of the self outside the self through data.

. The Education of Henry Adams would seem to share many formal qualities with W. E. B. Du Bois’s Dusk of Dawn.1 Both autobiographical works conceptualize the human life not as an unfolding narrative but as a chronological and psychic space to be surveyed as exhaustively as possible. For both, this scope of observation leads to such a deep contextualization of the self that the assumption of individuality gives way to a perception of relationality. Du Bois describes his autobiography as a record of “the interaction of the stream and change of my thought” (Dusk, 7) assembled in order to represent the “significance of [the] race problem by explaining it in terms of the one human life I know best” (viii). Adams similarly figures his as the observation of a “manikin,” himself in the third person, used for “the study of relation” (The Education, 8). The significance of their lives, so represented, derives not from any internal exceptionality but from the Page 78 →accident of birth in times of social transformation. Du Bois states, “My life had its significance and its only deep significance because it was part of a Problem . . . the central problem of the greatest of the world’s democracies and so the Problem of the future world” (Dusk, vii–viii). Adams similarly positions himself as situated in peculiarly uncertain times, describing the infant Henry Brooks Adams as being born into a moment in which “the old universe was thrown into the ash-heap and a new one created” (The Education, 10), as a child whose life would answer the question of what this new universe would be. Both of these life stories are data collections meant to address the question of the future.

. The two passages of life writing that form the epigraph to this chapter demonstrate how this uncertainty manifests in conceptions of narrative form. John Quincy Adams, writing in his diary as the twenty-eight-year-old son of a U.S. president, sees his life and education as seamlessly intertwined, coherent, and defined by an economy of effort that virtually guarantees his success. His “studies” are “all directed to one point,” a single Page 79 →end of his life story clearly “pointed out” by the “station” to which he is born. His grandson, Henry Brooks Adams, writing a century later from the vantage of midlife, wants to see education the same way: as a path to social and political prominence, a narrative trajectory confirming that “a system of society which had lasted since Adam would outlast one Adams more” (The Education, 21). But when he surveys the data of his own education, he sees only the collection of impressions. The impression as educational experience is a haphazard substitute for his grandfather’s coherent studies. Impressions imply the diffusion of effort and the accumulation of experience without a direction of development, and they promise nothing so stationary as a “station” for the erstwhile subject of education. Collective, rather than selective, proliferating points without pointing anywhere, the impression as form of education substitutes a data aesthetic for a developmental narrative.2

. An anxious relationship to the future is what data is meant to forestall through modeling and prediction. To combat uncertainty, Adams and many of his social and intellectual peers turn toward empirical social sciences. As Dorothy Ross has argued, the push toward empirical social sciences is especially pronounced in the U.S. context precisely because of the anxieties of a “gentry” class, “largely northeastern in residence, well-educated, liberal or heterodox in religion” (53), accustomed to wielding intellectual authority, ensconced in positions of social power, and highly invested in American exceptionalism.3 Adams participates in this search for “fundamental laws at work alike in nature and history” (Ross, 60) as a scientific historian, part of a nineteenth-century intellectual movement toward aligning methods of historical inquiry with those of science. Empirical methods are thought to not only provide the best knowledge of reality but also to provide the path back to certainty and security.

. Page 81 →For Adams, a rigorous data collector as both historian and autobiographer, data instead becomes the confirmation of uncertainty, due not only to its content but to the form of reality that it suggests. Specifically, the reality of history and self as presented to him by data collection takes on the form of self-conscious assemblage, in which the data-driven narrator must always choose which configuration of points to privilege. He may thus choose to privilege the United States as an exceptional republic and himself as an exceptional U.S. American, but the data suggesting alternate, equally true narrative assemblages will remain to confront him. I do not claim, here, that The Education’s textual affinity to data collection is literal. Adams is not actually presenting us with an unedited list of observations. As a number of critics have pointed out, Adams’s ironic, aphoristic style is highly polished.4 The perpetual frustration of narrative, the self-conscious claim that the story is not finished, does not imply that the text itself is less finished or that a data aesthetic requires that the text be fully non-selective. What a data aesthetic calls attention to is that Adams’s selections represent a self that cannot justify selecting.

. These writers reflect on our own data surround in unexpected ways, because they share the desire for, suspicion of, and aspiration to total data representation that animates much of contemporary data discourse, yet they negotiate these relationships to data in a pre-digital ecology of affordances for collection, manipulation, and interpretation. This means that their engagement with data is more literally an encounter with data—more embodied and more often on a point-by-point basis rather than parsed through filters and statistical summaries. As Sara Ahmed elaborates the Page 4 →significance of encounter, “The term encounter suggests a meeting, but a meeting which involves surprise and conflict” (6). Ahmed’s theorization of the encounter includes both face-to-face meetings and the more conceptual, but equally powerful “coming together of at least two elements,” such as “a meeting between reader and text” (7). Encounters are experiences that force us to ask, “how does identity become instituted through encounters with others that surprise, that shift the boundaries of the familiar, of what we assume we know?” (6–7). The data collection practices and technologies of the pre-digital era produce encounters that surprise and defamiliarize. When Du Bois, for example, collects data of African American life for The Philadelphia Negro, he is not querying a series of dot-gov databases with historical information on household make-up, income, and property values. He walks door to door to complete the five thousand questionnaires that make up his database. He personally enacts the encoding of perception and interaction into number and word. He then stores, organizes, and reorganizes this data in paper form. When Stein, as a medical student, works to model the human brain, she is not using imaging software to apply multiple comparison methods to a set of functional magnetic resonance imaging (fMRI) scans. She is drawing, by hand, every facet of every tissue section put in front of her. Because of these writers’ more intimate encounters with the data points they collect, their perspective on data collection as a representational form offers a vantage that has been obscured by contemporary interfaces.4 While so many of the tools we use as professional and lay readers of a data-enveloped world seem to flatter our pretensions of being able to see everything and know it all, these writers daily perform the lossy translation of experience into data and confront the impossibility of keeping every point in view. This perspective marks them as forebears of critical data studies, a field of “research and thinking that applies critical social theory to data to explore the ways in which they are never simply neutral, objective, independent, raw representations of the world, but are situated, contingent, relational, contextual, and do active work in the world” (Kitchin and Lauriault, 7). In their linkage of data collection and embodied selfhood, the authors considered in this book situate data in time and space, expose the creative processes that underlie its seeming rawness, consider the political ramifications of its collection and interpretation, and above all, challenge the idea that if we only collect enough data, it will narrate itself.

. In this chapter, I examine data collection as a method and form underlying Adams’s scientific history and his autobiography in order to grapple with the unfamiliarity of the self as data. I begin by reading Adams’s theorization of scientific history as a data-driven conception of reality, building from Taylor’s argument that “it is Adams’s science—not his biography—that provides the best explanation for his impoverished status in The Education” (61). I do not seek to discount or ignore the political dynamics of his biography, but I do focus on how empiricist science became his primary lens for attempting to understand and (ultimately failing) to contain those dynamics. Within the broader umbrella of Adams’s science, I argue that the concept of data and the practice of data collection are underlying, and typically overlooked,5 epistemological and formal dynamics driving The Education and its model of selfhood. Linking the narrative form of The Education to an aesthetics of data collection, I propose that Adams’s ostentatious failure to narrate his education marks his confrontation with data collection as a representational form of selfhood. Reading the dualism of unity and multiplicity as a figuration of the dualism of narrative and data collection, I situate The Education’s representation of selfhood at the crux of two ontological questions provoked by the epistemology of data: who am I, and what is the status of the human in a paradigm of pervasive empiricism? After repeated attempts to answer these questions through attempted theorization and practice of scientific history, though, Adams finds that any promise of greater clarity is continually frustrated by the addition of new data. Further, as he turns to scientific method to get at history’s underlying order, contemporary science begins to claim the reality of disordering forces. Turning this method of history upon his own life, and turning his life upon the pretensions of scientific history, he struggles with an unshakeablePage 83 → recognition that his modern self is formed not by developmental narrative but by anxious assemblage.

. Adams’s commitment to empiricism is manifest in his association with the movement toward scientific history. Though far from homogenous Page 85 →in theory and practice, U.S. scientific historians comprise a “broad trend away from what many came to consider the overemphasis on narrative by so-called literary historians like Macaulay, Michelet, and Prescott” (Jordy, 3). Methodologically, they drew heavily on the work of German historian Leopold Von Ranke (1795–1886). Ranke proposed that the key value of historical research was to “reveal history as it had actually happened” (Jordy, 2). The “actually” here implies a stark difference from the goals of narrative history.6 Known during his academic life in Berlin as an opponent of Hegelian teleology, Ranke eschews the assumption of an underlying developmental process (or narrative form) that can be projected onto history. Ranke theorizes historical method as primarily a practice of data collection. The historian’s goal, as he understands it, is exhaustive representation through the collection of primary source documents. The least mediated sources were to be sought and the historian was to refrain from applying preexisting schemas of understanding in presenting them. Ranke writes in 1839, “I see the time approach in which we shall no longer have to found modern history on the reports even of contemporary historians, except in so far as they were in possession of personal and immediate knowledge of facts; still less, on works yet more remote from the source; but on the narratives of eye-witnesses, and the genuine and original documents” (xi). Ranke’s influence on scientific history in the United States is evident in expressions of confidence in the project of data collection. The 1901 Annual Report of the American Historical Association, for example, reports that Edward Cheyney argues that the “simple but arduous task of the historian was to collect facts, view them objectively, and arrange them as the facts themselves demanded, without reference to any especial operating force beyond that clearly shown by actual conditions” (American Historical Association, vol. 1, 29). Albert Bushnell Hart, addressing the same body in 1910, calls for a “genuinely scientific school of history” modeled on the “fortunate analogy of the physical sciences” (232). He contends that if historians follow the lead of Darwin who “[spent] twenty years in accumulating data . . . before he so much as ventured a generalization,” then they will find that “in history, too, scattered and apparently unrelated data fall together in harmonious wholes” (232–33).7 Thus, the aspirations of scientific history echo the Laplacian desire for exhaustive data and anticipate current speculations around historical methods, in which historians have started contemplating what it might be like to have access to vast archives of data recording actual human behavior and thinking in real time.8 The Education’s descriptions of Adams as scientific historian confirm a similar Page 86 →conception of historian as data collector. Adams imagines historical data as discrete, truth-bearing points through which an ultimate reality outside the self will be revealed. History is the ceaseless generation of material traces that constitute an objective record of these points: “History set it down on the record—pricked its position on the chart—and waited to be led, or misled, once more” (The Education, 423). The scientific historian is one who seeks the totality of this record.

