Katherinebode.files.wordpress.com



PostprintCitation: Katherine Bode. 2021. “Data Worlds: Patterns, Structures, Libraries.” In The Cambridge History of World Literature Volume 2, ed. Debjani Ganguly. Cambridge University Press, 765–785. Data Worlds: Patterns, Structures, LibrariesBy Katherine BodeThe emergence of world literary studies and the digital turn in literary studies share a seminal text in Franco Moretti’s “Conjectures on World Literature.” Since that article was published in 2000, the ongoing digitization of cultural collections and emergence of new computational methods of analysis have only increased the excitement of some, and concern of others, that digital research will become a major, even a dominant, trajectory in world literary studies. But the associated claim – by critics and proponents alike – that digital approaches represent a paradigmatic shift ignores important continuities between digital literary research and the non-digital scholarship that precedes and persists alongside it, including in world literary studies. Much of the traction that Moretti’s “distant reading” has gained can be ascribed to such continuities, including in the conception of archives, attention to networks and patterns, and orientation to national and canonical frameworks. Pre-digital traditions, ideologies, and infrastructures also shape digital resources and methods in substantial and influential ways. Rather than breaking with literary traditions, the more pressing problem with digital literary studies might be its tendency to sustain past hierarchies and structures of knowledge, including those that have relegated non-Western literary cultures and communities to the margins of academic debate. Recognizing and interrogating this inheritance, an emerging group of digital projects are beginning to advance what Vanessa Smith describes as some key aims of world literary studies, including situating literature “within a vast transnational library,” expanding “the canon of the literary properly to reflect global diversity,” and offering new conceptual frameworks that “adjust our reading of the novel to world scale” (2016: 92).As most readers of this collection will know, “distant reading” – a term central to digital humanities – began its life as a “new critical method” for world literary studies. In his “Conjectures” essay, Moretti rejected the reliance of world literary studies on close reading, arguing that this method meant the field could only consider a small canon of literary works from a limited segment of the world literary system: from “Western Europe, and mostly revolving around the river Rhine … Not much more” (2000: 54). To solve this problem Moretti proposed that literary scholars look to “distance” as “a condition of knowledge … to focus on units that are much smaller or much larger than the text: devices, themes, tropes – or genres and systems” (57). These units were to be derived from the observations of scholars in multiple national fields, offering the basis for investigating a world literary system that, in Moretti’s analysis, constituted an unequal relationship between core and periphery literary cultures. Moretti defends the loss of the text, in such investigations, as a necessary precondition to explore that world literary “system in its entirety” (57).One thing that the “Conjectures” essay does not do is to advocate for computational or digital approaches to literature. This connection of “distant reading” to literary data, computational methods, and mass-digitized collections has instead emerged over time, in Moretti’s subsequently delineations of the method. Graphs, Maps, Trees: Abstract Models for Literary History, his 2005 book, translates the observations of other scholars into literary data and creates a link between the resulting analyses and world-systems theories by arguing that the history of literature “cannot be understood by stitching together separate bits of knowledge about individual cases, because it isn’t a sum of individual cases: it’s a collective system, that should be grasped as such, as a whole” (4). His 2013 book, Distant Reading, foregrounds digital methods, such as computational text and network analysis, as well as the digitized bibliographies, libraries, and archives that are increasingly substantial components of the knowledge infrastructure for literary studies. Moretti highlights the importance of digital infrastructure in reforming literary studies by noting that, where the field has previously experienced “the rise of quantitative evidence … without producing lasting effects … this time is probably going to be different, because this time we have digital databases and automatic data retrieval” (212).While many literary scholars have enthusiastically embraced the possibilities of “distant reading” (for example, Goodwin and Holbo 2011), others have rejected the approach. World literary studies scholars have been among the most prominent critics of Moretti’s method, in both its non-digital and digital forms. Katie Trumpener proposes that “distant reading” “violate[s] the individuality” and aesthetic qualities of literary works (2009: 160). Arguing that world-systems theory equates “economic with cultural systems,” Gayatri Spivak describes “distant reading” as replicating, in the scholarly arena, the core/periphery dynamics of that system by encouraging scholars at powerful, central sites, such as the United States, to amass and process close readings by “native informants” from predominantly non-anglophone literary cultures (2003: 108). As “distant reading” has become increasingly aligned with digital humanities, these earlier criticisms have been supplemented by claims that data-rich methods are unable to support new discoveries for literary studies and contribute to the corporatization of contemporary universities (Marche 2012; Kirsch 2014; Allington, Brouillette, and Golumbia 2016).Among the earliest critiques of digital approaches to comparative and world literary studies is Haun Saussy’s lead essay to the 2004 report of the American Comparative Literature Association, on the discipline in the context of globalization. Saussy argues that literature and reading – closely and carefully in the original language – are fundamentally opposed to contemporary flows of information and the informationist discourse it introduces. As “relic[s] of an earlier, data-poor, low-bandwidth era of communications … when details mattered” (2004: 32), literature and reading “frustrate the economy of information in which more data and faster access is always better” and thus offer a form of “internal resistance” to “information’s charms” (33). By contrast, the “world according to Google,” though “vast (and getting vaster all the time, now that whole libraries are being scanned into its database) and instantly searchable all the way to its farthest recesses,” offers a “flat” “intellectual landscape” (32). For Saussy, the digital context enables only “a positivist style of reading,” one that limits us to asking questions “of preestablished categories narrowing down to preestablished subcategories” (33). What Saussy’s chapter makes especially