Return to the MLA Commons
An Evolving Anthology

From Reading to Social Computing

Alan Liu

1 Leave a comment on paragraph 1 0 My topic here is social computing, which gives literary studies a chance to think anew about such apparently axiomatic literary phenomena as writing, publishing, reading, and interpreting. Or, rather, anew is the wrong word. It is more accurate to say that social computing—which includes, but is not limited to, social networking—does not so much make writing, publishing, reading, and interpreting new as it does recover within them, with fresh force, a broader repertoire of historical literary functions than modern literary studies usually needs to attend to. Social computing encourages literary scholars to remember and repurpose the long history of social writing, publishing, reading, and interpreting. I emphasize the perspective of reading. But, really, all the functions and roles of literary activity are in play.

2 Leave a comment on paragraph 2 0 Of course, the sociality of literature has recently been an important theme in a variety of literary-critical approaches—including new historicism, cultural criticism, multiculturalism, the new textual editing, the history of the book, and others I will later mention. But the concept, tools, and—not to be underestimated—growing audience of social computing make it tempting to think that such sociality might now also be implemented in new practices, and not just in themes, of literary scholarship.

Social Computing

3 Leave a comment on paragraph 3 0 When the Web was new, I used to tell my students to learn HTML because, as I put it, the Web is one of the most powerful publishing platforms ever invented, and they’d be crazy not to use it to become visible in their intended professional fields. “Look at what other humanists are publishing on the Web,” I said, which was essentially the mission statement of Voice of the Shuttle, the site for the humanities that I started in 1994. Since then, however, my English professor’s analogy of print publication has been superseded by technological advances that make the Internet a challenge to a literary scholar’s received understanding of the core circuit of literary activity: authors, publishers, readers (and interpreters) mediated by documents (see fig. 1).1

4 Leave a comment on paragraph 4 0
Figure 1. Core circuit of literary activity.

5 Leave a comment on paragraph 5 0 We can pick up the story with what might now be named Web 1.0. In the years between its invention and its first popularity in 1993–94, Web 1.0 structured a transmission act that seemed to map well over print publication. I visualize the information architecture of Web. 1.0 in figure 2. According to that architecture, an author wrote a work and uploaded it to a Web server (with or without the help of an editor or technical person, since originally not much proficiency was required to format a work as a static HTML file). Then, the server—technically, the Web-server program on the computer server—performed the act of publishing by fetching HTML files and associated image files from storage and delivering the assemblage through the HTTP transfer protocol to the end user or reader. This is not to say that there was nothing new. Besides accessibility and speed, Web 1.0 introduced new flexibility in the act of consumption. At the reader’s position in the circuit, for instance, the client program (usually a browser) rendered the HTML source code in dynamic adaptation both to local hardware and to users’ navigational and stylistic choices. As a result, the reader using a personal computer acquired an unprecedentedly active role, as envisioned by such early hypertext theorists as George Landow, who thought that digital media realized Roland Barthes’s idea of the “writerly” reader.

6 Leave a comment on paragraph 6 0
Figure 2. Web 1.0 information architecture.

7 Leave a comment on paragraph 7 0 Yet with few exceptions, there was no actual crossing on the early Web between the roles of author and reader, because the information delivery system partitioned those roles from each other, leaving any change in the reading act quarantined at the reader’s station. The reader was newly hyperactive, but at best such hyperactivity was authorship at one remove from what was really happening on the server. It was simulated authorship.

8 Leave a comment on paragraph 8 0 No sooner did the early Web take off, however, than Web 1.5 (as I call it for lack of a better designation) quietly supervened. After the Internet data backbone went commercial in 1995, an increasing number of corporate and institutional Web sites came online, bringing with them their databases. These they inserted into the circuit of transmission, as visualized in figure 3. The author wrote content but, instead of uploading to the Web directly, used a Web “form” to add the content to an underlying database, which ran on the server alongside the Web-server program. Then, when a reader clicked on a link to request content, an output template, or theme page—actually, a clustered series of files with HTML for Web formatting, scripting code for talking to databases, cascading style sheets for fine-tuning layout and styling, and JavaScript for active client behaviors—extracted material from the database, formatted it on the fly in the predesigned template, and delivered the overall construction to the reader.2 The result is that the author now really did require technical assistance, since the know-how needed to design databases and the “middleware” code that shuttles content between databases and the Web was much higher. But this itself did not (at least, at first) seem to change the fundamental paradigm. After all, Web 1.5 was in this respect no different from professional print publication, which also required specialized editorial, design, and other staff members to produce an author’s manuscript (a publication process that by this time had become increasingly digital).