. Adams’s commitment to empiricism is manifest in his association with the movement toward scientific history. Though far from homogenous Page 85 →in theory and practice, U.S. scientific historians comprise a “broad trend away from what many came to consider the overemphasis on narrative by so-called literary historians like Macaulay, Michelet, and Prescott” (Jordy, 3). Methodologically, they drew heavily on the work of German historian Leopold Von Ranke (1795–1886). Ranke proposed that the key value of historical research was to “reveal history as it had actually happened” (Jordy, 2). The “actually” here implies a stark difference from the goals of narrative history.6 Known during his academic life in Berlin as an opponent of Hegelian teleology, Ranke eschews the assumption of an underlying developmental process (or narrative form) that can be projected onto history. Ranke theorizes historical method as primarily a practice of data collection. The historian’s goal, as he understands it, is exhaustive representation through the collection of primary source documents. The least mediated sources were to be sought and the historian was to refrain from applying preexisting schemas of understanding in presenting them. Ranke writes in 1839, “I see the time approach in which we shall no longer have to found modern history on the reports even of contemporary historians, except in so far as they were in possession of personal and immediate knowledge of facts; still less, on works yet more remote from the source; but on the narratives of eye-witnesses, and the genuine and original documents” (xi). Ranke’s influence on scientific history in the United States is evident in expressions of confidence in the project of data collection. The 1901 Annual Report of the American Historical Association, for example, reports that Edward Cheyney argues that the “simple but arduous task of the historian was to collect facts, view them objectively, and arrange them as the facts themselves demanded, without reference to any especial operating force beyond that clearly shown by actual conditions” (American Historical Association, vol. 1, 29). Albert Bushnell Hart, addressing the same body in 1910, calls for a “genuinely scientific school of history” modeled on the “fortunate analogy of the physical sciences” (232). He contends that if historians follow the lead of Darwin who “[spent] twenty years in accumulating data . . . before he so much as ventured a generalization,” then they will find that “in history, too, scattered and apparently unrelated data fall together in harmonious wholes” (232–33).7 Thus, the aspirations of scientific history echo the Laplacian desire for exhaustive data and anticipate current speculations around historical methods, in which historians have started contemplating what it might be like to have access to vast archives of data recording actual human behavior and thinking in real time.8 The Education’s descriptions of Adams as scientific historian confirm a similar Page 86 →conception of historian as data collector. Adams imagines historical data as discrete, truth-bearing points through which an ultimate reality outside the self will be revealed. History is the ceaseless generation of material traces that constitute an objective record of these points: “History set it down on the record—pricked its position on the chart—and waited to be led, or misled, once more” (The Education, 423). The scientific historian is one who seeks the totality of this record.

. Adams’s commitment to empiricism is manifest in his association with the movement toward scientific history. Though far from homogenous Page 85 →in theory and practice, U.S. scientific historians comprise a “broad trend away from what many came to consider the overemphasis on narrative by so-called literary historians like Macaulay, Michelet, and Prescott” (Jordy, 3). Methodologically, they drew heavily on the work of German historian Leopold Von Ranke (1795–1886). Ranke proposed that the key value of historical research was to “reveal history as it had actually happened” (Jordy, 2). The “actually” here implies a stark difference from the goals of narrative history.6 Known during his academic life in Berlin as an opponent of Hegelian teleology, Ranke eschews the assumption of an underlying developmental process (or narrative form) that can be projected onto history. Ranke theorizes historical method as primarily a practice of data collection. The historian’s goal, as he understands it, is exhaustive representation through the collection of primary source documents. The least mediated sources were to be sought and the historian was to refrain from applying preexisting schemas of understanding in presenting them. Ranke writes in 1839, “I see the time approach in which we shall no longer have to found modern history on the reports even of contemporary historians, except in so far as they were in possession of personal and immediate knowledge of facts; still less, on works yet more remote from the source; but on the narratives of eye-witnesses, and the genuine and original documents” (xi). Ranke’s influence on scientific history in the United States is evident in expressions of confidence in the project of data collection. The 1901 Annual Report of the American Historical Association, for example, reports that Edward Cheyney argues that the “simple but arduous task of the historian was to collect facts, view them objectively, and arrange them as the facts themselves demanded, without reference to any especial operating force beyond that clearly shown by actual conditions” (American Historical Association, vol. 1, 29). Albert Bushnell Hart, addressing the same body in 1910, calls for a “genuinely scientific school of history” modeled on the “fortunate analogy of the physical sciences” (232). He contends that if historians follow the lead of Darwin who “[spent] twenty years in accumulating data . . . before he so much as ventured a generalization,” then they will find that “in history, too, scattered and apparently unrelated data fall together in harmonious wholes” (232–33).7 Thus, the aspirations of scientific history echo the Laplacian desire for exhaustive data and anticipate current speculations around historical methods, in which historians have started contemplating what it might be like to have access to vast archives of data recording actual human behavior and thinking in real time.8 The Education’s descriptions of Adams as scientific historian confirm a similar Page 86 →conception of historian as data collector. Adams imagines historical data as discrete, truth-bearing points through which an ultimate reality outside the self will be revealed. History is the ceaseless generation of material traces that constitute an objective record of these points: “History set it down on the record—pricked its position on the chart—and waited to be led, or misled, once more” (The Education, 423). The scientific historian is one who seeks the totality of this record.

. Adams’s criticism of Lyell’s and Spencer’s selective practices demonstrates his critical awareness of empiricism’s mandate to account for all the data, and this awareness becomes his crucial intervention in empiricism’s claiming of the human. Adams becomes more and more pessimistic about Page 90 →the viability of scientific history in light of the implications of contemporary physical science. Physical sciences were, in the early twentieth century of Adams’s later career, grappling with evidence that material reality was far more complex than Newtonian physics predicted, perhaps even ultimately chaotic. As Adams writes “Rule of Phase,” a treatise applying J. Willard Gibbs’s phase rule to history to demonstrate and explain the transition between distinct historical phases,9 work in chemistry and physics has effectively ended the dream of extending the neatly predictive formulas of Newtonian mechanics to the atomic level. Adams may not have fully understood the science, but he intuited the implications of the reality proposed by theoretical physics, specifically thermodynamics and the law of entropy. Not only does science seem to fail to provide predictive laws, the laws it does provide quash any potential for human creativity and agency in the face of inevitable decay.

. To understand why Adams’s relationship to assemblage selfhood is an anxious one, we might begin with biographer Edward Chalfant’s incisive observation: “Adams had not become a great historian out of special interest in the past. . . . It was because he needed to learn where America was going that he had given most of the years of his second life to a systematic survey of a selected sample of the past” (73).10 Adams sought to understand the past in order to predict, and potentially control, the future. Uncertainty is intrinsic to assemblage form. While narratives operate in terms of cause and effect, assemblages operate in terms of becoming, or unpredictable emergence through ongoing reconfiguration. Unsurprisingly, then, assemblage selfhood is for Adams a source of anxiety, an affective stance toward the future characterized by uncertainty.

. The doubled self can be read as an anxious version of parallel selfhood, as Sianne Ngai’s formulation of anxiety illuminates. Ngai argues that “while intimately aligned with the concept of futurity, and the temporal dynamics of deferral and anticipation in particular, anxiety has a spatial dimension as well” (210). The anxious subject cannot differentiate between a securely demarcated “here” and a threatening, unknown “yonder” (212).11 Ngai finds that this affect is particularly likely to manifest for the “knowledge-seeking subject” (212) who must secure “an auratic distance from the worldly or feminine sites of asignificance or negativity” (236). Anxiety can be seen in Page 100 →the desire to winnow parallelized perception down to doubleness, a clarifying “here” and “yonder” out of the destabilizing awareness of the equally present reality of other people and places.

. Charting the impressions that never amount to education, Adams must perceive himself as multiple and his future as uncertain. The perception of multiplicity extends to his view of others’ lives, too: “Between 1850 and 1900 nearly everyone’s existence was exceptional” (The Education, 40). In other words, Adams perceives every life during this period as a singular data point. Adams is not the only one who cannot contain his selfhood in a coherent narrative, as he observes, “No scheme could be suggested to the new American” (461). Adams finds himself ordinary because he is “exceptional” in the same sense that “everyone’s existence” is. He can no longer think of the narrative selfhood he receives from his family as “the” American model but must instead perceive himself as one of many models and as a model that must change. Early in The Education, he describes the end of education as narrative form by explicitly comparing his own education to that of “outsiders, immigrants, adventurers” (29). As a child of the first half of the nineteenth century, he bore the “stamp of 1848 . . . almost as indelible as the stamp of 1776, but in the eighteenth, or any earlier century, the stamp mattered less because it was standard, and everyone bore it” (29). But “men whose lives were to fall in the generation between 1865 and 1900” instead had to “take the stamp that belonged to their time. This was their education. To outsiders, immigrants, adventurers, it was easy, but the old puritan nature rebelled against change” (29). Instead of being stamped, or formed by a single, progressive/developmental educational narrative, the self must become re-stampable, constantly educable. It is not just a matter of trading one stamp for another but trading the idea of a finished, developed self for a readiness to be developed over and over again. While we should certainly question that it was “easy” for the other, less socially advantaged selves he notes, we can still find the perception meaningful.12 Granting that Adams is, as Kurt Albert Mayer points out, a racialist by the standards of his own time and a racist by the standards of our own, in The Education even this strain of narrow concern for self holds insight into a shifting awareness of the self’s relationship to social collectivities. Through perceiving the multiplicity of life paths that have come to represent U.S. American selfhood, Adams is thrust out of the idea that his is “standard.” He sees the nation as a sprawl of disconnected endeavors, just as he perceives himself as a bundle of memories with no inherent connection but those self-consciously constructed.