clear is how rejections of “distant reading” – whether as a critical or specifically a digital approach – consistently conceive of it as a fundamental break from earlier paradigms and practices.Yet in important ways, Moretti’s approach builds on and resonates with trajectories in non-digital research, including in world literary studies. The expansionist, informationist discourse that Saussy criticizes as an effect of digital technologies can be perceived in intellectual and political reconfigurations of the archive in the latter part of the twentieth century, well before Google became a company let alone a verb, and definitely prior to the widespread use of digital resources and methods for humanities research. The insight, by Marxist, feminist, and postcolonial studies from the 1960s, that literary scholarship had privileged certain voices, raised the profile of archival research and ascribed moral authority to the practice of extending the canon by studying and teaching new, particularly working-class, women, and non-white, authors. New historicist scholarship refined and strengthened this focus on the archive in emphasizing the “textuality of history and the historicity of texts” (Montrose 1992: 410). The subtle shift in emphasis from incorporating new actors and voices into literary and other histories to also – or especially – reimagining the nature and scale of the archive was reinforced and progressed by historical research oriented to transnational spaces and dynamics. For instance, describing the emergence of New Imperial History in the 1980s, Tony Ballantyne highlights growing recognition of the mobility of colonial knowledge and how this motivated a shift in focus from “enclosed, static, and discrete” archives – typically limited to a particular place and time – to a transnational archive conceived as “the product of the constant circulation of information and the heavy intertextuality of many forms of knowledge” (2003: 113).This conception of archives as extensive and interconnected sites of information flow – and the orientation it produces to phenomena such as networks and patterns – has been influential in non-digital as well as digital approaches to world literary studies. Although she criticizes Moretti’s world-systems approach, Wai Chee Dimock conceives of American literature as a networked configuration: not “a discrete entity” but “a much more complex tangle of relations … a crisscrossing set of pathways, open-ended and ever multiplying, weaving in and out of other geographies, other languages and cultures” (2006: 3). Roland Greene perceives the network as the implicit frame of reference in world literary studies, citing as evidence for this claim the focus on “networks” rather than “works” in David Damrosch’s as well as Moretti’s writing (2004: 214). While networks attain an obvious, concrete form in the visualizations that feature in some of Moretti’s experiments (for example, Distant Reading [2013: 213]), Greene argues that even in Damrosch’s studies of individual works, conceptions of literary value and meaning have become inseparable from a systemic view of literature and its connection to an international or transnational archive. Both authors also use patterns in flows of information – or of literary works, people, or ideas – to discern the latent meaning of the networks comprising the world literary system. Accordingly, in What Is World Literature? (2003) Damrosch foregrounds patterns in the modes and effects of translation of individual works of world literature, while in “Patterns and Interpretation” (2017) Moretti does the same with respect to word frequencies derived from hundreds of literary works. Both critics and proponents of “distant reading” stress its departure from non-digital approaches. Yet the impact of Moretti’s arguments is attributable, at least in part, to the equivalent importance of extensive archives and the relationships and repetitions they bring into view in literary and historical scholarship in the latter twentieth century in general, not least of all in world literary studies.While the turn to expansive archives might be a feature of both digital and non-digital scholarship, the capacity of digital methods to explore such archives in new ways, and specifically at new scales, is often taken to mean that digital approaches will inevitably challenge traditional – particularly national and canonical – approaches to literature. At least in the range of national literatures explored, some of Moretti’s experiments achieve his stated aim that “distant reading” be “a thorn in the side, a permanent intellectual challenge to national literatures” (2000: 68). Graphs, Maps, Trees opens with the rise and fall of the novel across a range of national contexts, including Britain, Japan, Italy, Spain, Nigeria, Denmark, France, and India, and Distant Reading reprises an earlier investigation of the international movement of American films. Yet British novels are the focus of most of Moretti’s “distant readings,” including analyses of trends in gender and genre and of formal features of detective and village stories in Graphs, Maps, Trees, and in studies of novel titles and dramatic character relationships in Distant Reading. Even his work on multiple national contexts frequently reinforces national boundaries. As I have argued elsewhere, for instance, in discussing the rise of the novel in Britain, Japan, Italy, Spain, and Nigeria, Moretti assumes that readers in those nations read only novels by authors of the same nation (Bode 2017: 90). With some important exceptions discussed at the end of the chapter, digital literary studies echoes the national orientation of Moretti’s scholarship, with most large-scale literary histories focusing on literary works by authors of individual, predominantly anglophone, nations, including America (Wilkens 2013, 2016), Australia (Bode 2012), England ( HYPERLINK \l "c039_r003" Bamman, Underwood, and Smith 2014), and Germany ( HYPERLINK \l "c039_r024" Erlin and Tatlock 2014).The focus on canonical authors and texts is another area where the framing of digital scholarship is at odds with its practice. While the expanding digital archive is repeatedly described as standing in contrast to the literary canon, most digital editorial work concerns canonical – predominantly white, male, British or American – authors and their works, including William Blake (Eaves, Essick, and Viscomi 2020), Emily Dickinson (Morris 2012), Dante Gabriel Rossetti (McGann 2008), William Shakespeare (Mowat et al. n.d.), and Walt Whitman (Folsom and Price n.d.). In Roopika Risam’s words, “[s]ome of the most developed digital humanities work” is devoted to “preserve the writing of dead white men, specifically individuals unlikely to be forgotten in Anglophone literary history even if these projects did not exist” (2015). Although a number of early digital projects and collections focus on women’s writing, including the Women Writers Project (1999–2016) and The Orlando Project (Brown, Clements, and Grundy, n.d.), Shawna Ross argues that such work has declined as a proportion of digital humanities scholarship since the 2000s, with