9 Leave a comment on paragraph 9 0
Figure 3. Web 1.5 information architecture.

10 Leave a comment on paragraph 10 0 The real game changer was that the effective range of the reader’s activity expanded. In Web 1.5, readers no longer just retrieved particular resources “as is” from the server to customize on their local computer. Instead, just as authors used Web forms to manage content in the server’s database, so readers used their version of Web forms (advanced search forms, “shopping cart” pages, and so on) to transact bidirectionally with that database, reading from the database but also writing to the database. After all, online commerce sites wanted customers to write in credit card numbers, mailing addresses, product reviews, and so on. In effect, the range of the reader’s agency expanded to the server at the halfway point between author and reader.

11 Leave a comment on paragraph 11 0 Web 2.0 arose when this potential for expanded reader agency awakened a new paradigm. About the time of the 2000 dot-com bust, it dawned on a new generation of Web application developers that the data architecture I have called Web 1.5 could be tapped in new ways. As long as so much technically demanding work was needed to design the database and middleware through which authors and readers separately interacted with a database, why constrain the system asymmetrically so that authors, usually one at a time, primarily input content while readers primarily requested the output of that content while writing only small bits into the database in little name, address, credit card number, or “review our product” fields? Why not give readers Web-input pages similar to those used by authors so that they can write more fluently into the identical database, thus effectively allowing readers to become prolific commentators and actual coauthors? In figure 4, I visualize in simplified form the data architecture of Web 2.0 with its explosion of blogs, wikis, social-networking sites, shared bookmarking sites, collaborative document work spaces, and so on. Take the example of a blog. A blog allows an author to use a Web form to write content into a database for presentation on the Web. But readers of that blog can use a “comment” Web form to write information into the same database for presentation below or alongside the author’s writing. In similar fashion, social-networking sites allow readers to post to an author’s Facebook “wall” or equivalent. Or consider the example of a wiki, which allows readers to use Web-input pages to alter an author’s writing directly and so become coauthors.

12 Leave a comment on paragraph 12 0
Figure 4. Web 2.0 information architecture.

13 Leave a comment on paragraph 13 0 Above all, the result is social computing, the hallmark of Web 2.0.3 Recall that when modern computers were invented in the World War II era, they were first thought of primarily as ballistics or scientific-calculation machines. Then, with the advent of corporate mainframes, calculation took a backseat to business functions originally peripheral to computing: storage, filing, sorting, and printing. With the personal computer and the Internet, computers next turned into universal communication devices and media players.4 Now, with Web 2.0 (abetted even more recently by ubiquitous mobile computing and communications), the computer enfolds all its previous functions in the increasingly dominant paradigm of social computing, which the interdisciplinary research group on the topic that I participated in on my campus defines most generally as “the use of technology in networked communication systems by communities of people for one or more goals,” even if that goal is as seemingly unfocused as building the community itself and one’s identity in it.5 Oft-cited examples include blogs and their derivatives (including all manner of sites powered by the WordPress blog engine and Drupal community-message-board engine that have evolved into general-purpose content management systems); microblogging platforms like Twitter and Tumblr; social-networking platforms like Facebook and LinkedIn; wikis like Wikipedia or such make-your-own wiki platforms as PBWorks (previously PBWiki); social-bookmarking or content-discovery sites like Delicious (previously del.icio.us) and StumbleUpon; shared or social news sites like Digg (in its heyday) and Reddit; and—at least in some of their shared or specialized community features—image- and video-sharing sites like Flickr or YouTube as well as collaborative-work-space sites like Google Docs. At the level of infrastructure, most such social-computing systems are based on the same back-end Web 2.0 data architecture I describe above, but with increasingly complex subarchitectures. Indeed, many are based on the identical LAMP open-source software platform (the acronym for the combination of the Linux operating system, Apache Web-server program, MySQL database program, and PHP scripting code). Just as important as the shared infrastructure are shared superstructural conventions, practices, and forms. Some of these may be likened to classical rhetorical devices—for example, common Web 2.0 topoi such as the profile page, post, comment, tag, and so on. At a higher level of convention, the various social-computing instances I have cited (blogs, wikis, social-networking sites, etc.) are akin to genres. And if we were to go platonic, the equivalent of transcendental, and not just Web, forms in the social-computing universe is the social graph. Often invoked with the definite article as if there were just the one social graph, this concept bridges the literal notion of a social-network diagram visualizing all one’s online social connections and a quasi-metaphysical theory of universal sociality.6