. Stein finds that this commitment to exhaustivity in representation, though, introduces formal potentials that interest her more than completing a schema. Stein claims that MoA along with Three Lives, also written in the decade after she leaves medical school, mark the origin of an aesthetic she calls the “continuous present.” She writes, “In these two books there was elaboration of the complexities of using everything and of a continuous present and of beginning again and again and again” (“Composition,” 220). The equation of the “continuous present” with “the complexities of using everything” and “beginning again and again and again” marks it, I Page 108 →will argue, as a data aesthetic. In its imagination of a mode of writing that uses “everything,” it implies a practice of exhaustive collection. The commitment to working out the “complexities of using everything” indicates a desire to keep every data point, not discarding those that would enable the production of a coherent narrative. More committed to exhaustivity in collection than coherence of result, Stein insists on a data epistemology of multiplicity and difference rather than revelation and certainty. Scholars have considered her literary writing in terms of science broadly conceived,1 in terms of specific theories put forward by her mentors and interlocutors,2 in terms of specific theories of personality and gender,3 and in terms of the epistemic virtue of objectivity.4 What these studies have not so far addressed is data collection as a method of inquiry and representational form underlying her work across scientific disciplines and how it informs her theorization of narrative. Data, the collection of discrete, coequal observations to represent the real, is an unexplored context5 for Stein’s descriptive, paratactic, and repetitive style. This style plays an especially important role in the early prose that she claims led to her “knowing that I was a genius” (Everybody’s Autobiography, 79) and specifically in “Melanctha” which she claims as “the story that was the beginning of her revolutionary work” (Autobiography of Alice B. Toklas, 82).

. Stein finds that this commitment to exhaustivity in representation, though, introduces formal potentials that interest her more than completing a schema. Stein claims that MoA along with Three Lives, also written in the decade after she leaves medical school, mark the origin of an aesthetic she calls the “continuous present.” She writes, “In these two books there was elaboration of the complexities of using everything and of a continuous present and of beginning again and again and again” (“Composition,” 220). The equation of the “continuous present” with “the complexities of using everything” and “beginning again and again and again” marks it, I Page 108 →will argue, as a data aesthetic. In its imagination of a mode of writing that uses “everything,” it implies a practice of exhaustive collection. The commitment to working out the “complexities of using everything” indicates a desire to keep every data point, not discarding those that would enable the production of a coherent narrative. More committed to exhaustivity in collection than coherence of result, Stein insists on a data epistemology of multiplicity and difference rather than revelation and certainty. Scholars have considered her literary writing in terms of science broadly conceived,1 in terms of specific theories put forward by her mentors and interlocutors,2 in terms of specific theories of personality and gender,3 and in terms of the epistemic virtue of objectivity.4 What these studies have not so far addressed is data collection as a method of inquiry and representational form underlying her work across scientific disciplines and how it informs her theorization of narrative. Data, the collection of discrete, coequal observations to represent the real, is an unexplored context5 for Stein’s descriptive, paratactic, and repetitive style. This style plays an especially important role in the early prose that she claims led to her “knowing that I was a genius” (Everybody’s Autobiography, 79) and specifically in “Melanctha” which she claims as “the story that was the beginning of her revolutionary work” (Autobiography of Alice B. Toklas, 82).

. The lens of modernist aesthetics allows us to register these writers’ innovative engagement with data collection and data representation on the level of both content and form. I bring together the terms “data-driven” and “modernism” to suggest a literary critical framework for apprehending the historical, formal, and epistemological uses of data. In present usage, “data-driven” denotes a process of decision-making that purports to look to data for its direction and justification, implying that decisions can be made more effectively and without the danger of bias or misguided tradition by referring to data.5 I put a different twist on the term to call attention to how imagining reality as a vast data collection drives that imagination toward certain formal parameters. Data-driven modernists, in my definition, commit to the form of data collection first and then see what narratives result. In this way, they highlight how data complicates, rather than clarifies, our narrative representations of reality. I approach modernism from a historical as well as aesthetic perspective, following interpretive pathways opened by the work of scholars such as Andreas Huyssen, Susan Stanford Friedman, Douglas Mao, Rebecca Walkowitz, and Werner Sollors, who have challenged critics to consider how our methodological concatenation of text and context must shift when we loosen the association of modernism with a few representative works and seek the “effects of synergy or friction [that] result when the many, sometimes contradictory criteria of high modernism are tested against less evidently experimental texts” (Mao and Walkowitz, Bad Modernisms, 2).6 Considering historical context alongside aesthetic practice is especially important for critical consideration of race and ethnicity as factors shaping literary production, as the work of Joshua Miller, Sarah Wilson, and Leif Sorensen has modeled.7 Focusing on data as a representational strategy with aesthetic as well as epistemological implications brings into a view a friction-generating range of texts and creators, embedded in multiple disciplines and literary communities.

. Stein finds that this commitment to exhaustivity in representation, though, introduces formal potentials that interest her more than completing a schema. Stein claims that MoA along with Three Lives, also written in the decade after she leaves medical school, mark the origin of an aesthetic she calls the “continuous present.” She writes, “In these two books there was elaboration of the complexities of using everything and of a continuous present and of beginning again and again and again” (“Composition,” 220). The equation of the “continuous present” with “the complexities of using everything” and “beginning again and again and again” marks it, I Page 108 →will argue, as a data aesthetic. In its imagination of a mode of writing that uses “everything,” it implies a practice of exhaustive collection. The commitment to working out the “complexities of using everything” indicates a desire to keep every data point, not discarding those that would enable the production of a coherent narrative. More committed to exhaustivity in collection than coherence of result, Stein insists on a data epistemology of multiplicity and difference rather than revelation and certainty. Scholars have considered her literary writing in terms of science broadly conceived,1 in terms of specific theories put forward by her mentors and interlocutors,2 in terms of specific theories of personality and gender,3 and in terms of the epistemic virtue of objectivity.4 What these studies have not so far addressed is data collection as a method of inquiry and representational form underlying her work across scientific disciplines and how it informs her theorization of narrative. Data, the collection of discrete, coequal observations to represent the real, is an unexplored context5 for Stein’s descriptive, paratactic, and repetitive style. This style plays an especially important role in the early prose that she claims led to her “knowing that I was a genius” (Everybody’s Autobiography, 79) and specifically in “Melanctha” which she claims as “the story that was the beginning of her revolutionary work” (Autobiography of Alice B. Toklas, 82).

. Stein finds that this commitment to exhaustivity in representation, though, introduces formal potentials that interest her more than completing a schema. Stein claims that MoA along with Three Lives, also written in the decade after she leaves medical school, mark the origin of an aesthetic she calls the “continuous present.” She writes, “In these two books there was elaboration of the complexities of using everything and of a continuous present and of beginning again and again and again” (“Composition,” 220). The equation of the “continuous present” with “the complexities of using everything” and “beginning again and again and again” marks it, I Page 108 →will argue, as a data aesthetic. In its imagination of a mode of writing that uses “everything,” it implies a practice of exhaustive collection. The commitment to working out the “complexities of using everything” indicates a desire to keep every data point, not discarding those that would enable the production of a coherent narrative. More committed to exhaustivity in collection than coherence of result, Stein insists on a data epistemology of multiplicity and difference rather than revelation and certainty. Scholars have considered her literary writing in terms of science broadly conceived,1 in terms of specific theories put forward by her mentors and interlocutors,2 in terms of specific theories of personality and gender,3 and in terms of the epistemic virtue of objectivity.4 What these studies have not so far addressed is data collection as a method of inquiry and representational form underlying her work across scientific disciplines and how it informs her theorization of narrative. Data, the collection of discrete, coequal observations to represent the real, is an unexplored context5 for Stein’s descriptive, paratactic, and repetitive style. This style plays an especially important role in the early prose that she claims led to her “knowing that I was a genius” (Everybody’s Autobiography, 79) and specifically in “Melanctha” which she claims as “the story that was the beginning of her revolutionary work” (Autobiography of Alice B. Toklas, 82).

. Stein finds that this commitment to exhaustivity in representation, though, introduces formal potentials that interest her more than completing a schema. Stein claims that MoA along with Three Lives, also written in the decade after she leaves medical school, mark the origin of an aesthetic she calls the “continuous present.” She writes, “In these two books there was elaboration of the complexities of using everything and of a continuous present and of beginning again and again and again” (“Composition,” 220). The equation of the “continuous present” with “the complexities of using everything” and “beginning again and again and again” marks it, I Page 108 →will argue, as a data aesthetic. In its imagination of a mode of writing that uses “everything,” it implies a practice of exhaustive collection. The commitment to working out the “complexities of using everything” indicates a desire to keep every data point, not discarding those that would enable the production of a coherent narrative. More committed to exhaustivity in collection than coherence of result, Stein insists on a data epistemology of multiplicity and difference rather than revelation and certainty. Scholars have considered her literary writing in terms of science broadly conceived,1 in terms of specific theories put forward by her mentors and interlocutors,2 in terms of specific theories of personality and gender,3 and in terms of the epistemic virtue of objectivity.4 What these studies have not so far addressed is data collection as a method of inquiry and representational form underlying her work across scientific disciplines and how it informs her theorization of narrative. Data, the collection of discrete, coequal observations to represent the real, is an unexplored context5 for Stein’s descriptive, paratactic, and repetitive style. This style plays an especially important role in the early prose that she claims led to her “knowing that I was a genius” (Everybody’s Autobiography, 79) and specifically in “Melanctha” which she claims as “the story that was the beginning of her revolutionary work” (Autobiography of Alice B. Toklas, 82).

. Both James and Münsterberg position psychology as a natural science, with methods built on data collection meant to reveal generalizable, causal law. Both, however, also recognize the tension between a desire for data and the goal of generalization. In Psychology: The Briefer Course, an abridgement of The Principles of Psychology which James assigned in the introductory course Stein took at Harvard in 1893,6 James formulates psychological research as part of an overarching knowledge project that will be driven by numerous processes of data collection. In the introduction, he proposes “Psychology is to be treated as a natural science” (Briefer Course, xxv), one of “a lot of beginnings of knowledge made in different places,” whose pursuit shall culminate in fully realized knowledge. As James avers, “Most thinkers have a faith that at bottom there is but one Science of all things, and that until all is known, no one thing can be completely known” (xxv). Each of these “beginnings” is a field of natural science, a “partial and provisional” (xxvi) piece of the larger puzzle that will be revealed when “all is known.” Given the immensity of the work required before “all is known,” though, the result of total knowledge is deferred and the work of preliminary collection continues indefinitely. James says as much of his own work: “The reader will seek in vain for any closed system in this book. It is mainly a mass of descriptive details” (The Principles of Psychology, vii).