a concomitant rise in a gendered construction of the field that emphasizes “the coolness of one’s tool, the bigness of one’s data, or the goodness of one’s intentions” (2018: 212). Many computational analyses of literature likewise consider canonical writers, including Jane Austen (Burrows 1987), Emily Dickinson ( HYPERLINK \l "c039_r065" Plaisant et al. 2006), and again, Shakespeare (Hirsch and Craig 2008). And while many of Moretti’s earlier experiments in “distant reading” reach beyond the so-called one percent of the canon, much of his recent work has a canonical emphasis. For instance, his investigation of dramatic character networks in Distant Reading considers Shakespeare’s Hamlet (2013: 211–29): arguably the most canonical play of the most canonical playwright. Again, the issue is not so much that some digital projects explore canonical authors. Rather, it is that the rhetoric of newness and disruption surrounding digital resources and methods directs attention away from such continuities and in doing so obscures the possibility that digital projects might be critiqued not for departing from non-digital approaches, but for replicating, perpetuating, and even solidifying earlier research paradigms.This disjunction between rhetoric and practice in digital approaches to literary canons brings to mind Damrosch’s description of the existence of a “hypercanon” in an ostensibly “postcanonical” age; and indeed, Damrosch’s account of pre-digital scholarship suggests some of the key reasons for the continuing preoccupation with canonical and national literatures in a supposedly globalized, distributed, and democratic digital world. Locating the emergence of this phenomenon in the 1980s, Damrosch defines the “hypercanon” as the tendency for traditionally major authors to hold their position or gain ground despite the widespread characterization of literary studies as expanding or deconstructing the canon. Damrosch ascribes the “hypercanon” predominantly to ideological and social causes. Not only do we inherit a way of thinking about literature in terms of individual, major authors, but these authors provide the point of connection between scholars: a shared reference that becomes even more essential as authors of the “counter canon” are shared only by specialist subsets within literary studies, and as authors of the “shadow canon” are known only to older scholars (Damrosch 2004).Similar social and ideological causes underpin the focus of digital research on national, predominantly anglophone, literatures and canonical authors. For scholars who began their research careers in literary fields largely delineated by national boundaries, the questions they know to ask of literature, with or without digital methods, are frequently those articulated in long-standing debates regarding national literary traditions. Likewise, the dominant way of thinking about literary history – as, in William St. Clair’s words, a “parade of authors,” with canonical authors filing past the commentator’s box in chronological order, taken as representative of the historical period in which they wrote (2004: 4) – has remained highly influential, even as technologies for accessing and analyzing literature have changed. Skye Bianco, for instance, argues that one of the ways in which digital literary studies projects make themselves “disciplinarily legible” to institutions and funding organizations is by invoking “an older humanism, a retro-humanism” precisely by focusing on national and/or canonical literatures (2015: 101).This orientation to national and canonical paradigms is also informed by institutional and infrastructural issues that preceded, but now profoundly shape, digital scholarship. The vast majority of the digital bibliographies and text collections that researchers use to pursue digital research are derived from pre-digital collections, typically those of university libraries and cultural institutions. Any wholesale or selective digitization from this pre-digital record is thus inevitably enacted in relation to the historically contingent notions of value, meaning, and purpose that informed earlier collecting practices. Put simply, because university and other institutional collections are more likely to hold literature by canonical authors and from prominent national literary traditions, these same literary categories will be more likely to characterize digital collections created on their basis. Even the largest digital libraries, such as Google Books and HathiTrust, often understood as global literary records, are predominantly based on the holdings of a relatively small and select group of American and British university libraries. These large digital libraries thus embed the particular cultural perspectives and linguistic orientation of those national contexts, and of their elite universities, into the infrastructure used for digital scholarship. Beyond book digitization programs, most mass-digitized collections arise from anglophone or European contexts: for instance, the largest digital historical newspaper collections, Chronicling America (Library of Congress), the British Newspaper Archive (British Library and findmypast), Europeana (Europeana), and Trove (National Library of Australia), represent American, British, European, and Australian periodicals, respectively. As ?lika Ortega writes, “Like their analog counterparts, digital archives impose forms of discursive authority, shape their reading and navigation, grant access to their holdings and obfuscate their deficiencies, give some rein to peruse them freely, but ultimately shape the knowledge that can be extracted out of them” (2018: 233).The issue of infrastructure is not new to world literary studies. In his account of the institutional separation of world literary studies from comparative literature in American universities in the 1990s, Damrosch notes that the intellectual merits of the respective approaches were foregrounded. Because they studied literature in the original languages, comparative literature programs in Ivy League colleges were held up as more prestigious and intellectually rigorous than programs of world literary studies in wheat-belt institutions, which read literature in translation (2013: 158). Yet the divide was also a matter of infrastructure: it was the extensive library holdings of the wealthier institutions that facilitated the study of literature in the original language, whereas the less wealthy colleges could only support the study of literature in translation. These infrastructural challenges have changed somewhat in the digital age. One of the things that digitization is most routinely celebrated for is overcoming such inequalities of access; and open-access resources allow students and scholars from rich and poor universities to view, for instance, ancient texts in the original language (Crane n.d.). Yet large – and arguably, increasing – sections of the digital archive are behind paywalls.Certain features of digital collections, and the methods used to explore them, are also likely not only to compound the selection effects arising from their basis in historically constituted collections, but