Social Reading

14 Leave a comment on paragraph 14 0 Meanwhile, in almost exactly the same decades that the Internet arose and eventually evolved social computing, literary scholarship followed similar principles of decentralization to evolve cultural criticism.7 I refer in particular to literary theory from the 1960s onward—precisely Landow’s reason for analogizing hypertext to Barthes and other poststructuralists. It would be foolish to generalize uniformly about literary theory in the past half century. What is germane here is simply that many such theories channeled—inflected, deflected, or reflected on—the social activism of the May 1968 era. As a consequence, whatever their differences, they manifested a common will to decentralize or democratize the traditional understanding of literary sociality I above called the core circuit of “authors, publishers, readers (and interpreters) mediated by documents” (fig. 1)—the purest transistor of which was the document, or text “itself,” as crystallized by New Criticism.

15 Leave a comment on paragraph 15 0 The exact size of the established core circuit I visualize here, I note, can be slightly incremented or decremented without essential difference to my present argument. For example, perhaps we can dispense with the publisher and interpreter, leaving the transistor of the document with just the single input of the author and the single output of the reader. The third, regulating wire of the transistor might then be tradition, pragmatist experience, Marxist class, psychoanalytic mind, or any other transcendental or immanental source.8 The main point is that modern literary criticism up to the 1960s processed interpretation through a core circuit of social functions that had been regularized by centuries of evolution in the underlying technology, economy, politics, law, and culture of print. The result, if not “standardization” and “fixity” (as Eisenstein puts it in her history of the print revolution [51–64, 78–89]), was at least the calming of unfixity (the epitome of which was “piracy,” as Johns calls it in his rebuttal of Eisenstein).9 Thus a working literary scholar up through the late 1960s could do interesting interpretive work with just this core circuit (really, a “lite” model) of literary sociality because literature’s who, where, when, why, and how had been standardized (puzzle key to the above—who: canonical authors; where: the major publishing cities; when: between Shakespeare and Joyce; why: art; and how: print).

16 Leave a comment on paragraph 16 0 After May 1968, however, literary theory democratized this core circuit of literary sociality. Partly, it did so by rebalancing the weights of the nodes in the core circuit so that the previous value placed on what Foucault called the “author function” diminished. Reader-response and reception theory, for instance, elevated reading activity in the circuit. Deconstruction elevated the role of the reader as interpreter (and, under the rubric of intertextuality, even reconceived the author as just another reader). Indeed, deconstruction implicitly socialized literary agency by distributing it throughout the literary transmission circuit in the name of language or discourse. It was language that wrote, read, and interpreted. Similarly, Mikhail Bakhtin’s idea of dialogism and the “novelistic” (making its impact on literary theory from the 1980s on) dispersed authority “centrifugally” among many cultural voices. Or, again, there was Michel de Certeau’s “practice of everyday life,” which envisioned a “tactical” active reader analogous to a jaywalking pedestrian taking it on his own authority to walk a path different from any official A to Z. The eventual upshot of such post–May 1968 theory was cultural criticism, including new historicism, cultural materialism, Althusserian Marxism, post-Marxism, race and ethnicity studies, and feminist and gender studies (with their cross-disciplinary allies in cultural anthropology, new cultural history, social linguistics, and so on). This big tent of cultural criticism democratized the core circuit of literary sociality even further. New cultural history and new historicism, for instance, imagined that readers could transgressively interpret authors, even as authors themselves could do the same to monarchs. It was all subversion, the new-historicist dream of jaywalking or being a hyperactive subject.10