. If gender was the question, type was a common form of the answer. Type as a formal structure of identity underlies both the psychological research in which Stein was engaged and the theories of sexual difference that circulated around her. Understandably, then, typological theories of identity loom large in Stein criticism.7 While the specifics of these theories vary, they share a desire to classify people and verify essential difference. The typological endeavor is a form of the empiricist move from observation to explanation and prediction. Viewed through the lens of type as essential nature, individual acts of “movement, speech, writing, and [idea]” are “only the fragmentary external expressions of some underlying, embodied character” (Will, 21). Stein clearly subscribes to this idea as she writes, in “Cultivated Motor Automatism,” that “habits of attention are reflexes of the complete character of the individual” (299). “Habits of attention” are just one observable feature of “the complete character of the individual,” but if one assumes that a unifying type lies behind all observable features, these limited observations are enough. The assignment of type subsumes discrete data points into an overarching identity that not only explains but predicts the subject’s actions and reactions. Through these predictions, type assigns potentials, including one’s potential to be a genius, a creative intellectual agent. As Will observes, the concept of type precipitates “a change in the perception of ‘genius’ throughout the course Page 117 →of the nineteenth century, from a universal capacity to an embodied type, visibly and measurably distinguishable from non-geniuses through the evidence of discrete physical and mental characteristics” (4). And, per the variability hypothesis and others, if you were female you were most likely not the genius type.

. Asked to revise, Stein declines, both because she is “going abroad for an extended period” and further claims that the work has “to a certain extent” achieved her goal of expressing a “very clear image which exists in my own mind of a region which the existing literature of the subject leaves in a hopeless mess” (from facsimile reproduction of correspondence, Meyer, 94–95). In this letter, Stein appears to believe she has practiced the selective, interpretive task demanded of her rather than the more collection-oriented method I argue is apparent. She claims her work is aimed at correcting textbooks that “tell so much more that one is confused” and making “a pretty careful selection of sections [sic]” to reproduce. That there could be such a disconnect between her assessment of the finished product and that of her colleagues suggests that Stein had a different conception altogether of what type of representation captured reality. Her idea of selection is far more collective than others consider coherent. As Cecire suggests, “Stein was, by the end of medical school, beginning to push her objectivity beyond the norms of usable scientific practice—not retreating from objectivity but, if anything, rather aggressively, even destructively, exploring it Page 122 →further” (101).8 Implicitly, Stein’s staunch attention to the samples before her in both cases challenges the clarifying ends of empiricist data collection, but she finds this challenge dismissed as the assertion of an amateur rather than considered as an alternative perspective.

. Just as they were returning to the town Adele stopped abruptly and faced Helen. “Tell me” she said “do you really care for me any more?” “Do you suppose I would have stayed on here in Sienna if I didn’t” Helen answered angrily. “Won’t you ever learn that it is facts that tell?” Adele laughed ruefully. “But you forget,” she said, “that there are many facts and it isn’t easy to know just what they tell.” (Three Lives and Q.E.D., 226)9

. At first pass, this might seem to put Stein in line with the generalizing tendencies of traditional empiricists and not with an iconoclastic commitment to data collection as I have argued for it. Most critical considerations of Stein’s interest in type and typology begin with her retrospective assessment in “The Gradual Making of The Making of Americans”: “I was supposed to be interested in their reactions but soon I found that I was not but instead that I was enormously interested in the types of their characters that is what I even then thought of as the bottom nature of them” (Lectures, 137). In readings that see Stein’s interest in typology as a rejection of an earlier, more empiricist bent, this statement indicates that interest in type, or “bottom nature,” supersedes attention to data, “their reactions,” collected during the course of experiments.10 Confidence in the eventuality of accounting for the human through a finite number of types is an affirmation of the empiricist project of achieving generalization and prediction. Meyer argues, “Certainly, Stein’s understanding of science was initially mechanistic; thus in The Making of Americans, written between 1902 and 1911, she attempted to describe the precise mechanisms of human personality in great detail, with the ultimate aim of describing every possible kind of human being” (3). Earnest commitment to the project of discovering Page 124 →universal law, however, by no means forecloses unexpected results along the way. If we choose not to see data collection and type determination as mutually exclusive, we can see that Stein both confronted and employed typological theories of self because, at different times and in different contexts, she was less interested in disproving type than proving herself as one who could elucidate type. Just because others had forwarded the wrong deterministic model, in Stein’s view, did not necessarily mean there was not a correct one, the discovery of which would confirm her genius. This is the same kind of investment that Adams makes in scientific history, and for him it also resulted in more narrative disruption than clarity.

. Melanctha’s racialized status is simultaneously referential and abstracted, historical and fictional. One line of critical thought has argued that Stein’s prose signals a degree of aesthetic invention that makes it clear her intention is not to realistically represent an actual mixed race woman. Her use of racial language, unmoored from reference, should instead be seen as either aesthetic play or part of her critique of essentialist notions of identity.11 Further, the romance plot (to the extent that there is one) of “Melanctha” is, by critical consensus, a rewriting of the highly autobiographical Q.E.D., cast in racialized terms in order to deflect attention from her own identities as woman, lesbian, and Jewish.12 The autobiographical element of the story muddies any argument that Stein attempts to portray the reality of a working class, mixed race woman’s life. Yet, as another robust line of criticism has established, Stein’s language in “Melanctha” is so clearly rooted in the mainstream racializing discourse and racist equations of skin color with health, morality, and intelligence that it cannot be disentangled from historical questions, and injustices, of race and representation.13 Stein’s narrator is, at best, “passive to the status quo” (Ruddick, 50), using the vocabulary at hand. As well, as seen in the cited passage above from The Autobiography of Alice B. Toklas, Stein at times suggests she gains inspiration from her encounters with African Americans in Baltimore. Seeing Three Lives as an experiment in the genre of biography and proposing to read “Melanctha” as a data collection, I am inferring that we should read it as if it seems to represent an actual Page 133 →person. These framings entail reading racial descriptors as referential, in which case there is no question that, as the latter group of critics has argued, the describer is uncritically employing racist language, freighted by exactly the kind of narrative projections of identity that Stein wants to avoid having projected onto herself as a woman. It is in part through the use of this language that Stein effects her own escape from narrative projections onto her gender and into the identity of genius, innovator, and modernist. My purpose in reading this language as part of a data aesthetic is not to absolve or explain, but to examine how it works to empower the narrator but entrap the narrated. “Melanctha” stages a battle for self narration in which the data aesthetic proves to be an ambivalent tool.

. Melanctha’s racialized status is simultaneously referential and abstracted, historical and fictional. One line of critical thought has argued that Stein’s prose signals a degree of aesthetic invention that makes it clear her intention is not to realistically represent an actual mixed race woman. Her use of racial language, unmoored from reference, should instead be seen as either aesthetic play or part of her critique of essentialist notions of identity.11 Further, the romance plot (to the extent that there is one) of “Melanctha” is, by critical consensus, a rewriting of the highly autobiographical Q.E.D., cast in racialized terms in order to deflect attention from her own identities as woman, lesbian, and Jewish.12 The autobiographical element of the story muddies any argument that Stein attempts to portray the reality of a working class, mixed race woman’s life. Yet, as another robust line of criticism has established, Stein’s language in “Melanctha” is so clearly rooted in the mainstream racializing discourse and racist equations of skin color with health, morality, and intelligence that it cannot be disentangled from historical questions, and injustices, of race and representation.13 Stein’s narrator is, at best, “passive to the status quo” (Ruddick, 50), using the vocabulary at hand. As well, as seen in the cited passage above from The Autobiography of Alice B. Toklas, Stein at times suggests she gains inspiration from her encounters with African Americans in Baltimore. Seeing Three Lives as an experiment in the genre of biography and proposing to read “Melanctha” as a data collection, I am inferring that we should read it as if it seems to represent an actual Page 133 →person. These framings entail reading racial descriptors as referential, in which case there is no question that, as the latter group of critics has argued, the describer is uncritically employing racist language, freighted by exactly the kind of narrative projections of identity that Stein wants to avoid having projected onto herself as a woman. It is in part through the use of this language that Stein effects her own escape from narrative projections onto her gender and into the identity of genius, innovator, and modernist. My purpose in reading this language as part of a data aesthetic is not to absolve or explain, but to examine how it works to empower the narrator but entrap the narrated. “Melanctha” stages a battle for self narration in which the data aesthetic proves to be an ambivalent tool.

. The lens of modernist aesthetics allows us to register these writers’ innovative engagement with data collection and data representation on the level of both content and form. I bring together the terms “data-driven” and “modernism” to suggest a literary critical framework for apprehending the historical, formal, and epistemological uses of data. In present usage, “data-driven” denotes a process of decision-making that purports to look to data for its direction and justification, implying that decisions can be made more effectively and without the danger of bias or misguided tradition by referring to data.5 I put a different twist on the term to call attention to how imagining reality as a vast data collection drives that imagination toward certain formal parameters. Data-driven modernists, in my definition, commit to the form of data collection first and then see what narratives result. In this way, they highlight how data complicates, rather than clarifies, our narrative representations of reality. I approach modernism from a historical as well as aesthetic perspective, following interpretive pathways opened by the work of scholars such as Andreas Huyssen, Susan Stanford Friedman, Douglas Mao, Rebecca Walkowitz, and Werner Sollors, who have challenged critics to consider how our methodological concatenation of text and context must shift when we loosen the association of modernism with a few representative works and seek the “effects of synergy or friction [that] result when the many, sometimes contradictory criteria of high modernism are tested against less evidently experimental texts” (Mao and Walkowitz, Bad Modernisms, 2).6 Considering historical context alongside aesthetic practice is especially important for critical consideration of race and ethnicity as factors shaping literary production, as the work of Joshua Miller, Sarah Wilson, and Leif Sorensen has modeled.7 Focusing on data as a representational strategy with aesthetic as well as epistemological implications brings into a view a friction-generating range of texts and creators, embedded in multiple disciplines and literary communities.