to do so in ways that specifically foreground anglophone authors, works, and national literatures. Such features include algorithms for sorting search results, which typically return to the top of such results the authors and titles that are searched for most often, with the consequence that well-known aspects of the literary historical record are likely to become more well known, while lesser-known aspects are likely to become more obscure. A range of past and present issues that make certain texts more suitable for digitization than others – ranging from historical unevenness in the global distribution of printing technologies to contemporary copyright laws – have made the “nineteenth and early twentieth century Anglophone world,” in Lara Putnam’s words, “ground zero of digitization” (2016: 389). Methods for computational text analysis are usually adapted from software created for industrial or scientific purposes, and the dominance of the English language in these contexts means that tools for digitizing and analyzing documents with non-Latin scripts lag behind those for European languages. Considerable technical challenges are involved in applying computational methods to the few multilingual corpuses that exist ( HYPERLINK \l "c039_r004" Bergenmar and Leppanen 2017: 234). As Risam notes, due to aggressive forms of identity politics operating on the World Wide Web, “the digital cultural record not only must contend with the colonial hangovers from the cultural record but also [with] the forces that are actively constructing its medium as a hostile environment where universities and the academy are under threat” (2017). For all of these reasons, the historical beneficiaries of colonialism are further benefited by contemporary cultural, political, economic, and technological conditions which make it more likely for literature from those countries and regions to be the focus of digitization and digital scholarship.If, as Greene argues, the aim of world literary studies is “to construct a politics of comparison that does not borrow the inequalities and disproportions of the colonial world itself” (2004: 221), then the digital remediation of the literary record arguably makes that aim more difficult to achieve in embedding and amplifying the presence of canonical authors, national literatures, and elite institutional power within new knowledge infrastructure. Yet in this sense, at least, instead of predetermined categories and subcategories being phenomena that are instituted by digital technologies, as Saussy suggests, that organization of knowledge is often inherited from pre-digital scholarship and the views of the world, and of literary value, that it embodies. The main difference is that those categories, instead of being wrapped in a liberal humanist discourse of global equality, now come packaged in techno-utopian claims of extensive access and democratization.Despite the constitutive force of this disciplinary, institutional, social, and infrastructural inheritance, digital projects are beginning to explore literature in ways that extend beyond dominant anglophone, canonical, and/or national frameworks. An increasing number of digital collections represent literary works from non-anglophone regions, at least partly in response to and in order to redress the anglophone emphasis and cultural exclusions of the largest digital libraries ( HYPERLINK \l "c039_r035" Jeanneney 2006). Such enterprises include the pan-Hispanic collection Biblioteca Virtual Miguel de Cervantes, which by 2006 had digitized over 22,000 works; the French digital library Gallica, with over 70,000 volumes digitized; and the Digital Library of India (Ministry), with approximately 7 million items. The World Digital Library (Library of Congress) has books, manuscripts, maps, and other primary materials in over a hundred languages and provides search functions, metadata, and data for all records in seven languages, including Arabic, Chinese, and English. Digital libraries have also been – or are being – created for other non-anglophone countries including Brazil, Denmark, Finland, France, Germany, Hungary, Japan, Korea, the Netherlands, and Singapore. In terms of scale, even the largest of these non-anglophone digital libraries is dwarfed by the predominantly anglophone Google Books (which has digitized over 130 million titles). Yet reports on digital literary studies in non-anglophone regions, including Scandinavia and Mexico, suggest that scholarship conducted on the basis of these collections is more likely to emphasize non-canonical authors and literatures and nondominant social groups (see Bergenmar and Leppanen 2017; Ortega 2018).Some editorial projects create collections specifically to explore how literature traveled across multiple national contexts. Among the earliest, and best known, of these projects is Stanford University’s Mapping the Republic of Letters (2008), which explores networks of correspondence during the Enlightenment, including of literary figures such as Francesco Algarotti and Voltaire. Influential in demonstrating the capacity of data visualization to contribute to humanities research, the project offers new perspectives on the nature of Enlightenment communities (showing, for example, that they were less cosmopolitan than the participants liked to claim) while explicitly acknowledging the limitations of its underlying data (for instance, noting that only 10 percent of Voltaire’s correspondences have been digitized).More recent editorial projects, including the Bodmer Lab’s A Digital World Literature and the University of Chicago’s Philologic, create digital environments for exploring literary works from multiple national and/or linguistic contexts. While the former is based on the “library of world literature” amassed by Martin Bodimer, and privileges a carefully curated group of canonical authors, the latter takes a much broader approach, enabling searches across fifty databases with many thousands of literary works in multiple languages, including Ancient Greek, Catalan, Chinese, English, French, German, Italian, Japanese, Korean, Latin, Portuguese, and Spanish. Despite concern that digital research into women’s writing is in decline, new projects in world literary studies, including Swedish Women Writers on Export in the 19th Century (Leffler et al. n.d.) and Travelling Texts 1790–1914: The Transnational Reception of Women’s Writing at the Fringes of Europe (Partzsch et al. n.d.), employ digital resources and methods to investigate the translation and transmission of women’s literary works across national and linguistic boundaries.While these editorial projects work predominantly with digitized book collections, others explore the transnational circulation of texts by analyzing mass-digitized periodicals. The Viral Texts project, led by Ryan Cordell and David Smith (2017), uses a machine-learning algorithm to identify frequently reprinted texts in eighteenth- and nineteenth-century newspapers. Focusing in its initial stages on American newspapers, the project is currently moving to incorporate reprinting in British, Dutch, Finnish, German, and Mexican