17 Leave a comment on paragraph 17 0 By way of corollary, we can turn to the new textual editing and new bibliography, the resurgent history of the book, and new media studies (including so-called media archaeology). These approaches not only redistributed the weights in the core circuit of literary social activity but also revealed that that core was indeed only lite by restoring to view other vital nodes in the circuit: editors, publishers, translators, booksellers, shippers, balladmongers or peddlers, annotators, censors, and so on (fig. 5).11 Indeed, in the recent mode of materialist book studies, even the apparatuses and substrates of literary activity (e.g., writing tablets as studied by Stallybrass and others) might be conceived as akin to actants, somewhat in the spirit of Bruno Latour’s actor-network theory of distributed human-machine agency. D. F. McKenzie’s influential Bibliography and the Sociology of Texts may be taken as exemplary of these new approaches to texts. McKenzie writes:

18 Leave a comment on paragraph 18 0 My argument therefore runs full circle from a defence of authorial meaning, on the grounds that it is in some measure recoverable, to a recognition that, for better or worse, readers inevitably make their own meanings. In other words, each reading is peculiar to its occasion, each can be at least partially recovered from the physical forms of the text, and the differences in readings constitute an informative history. What writers thought they were doing in writing texts, or printers and booksellers in designing and publishing them, or readers in making sense of them are issues which no history of the book can evade. (19)

19 Leave a comment on paragraph 19 0
Figure 5. Larger circuit of literary activity (examples).

20 Leave a comment on paragraph 20 0 I call attention not just to McKenzie’s expansion of the agency of the reader (“readers inevitably make their own meanings”) but to his insertion into the circuit of “printers and booksellers,” who—with their whole house or, in modern times, corporation (not to mention extended social networks of journeyman writers, materials suppliers, print jobbers, shippers, etc.)—repopulate the sparse, lite model of literary activity. The result of such approaches can be counted today in research projects like the English Broadside Ballad Archive, which digitizes early modern ballads in social context to refocus our understanding of literary transmission around such then-pivotal nodes as the printer or publisher (who often initiated or managed a run of ballads) rather than the author.12 Or consider such recent historians of the book as Ann Blair and William Sherman, who study annotation practices—for example, notes scribbled in the margins of a manuscript or book—to witness a whole zone of literary activity that is undecidably readerly and writerly. Important contributions have also been made by scholars of the prehistory of literacy. In what Walter Ong calls “primary oral culture,” after all, the literary act is perhaps best conceived not as transmission at all instead but as community—the original village behind Marshall McLuhan’s mediated global village.

21 Leave a comment on paragraph 21 0 In the end, we may say, the post–May 1968 literary-critical movements changed the game of literary study in two ways. First, while scholars played literary criticism on the same game board of the core circuit I define above, they did so with bold new moves, allowing the reader to usurp the author. Second, scholars expanded the game board by adding such new squares on it as the tavern, coffeehouse, Grub Street, and other scenes on which historical writing and reading were a hubbub of collective literary life and where there were no stable distinctions between primary and secondary players. As Michael Bérubé, Hester Blum, Christopher Castiglia, and Julia Spicher Kasdorf recently commented in their PMLA article, “Community Reading and Social Imagination” (borrowing from the work of David Shields), “[R]eading was not always so solitary. . . . [T]he Anglo-American world of the eighteenth and nineteenth centuries was filled with community reading: coffeehouses, literary salons, reading clubs, reform associations, tea tables, lyceums, sewing circles were all places where people read, and listened to others read, together” (422).13 Using the fascinating example of shipborne printing presses run by sailors on nineteenth-century Pacific and polar expeditions, they note that such “community reading [was] also, in important ways, community writing” (422), producing not just conversations and ideas about preexisting literary works (e.g., about plays performed on board) but also reviews, counterreviews, broadsides, newspapers, whole books, and other “new works for circulation, debate, and provocation” (423).14