. Melanctha’s racialized status is simultaneously referential and abstracted, historical and fictional. One line of critical thought has argued that Stein’s prose signals a degree of aesthetic invention that makes it clear her intention is not to realistically represent an actual mixed race woman. Her use of racial language, unmoored from reference, should instead be seen as either aesthetic play or part of her critique of essentialist notions of identity.11 Further, the romance plot (to the extent that there is one) of “Melanctha” is, by critical consensus, a rewriting of the highly autobiographical Q.E.D., cast in racialized terms in order to deflect attention from her own identities as woman, lesbian, and Jewish.12 The autobiographical element of the story muddies any argument that Stein attempts to portray the reality of a working class, mixed race woman’s life. Yet, as another robust line of criticism has established, Stein’s language in “Melanctha” is so clearly rooted in the mainstream racializing discourse and racist equations of skin color with health, morality, and intelligence that it cannot be disentangled from historical questions, and injustices, of race and representation.13 Stein’s narrator is, at best, “passive to the status quo” (Ruddick, 50), using the vocabulary at hand. As well, as seen in the cited passage above from The Autobiography of Alice B. Toklas, Stein at times suggests she gains inspiration from her encounters with African Americans in Baltimore. Seeing Three Lives as an experiment in the genre of biography and proposing to read “Melanctha” as a data collection, I am inferring that we should read it as if it seems to represent an actual Page 133 →person. These framings entail reading racial descriptors as referential, in which case there is no question that, as the latter group of critics has argued, the describer is uncritically employing racist language, freighted by exactly the kind of narrative projections of identity that Stein wants to avoid having projected onto herself as a woman. It is in part through the use of this language that Stein effects her own escape from narrative projections onto her gender and into the identity of genius, innovator, and modernist. My purpose in reading this language as part of a data aesthetic is not to absolve or explain, but to examine how it works to empower the narrator but entrap the narrated. “Melanctha” stages a battle for self narration in which the data aesthetic proves to be an ambivalent tool.

. Seeing the narrator as a data collector and Stein’s continuous present as data collection, we also, however, get a glimpse of how a commitment to representational exhaustivity opens space for dissident claims to knowledge and the power to narrate. Refracted through the lens of professional practices of observation and girded by claims of empiricism, the narrator’s omniscience is a rhetorical effect, not a final word. While this narrator’s declarative syntax makes her seem flatly comfortable with the condemnations her observations constitute,14 as a data collector she is also impelled to collect even what does not accord with the social types and narrative structures these observations are meant to substantiate. For this reason, the text her observations form exceeds the determination her terms seek to enact. While her terminology aligns her with a presumption of type and inherent racial difference, her commitment to exhaustive collection of data compels her research to include data that could undercut that presumed reality.

. Why did the subtle, intelligent, attractive, half white girl Melanctha Herbert love and do and demean herself in service to this coarse, decent, sullen, ordinary, black childish Rose, and why was this unmoral, promiscuous, shiftless Rose married, and that’s not so common either, to a good man of the negroes, while Melanctha with her white blood and attraction and her desire for a right position had not yet been really married. (Three Lives, 89)15

. Given the shifting perimeter of the narrator’s limited omniscience,16 though, we might question whether the desire “to tell a story wholly” is Page 135 →solely that of the narrator. If we saw it instead as the narrator’s report of Melanctha’s desire and we took that desire at its word, we might read its judgments another way. Melanctha’s alleged lack of intention to “leave out big pieces” could be the narrator’s uncomprehending description of a positive intention to include all the pieces. The equation of telling a story “wholly” with being able to represent “what had really happened . . . and what it was that she had really done,” bears a resonant resemblance to Stein’s plea on behalf of another narrative subject, Julia Dehning in MoA. Stein implores “those who read much in story books surely now can tell what to expect of her, and yet, please reader, remember that this is perhaps not the whole of our story either . . . for truly she may work out as the story books would have her or we may find all different kinds of things for her” (MoA, 15). To apply narratives formulated and circulated in “story books,” would be for Stein, “to take the character from our Julia” (15), foreclosing the self’s potential to hold “all different kinds of things.” Given the importance of having this whole story for understanding a woman’s life narrative, we might question the “Melanctha” narrator’s presumed monopoly on the desire for practices of representation that would more nearly realize it.

. Through her practice of “wanderings after wisdom,” Melanctha seeks to claim the status of knowledge worker in her own right. Ruddick argues that “references in ‘Melanctha’ to the heroine’s many ‘wanderings’ have rightly been considered part of a sustained euphemism for sex, but one might easily reverse the emphasis and say that sex itself stands in the story as a metaphor for a certain type of mental activity” (18). Melanctha also undertakes this activity in a manner that aligns with practices of exhaustive data collection. Ruddick continues, “Melanctha’s promiscuity is part of an experiential promiscuity, an inability or unwillingness to approach the world selectively” (18). While Ruddick links Melanctha’s non-selective attention to Jamesian psychology, which finds it pathological, we might also link it to the ideal of exhaustive data collection that drives empiricist inquiry, which would make Melanctha’s mode of perception not unscientific but radically so. Cecire describes Melanctha as doing “camera work, the mechanical female scientific labor of abject mimesis” (97), a description that unlocks a range of resonances with data collection and locates Melanctha in the historical context of professional science, which has typically assigned women to roles requiring the most precise hewing to objectivity but then devalued that objectivity as mindless automatism.17 In this light, she is Stein’s autobiographical double in her persistent fascination with and insistence on the value of exhaustive collection. Both collect data, and using that data both seek to construct life stories governed by an aesthetic of inclusion, or stylistic and thematic insistences on including more information than is necessary to tell a conventionally plotted story. Their aesthetics of inclusion are a willful expansion of realist representation that paradoxically seems like a diminution of significance. In place of traditional plots of marriage or self-discovery, there are records of multiple, temporarily realized and ultimately frustrated, desires for knowledge of self.

. On the other hand, this intellectual project also has disempowering consequences in the context of Melanctha’s social milieu. Seeking to know and tell this whole story of self leads Melanctha into conflict with the narratives of middle-class morality and masculinist empiricism.18 Her relationship with Jeff ends in part because she refuses to disavow an earlier version of herself, when she was in close relationship with friend Jane Harden, who repeatedly behaves in ways that violate Jeff’s ideal narrative of self and who intimates to Jeff that Melanctha took part in activities that would also violate his ideal. Jeff is unable to reconcile Melanctha’s past with the future wife he imagines, and Melanctha refuses to renounce this past, instead claiming the coherence of her entire history and what appear to be conflictual selves. Ultimately, “Melanctha is too many” (Three Lives, 147) for Jeff, and theirs is one of a series of relationships that cannot encompass Melanctha’s insistence on self as movement and multiplicity. Her allegiance to holding the seemingly contradictory aspects of her historical self in parallel precludes conforming to the shape of available narratives of womanhood.

. Under the gaze of both Jeff and the narrator, Melanctha is in the position of experimental subject as Stein describes it in her undergraduate composition theme, forced to account for herself in relationship to an externalized record. Embodied in Jane Harden, Melanctha’s “record is there she cannot escape it” (Stein, reproduced in Miller, 121). Her relationship with Jeff is constrained by the very existence of this record, and Melanctha’s claims that it is incomplete and unrepresentative cannot overcome Jeff’s insistence on its reality. The narrator likewise keeps accumulating data points, forming a text that continually displaces Melanctha from narrative frameworks because of its insistence on exhaustivity and parallelism, refusing to designate certain points as more significant than others. The consequences of portraying another’s life in a manner which represents “each part as important as the whole” (Stein, “Transatlantic,” 15) are different for the representer and the represented. As Lisa Ruddick has argued, in the context of Jamesian psychology losing, or willfully abandoning, a guiding narrative of self is a moral and practical failure.19 In the world of the story itself, Melanctha’s data aesthetic of self has similarly negative consequences.Page 141 → As Corrine Blackmer suggests, she is “an embodiment of the acute invisibility and vulnerability of those who belong to many worlds and whose inability to discover a ‘right position’ results from the failure of others to perceive and, therefore, to ‘read’ them competently” (232). Melanctha never finds a partner willing to share her approach to selfhood, a commentary on the exclusionary nature of social narratives and the illegibility generated by being excluded from them.

. “Record” is something of a keyword for Ida B. Wells-Barnett.1 She uses it in the title of her widely cited anti-lynching publication, A Red Record: Tabulated Statistics and Alleged Causes of Lynchings in the United States. She uses it to describe her own autobiography, written as a contribution to African American history. She uses it to name the core of her intellectual and activist project: to “reproduce a record” of white violence against African Americans. The Society of American Archivists defines a “record” as “data or information that has been fixed on some medium; that has content, context, and structure; and that is used as an extension of human memory or to demonstrate accountability.” Wells-Barnett’s anti-lynching2 writings suggest another definition, one that both intersects with and critiques the technical language offered by contemporary archival science. She writes, “no one who reads the record, as it is written in the faces of the million Page 144 →mulattoes in the South, will for a minute conceive that the southern white man had a very chivalrous regard for the honor due the women of his own race or respect for the womanhood which circumstances placed in his power” (Red Record, 13). In her figuration, “faces” are the medium on which crucial data has been fixed. These faces, described as mulatto to infer their origin in what were almost certainly non-consensual interracial sexual relations, evidence at least a “million” instances of white male sexual violence against Black women. Yet, if a record is meant to serve as a site of memory and therefore accountability, this one has so far failed. News accounts of the lynching of African Americans by white mobs repeatedly tell a story in utter contradiction to the facts this record should be able to represent. As Alison Piepmeier has delineated, the white “lynch narrative” (135) casts the Black male as “a rapist, bestial and savage” and the white male as “chivalric” hero. White supremacy is allowed to circulate as fact not, Wells-Barnett points out, because no refuting record is available for reading but because the existing record has not been read. The data of white supremacy is already recorded on the Black body, but data so embodied has also been deemed illegible and forcibly made ephemeral, in both the literal and representational sense. The pain of the duty to reproduce this record arises not only from the trauma of this violence but also from the recognition that, in its original form, the record will never be read. Wells-Barnett sees that Black data collectors face a non-negotiable “duty” to reproduce the data of Black life in a form sanctioned by the terms of white empiricism. In calling attention to data as reproduction of a reality white observers refuse to perceive, her use of data to intervene in the lynch narrative is also an intervention in data epistemology. Wells-Barnett foregrounds the labors of reproduction inherent in the data representation of Black lives, and in so doing challenges a data epistemology built on the imagined objectivity of the collector and the erasure of the work of data preparation, circulation, and preservation.