periodicals, so as to understand the mechanisms – material and conceptual – by which texts circulated globally and the effects of this process on the formation of literary cultures (Oceanic Exchanges Project Team). My own editorial work, with Carol Hetherington, identifies fiction in digitized nineteenth- and early- twentieth-century Australian newspapers, held by the National Library of Australia’s Trove database. The resulting collection, To Be Continued: The Australian Newspaper Fiction Database, contains over 23,000 publications of novels, novellas, and short stories that suggest the cosmopolitanism of early Australian reading practices. While many of these stories originated in America, Australia, and Britain, there is also a significant amount of fiction from other anglophone countries, including Canada, New Zealand, and South Africa, as well as works in translation, including from Austria, France, Germany, Holland, Hungary, Italy, Japan, and Russia. Such digital editorial projects function in one respect like print anthologies in world literary studies; in presenting literary works from multiple contexts they expand the range of national literatures that can be studied and included on the curriculum. Yet owing to the expansive capabilities of digital platforms, they are able to collect a much larger range of fiction than print anthologies, whether in terms of the authors, genres, or languages represented.A number of projects use digital methods to ask the type of contextual and comparative questions that have long characterized world literary studies, but in relation to a wider sample of texts and/or contexts than has been possible previously. Chengzhou He searches databases and e-texts to find references to Isben’s A Doll’s House in Chinese literature from 1911 to 1949, so as to understand how “Isben is received, localized, and turned productive in the Chinese social and cultural spheres” (2017: 158). In connecting empirical historical trends as well as specific literary moments to the political and artistic context in China, He aligns digital research with an event-based conception of world literary studies that operates at both “micro” and “macro” levels. The WorldLiterature@UCLA project is likewise concerned with the ways in which particular texts are translated across cultures, languages, and nations as exemplars of world literature (Kim, Shepard, and Wan n.d.). Focusing on Johann Wolfgang von Goethe’s The Sorrows of Young Werther, the project visualizes correspondence networks, literary translations, and influences on contemporary thinkers in order to locate Goethe and his novel in a global context.Other projects look beyond the global circulation of specific authors or works, applying computational methods to large bibliographical or textual collections of literature to explore the world literary system in terms of communities and circuits of reception. Richard Jean So and Hoyt Long use the poetry journals in which authors are published to explore how modernist literary communities developed in America, Japan, and China (2013). Because these journals functioned as the institutional sites through which a readership and a market for modernist poetry developed, these patterns of publication indicate differences in the nature of these literary communities, including a contrast between a centralized and interconnected modernist network in America, a bifurcated Japanese field, and discrete and disconnected modernist literary communities in China. I have used the transnational collection of fiction in the To Be Continued database to explore the transnational movement and influence of literary works in the nineteenth century. For instance, network analysis of stories that are published on multiple occasions in Australian newspapers indicates the existence of previously unrecorded syndication agencies bringing fiction to, and circulating it within, the Australian colonies (Bode 2017b).Increasingly literary scholars use forms of computational modelling to explore literary forms, cultures, and understandings that exist in multiple national contexts. Much of this research adapts methods from computational linguistics. One of the earliest of these studies is Matthew Jockers’s Macroanalysis, which investigates thousands of nineteenth-century British and American novels, employing computational stylistic analysis to show that multiple aspects of this fiction – including genre, time of composition, and the gender of authors – display discernible characteristics across the two national contexts. Where Jockers’s focuses on two anglophone literary cultures, Andrew Piper investigates multi-language corpuses by defining and modeling formal features of literary works. For instance, Piper uses vector-space analysis of the pre- and post-conversion books of Augustine’s Confessions to define its narrative structure in terms of distances (such as between the earlier and later parts of the work). Piper uses this modeling of features to identify occurrences of this conversion plot in a collection of novels and autobiographies in English, French, and German (2015).In a different article to the one mentioned above, Long and So also model a formal literary feature – in their case, stream of consciousness – in order to explore its transmission across different national literatures and languages (2016). Defining this literary device in terms of thirteen elements (including average sentence length and proportion of sentences without verbs), they track the diffusion of stream of consciousness not only from avant-garde to popular works from a number of anglophone nations, but into Japanese literature. While the construction of models might still incorporate cultural biases from the collections used – for instance, Long and So define anglophone literary cultures by sampling from 10,000 titles most commonly held by American libraries – the practice of computationally defining and exploring formal features of literature is an important development in digital approaches to world literary studies. Resonating with the formalist emphasis in Moretti’s work, and with the exploration, in comparative literary studies, of literary phenomena that exist across cultures, these modeling projects propose literary and linguistic elements that are consistent enough to identify computationally despite the different contexts and languages of their production.Other projects employ computational modeling in ways that align linguistic features of works with the contexts of their reception in order to explore the social and cultural consequences of the global circulation of literature and literary values. In other work relating to the transnational collection of fiction in the To Be Continued database, I demonstrate and explore the capacity of machine-learning models to predict the national origins of stories. While such predictive capacity poses the existence, within this transnational market, of discernible national literary traditions, relationships between the different national literatures suggest ways in which an Australian literary tradition developed by incorporating and adapting particular features of British