. “Record” is something of a keyword for Ida B. Wells-Barnett.1 She uses it in the title of her widely cited anti-lynching publication, A Red Record: Tabulated Statistics and Alleged Causes of Lynchings in the United States. She uses it to describe her own autobiography, written as a contribution to African American history. She uses it to name the core of her intellectual and activist project: to “reproduce a record” of white violence against African Americans. The Society of American Archivists defines a “record” as “data or information that has been fixed on some medium; that has content, context, and structure; and that is used as an extension of human memory or to demonstrate accountability.” Wells-Barnett’s anti-lynching2 writings suggest another definition, one that both intersects with and critiques the technical language offered by contemporary archival science. She writes, “no one who reads the record, as it is written in the faces of the million Page 144 →mulattoes in the South, will for a minute conceive that the southern white man had a very chivalrous regard for the honor due the women of his own race or respect for the womanhood which circumstances placed in his power” (Red Record, 13). In her figuration, “faces” are the medium on which crucial data has been fixed. These faces, described as mulatto to infer their origin in what were almost certainly non-consensual interracial sexual relations, evidence at least a “million” instances of white male sexual violence against Black women. Yet, if a record is meant to serve as a site of memory and therefore accountability, this one has so far failed. News accounts of the lynching of African Americans by white mobs repeatedly tell a story in utter contradiction to the facts this record should be able to represent. As Alison Piepmeier has delineated, the white “lynch narrative” (135) casts the Black male as “a rapist, bestial and savage” and the white male as “chivalric” hero. White supremacy is allowed to circulate as fact not, Wells-Barnett points out, because no refuting record is available for reading but because the existing record has not been read. The data of white supremacy is already recorded on the Black body, but data so embodied has also been deemed illegible and forcibly made ephemeral, in both the literal and representational sense. The pain of the duty to reproduce this record arises not only from the trauma of this violence but also from the recognition that, in its original form, the record will never be read. Wells-Barnett sees that Black data collectors face a non-negotiable “duty” to reproduce the data of Black life in a form sanctioned by the terms of white empiricism. In calling attention to data as reproduction of a reality white observers refuse to perceive, her use of data to intervene in the lynch narrative is also an intervention in data epistemology. Wells-Barnett foregrounds the labors of reproduction inherent in the data representation of Black lives, and in so doing challenges a data epistemology built on the imagined objectivity of the collector and the erasure of the work of data preparation, circulation, and preservation.

. Page 147 →Reproductive labor is particularly important to data collection and analysis, but it is also rarely recognized. As Jean-Christophe Plantin notes, “data never comes as raw, pristine, or ready to use” but rather, “multiple interventions are always needed before data can be reused” (67). The work of the data curators who draw on both skill and judgment to perform these interventions, though, is made doubly invisible. It is devalued within the research community and intentionally repressed from public view to promote the ideal of the “pristine” data set, uniformly formatted and stripped of local idiosyncrasy.3 Making data appear “pristine,” on the one hand, makes it more easily usable, but it also perpetuates a sense of data as disembodied and unmarked by historical situation and human choice. In claiming that her work is “to reproduce a record,” Wells-Barnett calls attention to labor that the rhetoric of empiricism wishes to disavow. No record of reality, she suggests, simply exists.

. The lens of modernist aesthetics allows us to register these writers’ innovative engagement with data collection and data representation on the level of both content and form. I bring together the terms “data-driven” and “modernism” to suggest a literary critical framework for apprehending the historical, formal, and epistemological uses of data. In present usage, “data-driven” denotes a process of decision-making that purports to look to data for its direction and justification, implying that decisions can be made more effectively and without the danger of bias or misguided tradition by referring to data.5 I put a different twist on the term to call attention to how imagining reality as a vast data collection drives that imagination toward certain formal parameters. Data-driven modernists, in my definition, commit to the form of data collection first and then see what narratives result. In this way, they highlight how data complicates, rather than clarifies, our narrative representations of reality. I approach modernism from a historical as well as aesthetic perspective, following interpretive pathways opened by the work of scholars such as Andreas Huyssen, Susan Stanford Friedman, Douglas Mao, Rebecca Walkowitz, and Werner Sollors, who have challenged critics to consider how our methodological concatenation of text and context must shift when we loosen the association of modernism with a few representative works and seek the “effects of synergy or friction [that] result when the many, sometimes contradictory criteria of high modernism are tested against less evidently experimental texts” (Mao and Walkowitz, Bad Modernisms, 2).6 Considering historical context alongside aesthetic practice is especially important for critical consideration of race and ethnicity as factors shaping literary production, as the work of Joshua Miller, Sarah Wilson, and Leif Sorensen has modeled.7 Focusing on data as a representational strategy with aesthetic as well as epistemological implications brings into a view a friction-generating range of texts and creators, embedded in multiple disciplines and literary communities.

. Characterizing Wells-Barnett’s project as reproductive also highlights her challenge to longstanding concepts of modernism, a field long defined by valuation of authorial autonomy and innovation. Wells-Barnett’s writings, on first examination, seem to fail both requirements, being more or less traditional at the level of prose diction and clearly engaged in political advocacy, making it, to earlier generations of modernist scholars, one of the “at first quite unliterary promotions of feminism, socialism, nationalism, and other forms of social change” (Mao and Walkowitz, “New Modernist Studies,” 744). More recently, though, scholars have begun to recognize a strain of Black women’s modernist authorship that is characterized by reproductive labor. Teresa Zackodnik, examining the editorial practices of Jessie Fauset and Amy Jacques Garvey, has argued for recognition of the “black feminist press practice of recirculation” (439) as literary technique and political strategy. Fauset’s work as the literary editor of The Crisis and Jacques Garvey’s use of the “collage column” form in The Negro World demonstrate that reprinting others’ writing is not merely a labor-saving device or perfunctory editorial duty but an intentional, creative method. Their careful selection and layout of previously published material creates “dissonant juxtaposition” (440), demanding interpretation and thereby working to “effectively produce other politicized, critical, and conscious readers” as part of a widening network of Black thought. Zackodnik cites Wells-Barnett as an important precursor to this tradition.4

. Wells-Barnett’s work, while a form of recirculation, is also distinct in its reliance on data aesthetics. While Fauset and Jacques Garvey derive their primary impact from careful selection, Wells-Barnett emphasizes processes Page 148 →of collection aiming at exhaustivity. Formally, her writing embodies the characteristics of data-driven modernism with its aesthetics of collection and exhaustivity deployed to disrupt the white lynch narrative. Thematically, her writings theorize the role of data collection in social change and present the labor of circulation as crucial to any power it might have. Viewed as an example of data-driven modernism, her advocacy itself is intimately connected to the modernist representational project not only because it represents a critical encounter between race and modernity5 but also because its political intervention relies on a self-conscious relationship to narrative form.

. For Du Bois, encountering the news of Sam Hose’s lynching is figured as an eruption of irrational violence that casts his work and belief in empiricism into absurdity.6 For Wells-Barnett, close encounter with an act of lynching is the revelation of a crucial object of inquiry. The lynching of Moss, McDowell, and Stewart becomes data point number one in the collection that will define Wells-Barnett’s life and career. In response to their lynching, she “began an investigation of every lynching I read about” (Crusade, 64). In other words, she undertook the exhaustive collection of data to represent the reality of lynching. To register how her approach marks a significant shift toward data epistemology, it is useful to consider the other ways she might have represented the life—and death—stories of Moss, McDowell, and Stewart. She might have started a campaign based on the innocence and social worthiness of these three entrepreneurs and family men. She might have focused on their widows and orphans. Instead, she adopts the method of a data-driven modernist, undertaking exhaustive collection before forming any narrative representation.

. At the beginning of 1892, Wells-Barnett is a full time, self-supporting, nineteenth-century journalist. Historian of nineteenth-century journalism Hazel Dicken-Garcia suggests that the word “journalist” in this period be used to refer to “all those involved in gathering, preparing, and presenting newspaper content” (8). This description captures Wells-Barnett’s varied and intertwined roles throughout her writing career. She was undoubtedly a journalist, but what it meant to be a journalist was more defined by practice than credential. Even compared to the relatively young fields of sociology, psychology, and scientific history, journalism as a standardized discipline was in its nascency. The first university degree program was not launched until 1893, and the first widely published textbook was not Page 156 →released until 1894.7 Wells-Barnett had no formal training, but then again, neither did most working journalists. To say that she was a journalist does not imply that she ascribed to an explicit set of principles taught in a formal program, but rather points to the focus of her writing, the source of her livelihood, and her commitment to newspapers and print periodicals as vehicles for communal education. In this commitment, she was not alone. As Jean Marie Lutes notes in her survey of periodical studies in nineteenth-and early twentieth-century literature, “Between 1895 and 1915, more African American newspapers—some twelve hundred—were launched than in any other era of American history” (337). While formal training was not widely available, African American journalists, as documented by I. Garland Penn’s 1891 compendium, The Afro-American Press and Its Editors, formed a flourishing network of writers, editors, and publications. The Afro-American Press Association held an annual convention, which Wells-Barnett attended, and she also served a term as secretary to the Association.

. The turn toward data collection is also evident in the emergence of objectivity as a keyword of journalistic practice. Schudson writes, “‘Objectivity’ is at once a moral ideal, a set of reporting and editing practices, and an observable pattern of news writing’” (149). The 1890s mark the codificationPage 157 → of the “inverted pyramid” (Schudson, 149) as the standard form for writing news.8 The inverted pyramid form puts basic referential details in the first paragraph, often only a sentence long. The longer, more detailed paragraphs that may follow are subordinate to this seemingly straightforward recitation of primary facts. Events and people are taken out of historical context as the news is threshed through a template of “who/what, how, when, where?,” as Theodore Dreiser recalls having it presented to him in 1892 (52). This fact-forward form was a departure from prior practice. As David Mindich notes, “Until the end of the nineteenth century the telling of news nearly always took the form used in classical storytelling: first, an announcement of the utility or importance of the story . . . leaving the surprise, or what Aristotle called ‘Reversal of the Situation,’ for last” (64). The inverted pyramid exchanges temporal unfolding framed with moralistic interpretation for a set of summary facts. Mindich underscores that this formal change to “a system that appears to strip a story of everything but the ‘facts’” has epistemological consequences, “chang[ing] the way we process news” (65). Presenting human reality through discrete and concrete points of information, it facilitates faster, more superficial reading and acceptance of reported fact as transparent truth. Put another way, formally and epistemologically, the inverted pyramid form turns news into a kind of data collection, and it allows the appearance of objectivity to elide the still-present human interventions of selecting the scope of observation and language of description.

. The advent of the telegraph and wire service reporting also fomented the trend toward prose that conformed to an objective mode. The telegraph allowed for events reported in one location to be transmitted quickly enough to be printed in newspapers everywhere, omitting any context or analysis that would be deemed irrelevant to a national audience. Stripped of local detail, distinct events were represented using templated patterns, unfolding in roughly equivalent ways across time and space.9 These objectivizing tendencies form a background against which Wells-Barnett’s turn to data collection becomes legible as a critique of empiricism as performative, a critique executed by employing empiricist method more rigorously and therefore more radically.

. These narrative shifts in her own life are a direct consequence of her commitment to using data to disrupt the white narrative of lynching. The lynch narrative casts the Black male as “a rapist, bestial and savage,” the white male as “chivalric” hero, and the white female as “passive, virtuous, and terrified” (Piepmeier, 131).10 Although the lynch narrative was reiterated frequently across numerous media, it also asserted that lynching was fundamentally unpredictable and rare. As Wells-Barnett notes, lynch narrative presented lynching as “irregular and contrary to law and order,” provoked by “unreasoning anger over the terrible crime of rape,” and “justified” as an act of communal self-defense (Crusade, 64), insinuating that lynching was such an exceptional response to such an exceptional crime that no intervention could avail. The lynch narrative was so entrenched that according to Wells-Barnett, after the publication of her initial article in the New York Age, Frederick Douglass himself “came from his home in Washington to tell me what a revelation of existing conditions this article had been to him. He had been troubled by the increasing number of lynchings, and had begun to believe it true that there was increasing lasciviousness on the part of Negroes” (Crusade, 72). Joanne Braxton notes there is no direct evidence verifying this exchange and questions whether Douglass ever believed any change in African American character contributed to lynching. Even as a fabrication, though, it is telling that Wells-Barnett felt it rhetorically necessary to have this assessment come from the figure of Douglass. Douglass’s credulity serves as proxy for the narrative’s degree of acceptance as fact.

. Being able to “tell the whole truth” not only requires collecting data, it also requires a representational form that will go beyond the selective contours of personal testimony or argumentative essay and a format in which it can be circulated. With Southern Horrors: Lynch Law in All Its Phases, Wells-Barnett shifts from newspaper column to pamphlet, a form that she will go on to employ often throughout her writing. The pamphlet as format is integral to her ability to circulate her data collections and develop her own data aesthetic. In material terms, a pamphlet is simply a set of stitched or glued pages, commonly printed in a limited run at the behest (and usually the expense) of the author. Because they are simply printed paper, pamphlets are comparable to newspapers in cost and speed of production, but because they are bound and typically devoted entirely to one essay or subject, they also function to “preserve words and deeds in a discrete, individual, and long-lived object” (Rael, Newman, and Lapsansky, 2). Pamphlets hold a special role in U.S. print culture as a primary means of disseminating protest writing, with Thomas Paine’s Common Sense being perhaps the most famous example, but there was also a robust tradition of African American pamphleteering.11

. Wells-Barnett’s first pamphlet, Southern Horrors is published in late 1892, using funds raised by African American community leaders Victoria Earle Matthews and Maritcha Lyons.12 It includes a preface by Frederick Douglass. In their prefatory materials, both Wells-Barnett and Douglass work to recast the African American witness as data collector. They emphasize Wells-Barnett’s narratorial position as one who gathers and presents. Wells-Barnett describes the work as “a contribution to the truth, an array of facts” (1). “Array” connotes collection, an ordered multiplicity of coequal points brought together to serve some purpose. While the data collector as active arranger may seem to conflict with the figure of the data collective as receptive observer as I have discussed in prior chapters, collecting data does not mean one refrains from ordering it. Data collections, through the sorting of different metadata fields, are constantly ordered and reordered to pursue a line of inquiry.13 No matter what acts of ordering follow, data collectors maintain primary commitment to collection. Douglass’s prefatory description of Wells-Barnett’s work in Southern Horrors also acknowledges her data collection as a distinctive mode of authorship. He observes, “You [Wells] give us what you know and testify from actual knowledge. You have dealt with the facts with cool, painstaking fidelity and left those naked and uncontradicted facts to speak for themselves” (3). Rather than emphasizing her style or her story, he emphasizes her role as a conduit for the circulation of testimony. He also notes that this testimony takes a particular form: the putting forth of “naked and uncontradicted facts” (3). Douglass’s turn to the commonplace of facts “speak[ing] for themselves” can be seen as a savvy deployment of data rhetoric, legitimizing Wells-Barnett’s testimony by marking it as an objective, impersonal mode of authorship not typically accorded to women of color. Declaring that “there has been no word equal to it in convincing power” (3), Douglass tacitly endorses data collection as a liberatory practice in Wells-Barnett’s hands.

. Wells-Barnett’s first pamphlet, Southern Horrors is published in late 1892, using funds raised by African American community leaders Victoria Earle Matthews and Maritcha Lyons.12 It includes a preface by Frederick Douglass. In their prefatory materials, both Wells-Barnett and Douglass work to recast the African American witness as data collector. They emphasize Wells-Barnett’s narratorial position as one who gathers and presents. Wells-Barnett describes the work as “a contribution to the truth, an array of facts” (1). “Array” connotes collection, an ordered multiplicity of coequal points brought together to serve some purpose. While the data collector as active arranger may seem to conflict with the figure of the data collective as receptive observer as I have discussed in prior chapters, collecting data does not mean one refrains from ordering it. Data collections, through the sorting of different metadata fields, are constantly ordered and reordered to pursue a line of inquiry.13 No matter what acts of ordering follow, data collectors maintain primary commitment to collection. Douglass’s prefatory description of Wells-Barnett’s work in Southern Horrors also acknowledges her data collection as a distinctive mode of authorship. He observes, “You [Wells] give us what you know and testify from actual knowledge. You have dealt with the facts with cool, painstaking fidelity and left those naked and uncontradicted facts to speak for themselves” (3). Rather than emphasizing her style or her story, he emphasizes her role as a conduit for the circulation of testimony. He also notes that this testimony takes a particular form: the putting forth of “naked and uncontradicted facts” (3). Douglass’s turn to the commonplace of facts “speak[ing] for themselves” can be seen as a savvy deployment of data rhetoric, legitimizing Wells-Barnett’s testimony by marking it as an objective, impersonal mode of authorship not typically accorded to women of color. Declaring that “there has been no word equal to it in convincing power” (3), Douglass tacitly endorses data collection as a liberatory practice in Wells-Barnett’s hands.

. Page 7 →I use the analytic of narrative in this project in three senses: as a traditional literary form, the realism of which is being challenged and re-formed during this period and which individual texts may be read to seek, resist, and/or revise; as an intervention in the representation of lived experience through which we can read for ethical arguments and engagements; and as an act of self-representation that not only represents but, in the minds of many critics and theorists, also constitutes self.8 Because narrative is a kind of epistemological model for reality, the effects of the formal shifts provoked by data collection as a mode of representing the world are not solely aesthetic. As Michael Elliot notes, “narratives do more than place events into chronological sequence; they arrange those events according to patterns of causation in a way that enables the author and reader to create order out of the chaos of everyday life” (xxiii). Narrative presents an epistemological model for reality, implicitly staking claims about causation and relationship. In this way, the status of narrativity in life writing is a register of concepts of self, agency, and social order. I demonstrate the epistemological effects of conceptualizing lives as data by reading the formal effects data entails for life narrative.

. Describing her “unvarnished account” as “an array of facts,” Wells-Barnett shifts the formal terms of autobiographical text and the subjectivity traditionally accorded to African American autobiographical narrators. Page 164 →Claiming the position of the data collector is a crucial intervention in the constraints placed upon African American narration to that point, most visibly evidenced in slave narratives. As scholars have pointed out, the slave narrative, typically commissioned and published by white abolitionists, is bound to a performance of authenticity in exchange for the hope of political efficacy. Acceptable metrics of authenticity shift over time,14 but all require the discursive “carting out of black bodies onto the stage to bear witness to their authentic experiences of slavery” (McBride, 4). Thus, the possibility of a Black person’s testimony was premised on the (re)sacrifice of the body and assent to be seen primarily as a body, an object upon which others would gaze for their own edification. By claiming that her collecting work forms a “true, unvarnished account” (Southern Horrors, 1), Wells-Barnett articulates her narratorial position as that of the data collector, downplaying her own intervention and holding herself up as an impartial conduit. This links her method to the tradition of the autobiographical slave narrative, constructed, received, and valued as evidence against the practice of enslavement. Applying the value of a “true, unvarnished account” to a data collection rather than a more typical autobiographical account extends, rather than sacrifices, her epistemic authority by allowing her to marshal data beyond her personal experience. By decentering her own experience in this way, she claims the role of the data collector, distancing herself from the stance of the victim. Yet, there is no question that her own experience is in the data set of white violence against African Americans, so the role of the data collector is also repositioned as a grassroots observer. Southern Horrors is a re-formation of the autobiographical as one data source in a text that oscillates between point and collection, self and collectivity.