and American fiction (Bode 2018: 157–97). Ted Underwood and Jordan Sellers use machine-learning methods to consider how conceptions of literary value developed in Britain and American in the nineteenth and early twentieth centuries. Using a machine-learning model to predict whether British and American poetry collections were reviewed or not, they suggest a correlation between the linguistic features of a work and the literary esteem in which it was held. In doing so, they hypothesize the existence of common, transatlantic standards for literary evaluation that developed gradually over the nineteenth century, rather than changing suddenly at the end of this period, as has been widely argued. While networks and patterns are still central forms in digital literary studies, as the prominent use of modeling and machine- learning methods in these digital projects shows, the emphasis has largely moved away from the intuitive recognition of connections and resonances that Moretti’s work often foregrounds.As digital and world literary studies are increasingly integrated, there are opportunities for theoretical perspectives from both fields to inform and enrich the other. The profound ways in which social, ideological, and economic factors shape the digital collections with which literary research is now increasingly conducted makes theoretical reflections on information and infrastructure an essential foundation for contemporary literary studies, including world literary studies. Alan Liu’s work, which examines the relationships between epistemological and infrastructural regimes, is key in this respect including his 2004 book, The Laws of Cool, and his 2008 book, Local Transcendence. His recent claim that humanities researchers must contribute to an emerging “critical infrastructure studies” – indeed, that contemporary cultural studies has, in essence, already become this – points to a major field of inquiry with which world literary studies, and other areas of the humanities, will have to contend in coming decades (e.g., Liu 2017a; Liu 2017b). Early steps along the path to critical infrastructure studies include researchers using existing frameworks from textual studies – particularly scholarly editing and bibliography – to theorize the entities we analyse in digital literary studies (e.g., Bode 2018: 37–57; Cordell 2017; McGann 2014) or offering detailed histories of digital collections such as ProQuest’s Early English Books Online ( HYPERLINK \l "c039_r047" Mak 2014) and Gale’s British Nineteenth-Century Newspapers (Fyfe 2016). While the former studies seek to provide a critical language for understanding the digital entities that literary scholars investigate, the latter emphasize the major implications that the construction of digital collections has for the form and validity of scholarly arguments.Significant potential also exists for critical discussion of translation in world literary studies to enrich understandings of digitization. To return to Saussy’s survey of comparative and world literary studies in the age of globalization, his insistence on “the language of the original, [and] the language of translation … as something more than delivery systems for content … as having a weight and resistance of its own” is acutely relevant to digitization and its effects (2004: 14). Mechanistic accounts of digitization – as much as translation – overlook or obscure such weight in dividing these practices from the “the production of meaning,” rendering them “invisible … an extension – faithful or unfaithful – of the original work attributed to the author” (Ungar 2004: 129). In furthering this analogy, what parallels might exist, for instance, between contemporary practices of mass-digitization** and Spivak’s claim that “all the literature of the Third World gets translated into a sort of with-it translates, so that the literature by a woman in Palestine begins to resemble, in the feel of its prose, something by a man in Taiwan” (1993: 182)? Scholars such as Andrew Piper have begun to draw links between linguistic translation and the transition from words to numbers in digital literary studies (2018). But understandings of digitization could be substantially enhanced by further attention to the political and ethical dimensions of translation explored in comparative and world literary studies.Some of this potential might be seen in applying, to digitization, Lawrence Venuti’s rejection of mechanistic or literalist translation. While Bergenmar and Leppanen note, with reference to digitization, that any translation is an adaptation to culture and language (2017: 239), Venuti’s idea of translation as having ethical effects on and a performative relationship to the receiving culture and its dominant resources and ideologies suggests avenues for investigating the complexities of this process. For instance, we might ask whether the adaptations of digitization always domesticate the source text (a claim that Venuti makes of translation because of its inevitable aim to interpret that text in ways that are intelligible and interesting to the receiving situation)? As in Venuti’s case studies of translated works, addressing this question for digitization would require an integration of textual analysis and archival research. Such an approach would enable exploration not only of textual contents but of the selection of texts, their formal and discursive features, and the values and beliefs, individual and institutional actors, and representational and medial contexts involved in remediation. The range of factors encompassed by such analysis resonates well with critical infrastructure studies but encourages an extension of the current focus in that field on the conditions of the production of a digitization to the always emergent conditions of reception.Venuti’s focus on the centrality and invisibility of translators to translation also relates to emerging analyses of labor and power relations in digitization (for example, Warren 2018), while signaling interesting and potentially critically productive tensions between translation and digitization in this respect. For instance, while the figure of the translator is an established one in literary culture, who is the digitizer? Is it the librarian or literary scholar who selects the text; the (often female, often black) worker who conducts the scanning; or even the optical character recognition algorithm that transforms image into text? What does it mean, for a political and ethical conception of digitization, and for understandings of the contemporary humanities more broadly, that this mediating role is thus distributed, including across human and nonhuman actors? Finally, the challenge that Venuti’s framework poses to imagine a foreignizing digitization – one that remains legible while disclosing the digitized status of the text and the digitizer’s intervention – might prompt questions for humanities research, including for world literary studies, about the relationship between interpretation and political efficacy, on the one hand, and the scale and standardized workflows of digitization, on the other.My point is not to pose a direct analogy