. Chapters three through six slice15 this base data collection along the facets of different alleged offenses and types of victim to drive home the multiplicity of alleged offenses and the barbarity of the lynchings that follow. Chapter two’s list format emphasizes the sheer quantity of allegations that were seen as an excuse for lynching, but it also lulls perception of the vast differences between these allegations and the fact that lynch victims had rarely been tried in court, let alone proven guilty. Each chapter’s dive into a specific facet of the data set allows a different dynamic to emerge in focus. For each chapter, Wells-Barnett collects and reproduces additional data, most often in the form of newspaper coverage and sometimes in the form of correspondence with local investigators. Chapter three, “The Lynching of Imbeciles,” for example, documents two cases in which the victims had a history of mental disturbance. These chapters continue to use aesthetics of data collection to construct their alternate narrative of lynching, specifically through extensive quotation of news accounts, emphasizing collection over selection, and the use of re-categorization as epistemological intervention.

. Page 183 →While data collection might seem like a method of representation confined to knowing the past, Wells-Barnett’s final published pamphlet, The Arkansas Race Riot, demonstrates data’s potential role in understanding how white violence ramifies across the futures of African American communities. The pamphlet’s opening sentences, cited above, waste no time in establishing that the harmful representational dynamics she fought in her earlier work still obtain. The “press” has the power to circulate an account of events in Elaine, Arkansas that cast them in the predictable narrative of white self-defense against African American violence. Wells-Barnett travels to Elaine and investigates for herself. She learns that the “riots” began when white men attacked a meeting of African American tenant farmers who sought to form a chapter of the Farmer’s Union. As a group, they planned to refuse to sell their bumper cotton crop to their white landlords and instead take it directly to market. While there are enduring historical questions about what happened, including the actual number of African Americans who were killed during these riots, historians have established that the reality is much closer to what Wells-Barnett reports than what the white press reported at the time.16

. As the QS practitioners cited above demonstrate through their imagination of self as discrete bits, selfhood as assemblage is not just a formal effect or theoretical potential. It is both technological reality and lived experience. The conceptualization of self as parts and emergent wholes is, arguably, a part of the epistemological shift enacted by a data-driven perception of the world but also certainly a byproduct of ubiquitous networked computing, filled with technologies that enable and promote mass collection of life data. This, of course, is also known as surveillance. Some of this collection proceeds actively, through more or less formal instruments like surveys or job applications, and an increasingly vast amount operates passively, capturing traces of our digital actions as a matter of course. Further, as Kevin Haggerty and Richard Ericson have described, the stores of data collected about us are tending toward convergence, becoming accessible to each other across “centres of calculation” such as “forensic laboratories, statistical institutions, police stations, financial institutions, and corporate and military headquarters” where the discrete elements of a life captured by data are “reassembled and scrutinized in the hope of developing strategies of governance, commerce and control” (613).1

. Page 192 →I have so far focused on and argued for continuities between the collection of data in the period of this study and today, but at this juncture, it is important to note the difference that digital, networked affordances make in the ramifications of data collection. The writers of this study collected data in analog formats—surveys, documents, lab notes, anatomical drawings, newspaper clippings—and they did so actively. Today, the data of our lives is collected mostly in digital format and mostly passively, as a record of some type of digital interaction. Because it is digital, it is much easier to aggregate, although certainly not effortless. The potential for aggregation is a constant threat to privacy, because even data that is functionally anonymized in one collection could be reidentified when joined with another.2 Although we might think our data is partitioned, with some actors only having access to our Internet browsing habits and others only having access to medical records, those boundaries can become permeable through official (e.g., cross-institutional collaboration, subpoena) and unofficial channels (e.g., data breaches). Most fundamentally, and perhaps most ironically, the epistemological framework of Big Data has, in many contexts, altered the definition of useful knowledge. Eric Siegel notes that when using predictive analytics, “we generally don’t have firm knowledge about causation, and we often don’t necessarily care. . . . It just needs to work; prediction trumps explanation” (120). Where Bacon once sought prediction through the clarity of universal laws of nature, Big Data shrugs at explanation, satisfied with the instrumental uses of correlation.

. Data driven modernists illuminate the space between data and narrative to remind us that there are always multiple ways to move across it. Formally and thematically, their modes of narrating with data constitute interventions in the conflation of data and narrative. These narrative interventions anticipate and intersect with the flourishing work of critical data studies,3 Black data theory,4 queering data,5 Indigenous statistical methods,6 and feminist data practice.7 In closing, I wish to briefly underscore the primary intervention modeled by each of the writers I have examined. To some extent all of these writers employ all of these interventions, but some more centrally than others. Their writing of critical data narratives suggests strategies for becoming critical data readers. They expand the scope with care, expect multiplicity, ask who is narrating, and see data as labor.

. Data driven modernists illuminate the space between data and narrative to remind us that there are always multiple ways to move across it. Formally and thematically, their modes of narrating with data constitute interventions in the conflation of data and narrative. These narrative interventions anticipate and intersect with the flourishing work of critical data studies,3 Black data theory,4 queering data,5 Indigenous statistical methods,6 and feminist data practice.7 In closing, I wish to briefly underscore the primary intervention modeled by each of the writers I have examined. To some extent all of these writers employ all of these interventions, but some more centrally than others. Their writing of critical data narratives suggests strategies for becoming critical data readers. They expand the scope with care, expect multiplicity, ask who is narrating, and see data as labor.

. Data driven modernists illuminate the space between data and narrative to remind us that there are always multiple ways to move across it. Formally and thematically, their modes of narrating with data constitute interventions in the conflation of data and narrative. These narrative interventions anticipate and intersect with the flourishing work of critical data studies,3 Black data theory,4 queering data,5 Indigenous statistical methods,6 and feminist data practice.7 In closing, I wish to briefly underscore the primary intervention modeled by each of the writers I have examined. To some extent all of these writers employ all of these interventions, but some more centrally than others. Their writing of critical data narratives suggests strategies for becoming critical data readers. They expand the scope with care, expect multiplicity, ask who is narrating, and see data as labor.

. Data driven modernists illuminate the space between data and narrative to remind us that there are always multiple ways to move across it. Formally and thematically, their modes of narrating with data constitute interventions in the conflation of data and narrative. These narrative interventions anticipate and intersect with the flourishing work of critical data studies,3 Black data theory,4 queering data,5 Indigenous statistical methods,6 and feminist data practice.7 In closing, I wish to briefly underscore the primary intervention modeled by each of the writers I have examined. To some extent all of these writers employ all of these interventions, but some more centrally than others. Their writing of critical data narratives suggests strategies for becoming critical data readers. They expand the scope with care, expect multiplicity, ask who is narrating, and see data as labor.

. Data driven modernists illuminate the space between data and narrative to remind us that there are always multiple ways to move across it. Formally and thematically, their modes of narrating with data constitute interventions in the conflation of data and narrative. These narrative interventions anticipate and intersect with the flourishing work of critical data studies,3 Black data theory,4 queering data,5 Indigenous statistical methods,6 and feminist data practice.7 In closing, I wish to briefly underscore the primary intervention modeled by each of the writers I have examined. To some extent all of these writers employ all of these interventions, but some more centrally than others. Their writing of critical data narratives suggests strategies for becoming critical data readers. They expand the scope with care, expect multiplicity, ask who is narrating, and see data as labor.

. Provoking critical awareness of the gap between data and narrative is a project that has only become more urgent a century or so after these modernist experiments. Data is9 now constantly being used to narrate our lives. Categorizing algorithms draw from amassed personal data to assign narrative destinies to individuals at crucial junctures, simultaneously predicting and shaping the path of their lives’ unfolding. Data is commonly assumed to bring us closer to objective knowledge of reality, perhaps relieving the need for human interpretation altogether if we can only collect enough. In 2008, Chris Anderson, a writer for Wired magazine, went so far as to assert, “With enough data, the numbers speak for themselves” (108). But the narrative paths categorizing algorithms have assigned to human lives in areas ranging from public policy to personal finance seem, more often than not, to replicate biases about who an individual is and could become.10 It has become increasingly clear that, as John Cheney-Lippold cautions, “Who we are in terms of data depends on how our data is spoken for” (We Are Data, 48). The question of who we are hinges not on what data says, but on who is granted the authority to speak for it and what forms of representation they employ. The data-driven modernists that populate the juncture between life, data, and story model critical data narrative in ways that prepare us to challenge the claims of algorithmic identification and demand alternative modes of narrating our collected lives.

. We who practice literary studies might also see and seize upon the proliferating proposals for methods of reading in the twenty-first century—such as distant reading, surface reading, hyper reading, and machine reading8—as attempts to define literary scholarship and humanist studies not just in reaction to a cultural climate that accepts too credulously the allure of the quantitative,Page 198 → but also in relationship to the new awareness that the data of literature provokes: the reality of a vaster textual domain than one reader, one method, one algorithm, or any number of canons could ever account for. Important work arises from engaging with this data as data. Important work also arises from troubling the perceived tenability of knowing something about whole centuries or national corpora of literature through data by raising awareness of archival exclusions and the limits of encoding.

. Approaches to representing COVID-19 data display the same striking oscillation between point and collection employed by the writers examined in this study. The May 24, 2020, front page of the New York Times was covered by a list of names, followed by age, hometown, and a one-sentence obituary. This list of nearly a thousand names was roughly one percent of the lives lost to COVID-19 in the United States at that time. On Sunday, February 21, 2021, the front page was again dominated by an attempt to represent both the scope and particularity of loss in a one-dot-per-death vertical timeline to represent the nearly 500,000 lives lost to COVID-19 by that point. Yet, at the same time as efforts to regain a sense of scope by reminding us of the sheer number of individuals lost relied on list and dot forms, certain clear narrative lines were emerging from the collected points that should also have been able to move us, to mourning and to action. Indigenous, Black, Asian Pacific Islander, and Latino/a communities in the United States were, and as of this writing are, experiencing a death rate of double or more than that of whites and Asian Americans.9 Our understandings and actions must constantly shift between the formal frames of point, collection, and narrative if we are to realize data’s epistemological potential to improve not only our understanding of reality but our support for lives, human and nonhuman, within it. As well, we must reckon with the historical and political contexts that make the processes of turning data into meaning into just action anything but frictionless.