between practices of translation and of digitization; rather, it is to suggest some of the ways in which questions that have long been asked about the effects of translation in a world where resources are unevenly distributed might equally be asked of digitization. While research in critical infrastructure studies is challenging the notion that digitization reproduces the same text, the simultaneously ethical and political orientation of translation studies might be adapted to enable a fuller understanding of the difference that digitization makes by investigating how relations of power – ideological, disciplinary, cultural, historical – shape and constitute digitized documents and mass-digitized collections. Perhaps a combination of critical infrastructure and translation studies could supply an effective foundation for world literary studies in this age of digitization: on the one hand, for interrogating assumptions embedded in the foregrounding of extensive archives, networks, and patterns in both digital and non-digital expressions of world literary studies; on the other, for exploring alternatives to the flat and distributed interconnections proposed by such formal categories in ways that are more attentive to the disparities of power that occur when different communities, cultures, and languages as well as media, technologies, and communication systems intersect.In the past two decades, “distant reading” has been one of the most influential paradigms in literary studies, and its impact on world literary studies is undeniable. Both within and beyond this field, the idea of distance as a condition of knowledge, and of literary data and computational methods as offering radically new approaches to literature and culture, has been proclaimed and condemned in equal measure. In a surprising number of ways, the debate about “distant reading” reiterates issues prominent in long-standing discussions about the relationship between comparative and world literary studies: regarding the role of the canon and of national literatures in enabling and limiting our understanding of how literature exists and operates in the world, as well as the meaning and implications of scale – and of a world literary system – in shaping that understanding. The tendency for both critics and advocates of digital resources and methods to emphasize their difference from earlier approaches makes it especially important to consider how the digital infrastructure through which we increasingly study world literature – whether through computational or non-computational means – might re/inscribe inequalities that have shaped and continue to shape the production, circulation, and reception of literature. Although they might amplify, digital technologies do not inevitably reinforce such inequalities, and multiple projects show how digital resources and methods can be used to explore how literary works, communities, traditions, and/or concepts circulate and generate meaning in spaces not aligned with the nation-state.References Allington, Daniel, Sarah Brouillette, and David Golumbia. 2016. “Neoliberal Tools (and Archives): A Political History of Digital Humanities.” LA Review of Books, May 1, 2016, , Tony. 2003. “Rereading the Archive and Opening up the Nation-State: Colonial Knowledge in South Asia (and Beyond).” In After the Imperial Turn: Thinking With and Through the Nation, ed. Antoinette Burton, Durham University Press, 102–21.Bammam, David, Ted Underwood, and Noah A. Smith. 2014. “A Bayesian Mixed Effects Model of Literary Character.” In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Baltimore, Maryland, USA, June 23-25 2014). Association for Computational Linguistics, 370–79, , Jenny, and Katarina Leppanen. 2017. “Gender and Vernaculars in Digital Humanities and World Literature.” NORA – Nordic Journal of Feminist and Gender Research, vol. 25, no. 4, 232–46.Bianco, Jamie “Skye.” 2015. “This Digital Humanities Which Is Not One.” In Debates in the Digital Humanities, ed. Matthew K. Gold, University of Minnesota Press, 98–112, Minnesota Scholarship Online, doi: 10.5749/minnesota/9780816677948.003.0012.Biblioteca Virtual Miguel de Cervantes. N.d. . Bode, Katherine. 2012. Reading by Numbers: Recalibrating the Literary Field. Anthem Press.Bode, Katherine. 2017a. “The Equivalence of ‘Close’ and ‘Distant’ Reading; Or, Toward a New Object for Data-Rich Literary History.” Modern Language Quarterly, vol. 78, no. 1, 77–106, doi: 10.1215/00267929-3699787.Bode, Katherine. 2017b. “Fictional Systems: Mass-Digitisation, Network Analysis, and Nineteenth-Century Australian Newspapers.” Victorian Periodicals Review, vol. 50, no. 1, 100–38, doi: 10/1353/vpr.2017.0005.Bode, Katherine. 2018. A World of Fiction: Digital Collections and the Future of Literary History. University of Michigan Press. Bode, Katherine, and Carol Hetherington, eds. N.d. To be continued: The Australian Newspaper Fiction Database, http:cdhrdatasys.anu.edu.au/tobecontinued/ Bodmer Lab. N.d. A Digital World Literature. University of Geneve and Martin Bodmer Foundation. Library and findmypast. N.d. British Newspaper Archive. , Susan, Patricia Clements, and Isobel Grundy, directors. N.d. The Orlando Project: Feminist Literary History and Digital Humanities. , John F. 1987. Computation into Criticism: A Study of Jane Austen’s Novels and an Experiment in Method, Clarendon Press.Cordell, Ryan. 2017. “‘Q i-tjb the Raven’: Taking Dirty OCR Seriously.” Book History, vol. 20, 188–225, doi: 10.1353/bh.2017.0006.Cordell, Ryan, and David Smith, eds. 2017. The Viral Texts Project: Mapping Networks of Reprinting in 19th-Century Newspapers and Magazines, , Gregory R., editor in chief. N.d. Perseus Digital Library. , David. 2003. What is World Literature? Princeton University Press. Damrosch, David. 2004. “World Literature in a Postcanonical, Hypercanonical Age.” In Comparative Literature in an Age of Globalization, ed. Haun Saussy, Johns Hopkins University Press, 43–53.Damrosch, David. 2013. “World Literature in a Postliterary Age.” Modern Language Quarterly, vol. 7, no. 2, 151–70, doi: 10.1215/00267929-2072971.Dimock, Wai Chee. 2006. Through Other Continents: American Literature Across Deep Time. Princeton University Press.Eaves, Morris, Robert Essick and Joseph Viscomi, eds. 2020. William Blake Archive. , Matt, and Lynne Tatlock, eds. 2014. Distant Readings: Topologies of German Culture in the Long Nineteenth Century. NED – New edition ed., Boydell and Brewer, 2014. JSTOR, stable/10.7722/j.ctt5vj848.Europeana. Europeana Collections. , Ed, and Kenneth M. Price, eds. N.d. The Walt Whitman Archive. , Paul. 2016. “An Archaeology of Victorian Newspapers.” Victorian Periodicals Review, vol. 49, no. 4, 546–77, doi: 10.1353/vpr.2016.0039.Gallica. N.d. Gallica. , Jonathan, and John Holbo, eds. 2011. Reading Graphs, Maps, Trees: Critical Responses to Franco Moretti. Parlor Press, , Roland. 2004. “Not Works But Networks; Colonial Worlds in Comparative Literature.” In Comparative Literature in an Age of Globalization, ed. Haun Saussy, Johns Hopkins University Press, 212–23.Google Books. N.d. Google Books. books.HathiTrust. N.d. HathiTrust Digital Library. He, Chengzhou. 2017. “World Literature as Event: Isben and Modern Chinese Fiction.” Comparative Literature Studies, vol. 54, no. 1, 141–60, JSTOR, stable/10.plitstudies.54.1.0141.Hirsch, Brett, and Hugh Craig, eds. 2014. “Special Section: Digital Shakespeares.” The Shakespearean International Yearbook, vol. 14. Ashgate. Jeanneney, Jean-No?l. 2006. Google and the Myth of Universal Knowledge: A View from Europe. University of Chicago Press.Jockers, Matthew L. 2013. Macroanalysis: Digital Methods and Literary History. University of Illinois Press.Kim, David, David Shepard, and Jun Wan, with contributions by David Leynov and Elaine Li, and past contributions by Viola Ardeni, Lourdes Arevalo, Nickolas de Carlo, C. Myles Chykerda, Wendy Kurtz, Amber Williams, and Harry Xie. N.d. WorldLiterature@UCLA: Translation Network Analysis in the Digital Humanities. University of California, Los Angeles, , Adam. 2014. “Technology is Taking over English Departments: The False Promise of the Digital Humanities.” New Republic, May 3, , Yvonne, ?sa Arping, Jenny Bergenmar, Gunilla Hermansson, and Birgitta Johansson. N.d. Swedish Women Writers on Export in the 19th Century, Library of Congress. N.d. Chronicling America. of Congress. N.d. World Digital Library. , Alan. 2004. The Laws of Cool: Knowledge Work and the Culture of Information. University of Chicago Press.Liu, Alan. 2008. Local Transcendence: Essays on Postmodern Historicism and the Database. University of Chicago Press.Liu, Alan. 2017a. “Assessing Data Workflows for Common Data ‘Moves’ Across Disciplines.” Alan Liu blog, May 6, doi: 10.21972/G21593.Liu, Alan. 2017b. “Toward Critical Infrastructure Studies: Digital Humanities, New Media Studies, and the Culture of Infrastructure.” Alan Liu blog, Feb. 23, , Hoyt, and Richard Jean So. 2016. “Turbulent Flow: A Computational Model of World Literature.” Modern Language Quarterly, vol. 77, no. 3, 345–67, doi: 10.1215/00267929-3570656.Mak, Bonnie. 2014. “Archaeology of a Digitization.” Journal of the Association for Information Science and Technology, vol. 65, no. 8, 1515–26, doi: 10.1002/asi.23061.Marche, Stephen. 2012. “Literature is not Data: Against Digital Humanities.” LA Review of Books October 28, , Jerome, ed. 2008. The Complete Writings and Pictures of Dante Gabriel Rossetti. , Jerome. 2014. A New Republic of Letters: Memory and Scholarship in the Age of Digital Reproduction. Princeton University Press. Ministry of Human Resource Development. N.d. Digital Library of India. . Montrose, Louis. 1992. “New Historicisms.” In Redrawing the Boundaries: The Transformation of English and American Literary Studies, ed. Stephen Greenblatt and Giles Gunn. Modern Language Association, 392–417.Moretti, Franco. 2000. “Conjectures on World Literature.” New Left Review, vol. 1, 54–68. , Franco. 2005. Graphs, Maps, Trees: Abstract Models for Literary History. Verso.Moretti, Franco. 2013. Distant Reading. Verso.Moretti, Franco. 2017. “Patterns and Interpretation.” Stanford Literary Lab Pamphlet, vol. 15 (Sept.). , Leslie A., general ed. 2012. Emily Dickinson Archive. . Mowat, Barbara, Paul Werstine, Michael Poston, and Rebecca Niles, eds. N.d. Folger Shakespeare library: Shakespeare’s Plays from Folger Digital Texts. Folger Shakespeare Library, National Library of Australia. N.d. Trove. Exchanges Project Team. 2017.?Oceanic Exchanges: Tracing Global Information Networks In Historical Newspaper Repositories, 1840-1914. Doi: 10.17605/OSF.IO/WA94S.?Ortega, ?lika. 2018. “Archives, Libraries, Collections, and Databases: A First Look at Digital Literary Studies in Mexico.” Hispanic Review, vol. 86, no. 2, 229–47, Project MUSE, doi: 10.1353/hir.2018.0016.Partzsch, Henrietta, P?ivi Lappalainen, Katja Mihurko Poni?, Marie Nedregotten S?rb?, and Suzan van Dijk. N.d. Travelling Texts 1790-1914: The Transnational Reception of Women’s Writing at the Fringes of Europe, , Andrew. 2015. “Novel Devotions: Conversional Reading, Computational Modeling, and the Modern Novel.” New Literary History, vol. 46, no. 1, 63–98, Project MUSE, doi: 10.1353/nlh.2015.0008.Piper, Andrew. 2018. Enumerations. University of Chicago Press.Plaisant, Catherine, James Rose, Bei Yu, Loretta Auvil, Matthew G. Kirshenbaum, Martha Nell Smith, Tanya Clement, and Greg Lord. 2006. “Exploring Erotics in Emily Dickinson’s Correspondence with Text Mining and Visual Interfaces.” In Proceedings of the 6th ACM/IEEE-CS Joint Conference on Digital Libraries, Association for Computing Machinery, 141–51, , Lara. 2016. “The Transnational and the Text-Searchable: Digitized Sources and the Shadows They Cast.” American Historical Review, vol. 121, no. 2, 377–402, doi: 10.1093/ahr/121.2.377.Risam, Roopika. 2015. “Beyond the Margins: Intersectionality and the Digital Humanities.” Digital Humanities Quarterly, vol. 9, no. 2, , Roopika. 2017. “Colonial and Postcolonial Digital Humanities Roundtable.” Roopika Risam blog, October 27. , Shawna. 2018. “Toward a Feminist Modernist Digital Humanities.” Feminist Modernist Studies, vol. 1, no. 3: 211–29. doi: 10.1080/24692921.2018.1505821.Saussy, Haun. 2004. “Exquisite Cadavers Stitched from Fresh Nightmares: Of Memes, Hives, and Selfish Genes.” Comparative Literature in an Age of Globalization, edited by Haun Saussy, Johns Hopkins University Press, 3–42.Smith, Vanessa. 2016. “Novel Worlds.” Journal of Language, Literature and Culture, vol. 63, no. 2–3, 91–95, doi: 10.1080/20512856.2016.1244906.So, Richard Jean, and Hoyt Long. 2013. “Network Analysis and the Sociology of Modernism.” Boundary2, vol. 40, no. 2, 147–82.Spivak, Gayatri. 1993. Outside in the Teaching Machine. Routledge.Spivak, Gayatri. 2003. Death of a Discipline. Columbia University Press.St. Clair, William. 2004. The Reading Nation in the Romantic Period. Cambridge University Press.Stanford University. 2008. Mapping the Republic of Letters. , Katie. 2009. “Critical Response I. Paratext and Genre System: A Response to Franco Moretti.” Critical Inquiry, vol. 36, no. 1, 159–71. JSTOR, ww.stable/10.1086/606126.Underwood, Ted, and Jordan Sellers. 2016. “The Longue Durée of Literary Prestige.” Modern Language Quarterly, vol. 77, no. 3, 321–44, doi: 10,1215.00267929-3570634.Ungar, Steven. 2004. “Writing in Tongues: Thoughts on the Work of Translation.” In Comparative Literature in an Age of Globalization, ed. Haun Saussy, Johns Hopkins University Press, 127–138.University of Chicago. N.d. PhiloLogic: The University of Chicago Full-Text System. , Lawrence. 1999. The Scandals of Translation: Towards an Ethics of Difference. Routledge.Warren, Christopher N. 2018. “Historiography’s Two Voices: Data Infrastructure and History at Scale in the Oxford Dictionary of National Biography (ODNB).” Journal of Cultural Analytics, November 2, , Matthew. 2013. “The Geographic Imagination of Civil War-Era American Fiction.” American Literary History, vol. 25, no. 4, 803–40. Project MUSE, muse.jhu.edu/article/527788.Wilkens, Matthew. 2017. “Genre, Computation, and the Varieties of Twentieth-Century U.S. Fiction.” Journal of Cultural Analytics, November 1, Writers Project. 1999–2016. Women Writers Project. Northeastern University. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download