22 Leave a comment on paragraph 22 0 Yet ultimately it is too schematic to say that there were two separate game boards, one core and the other expanded. Really, there was one game board that all the new decentralizing literary-critical approaches I mention above skewed into a new social geometry by adding what can be generalized as a margin. In various ways, for example, deconstruction, cultural criticism, and the field of the history of the book defined marginal zones of literary activity that renegotiated the roles of literary sociality. Thus Jacques Derrida famously created such marginal interpretive paratexts as the single, running footnote that extends through his essay “Living On / Border Lines.” Cultural and multicultural criticism attends to the writings of marginalized peoples. And the field of the history of the book, as earlier mentioned, studies annotations that are literally in the margin, not to mention such “ephemera” as ballads figuratively marginal to canonical literature.

23 Leave a comment on paragraph 23 0 Today, the margin is the “sidebar” of such social-computing sites as blogs, where all the blog rolls, track-backs, and other signs of the vitality of communal communications manifest. Consider as an illustration, for instance, a blog post such as that in figure 6. As I have written about this example, “[t]he actual post in the main document window is minimal (and, in fact, does no more than link to a document elsewhere: ‘A good post on trust, gaming and social networking!’). The real action is in the sidebar that unfurls—as if from hidden dimensions—a long blog roll with links to other blogs in the relevant social universe” (Liu, “End” 514).

24 Leave a comment on paragraph 24 0
Figure 6. Post of 4 September 2007 from Robin Hunicke’s blog Gewgaw. The full, scrollable length of this post, including the sidebar, is here split in two for static display.

Social Computing Plus Social Reading

25 Leave a comment on paragraph 25 0 Why should contemporary literary scholarship take an interest in contemporary social computing? How should it take an interest?

26 Leave a comment on paragraph 26 0 Let me first address the why? question. In general, as I have taken to saying to colleagues suspicious of all the buzz about digital technologies, the stake for literary studies in the digital age is not first of all technological. It is to follow the living language of human thought, hope, love, desire—and hate too—wherever it goes and wherever it has the capacity to be literary, even if the form, style, or grammar of such literariness does not always conform to canonical standards. Speaking for a moment in personal voice—that is, in a social-networking style suited to my topic—I have cried for joy, sorrow, and shame over literature in print. That is part of what made me want to profess literature. In recent years, I have also done so over writings in Web 2.0. Of course, vast stretches of the latter—in any of the usual online modes of volunteer journalism and reviews, blog entries, epistolary comments, satires and whimsy, and sometimes collaborative fiction—are at present only a colloquial approximation (if that) of high literate language or literary sensibility. (To extend an analogy to the eighteenth century that Nunberg makes, an apt comparison might be literature situated somewhere between the coffeehouse and Grub Street, with the new junk form of the novel thrown in.)15 Nevertheless, I recognize that much of what provoked me to turn to literature in the first place—vital, daring, and meditative expressions of human experience—is there. It is there in the naked lyric of a blog post celebrating or mourning some personal or public event. It is there in the classical drama of a brawling, controversial Wikipedia article whose behind-the-scenes “talk” page stages the chorus of the “rule of many” or “wisdom of crowds.”16 And it is there in the epic of all the social-news, shared-bookmark, or similar sites that build a portrait of collective life from constantly reshuffled excerpts, links, and tags from that life akin to Homeric formulae. Above all, as a literature professor, I recognize that—viral YouTube videos aside—the vast preponderance of Web 2.0 is an up-close and personal experience of language. Much of that language, most of which is textual, may be demotically raw, even feral, but not all. Much is merely informational or metainformational (comments on information). But some is “cool,” which—as I have discussed in my Laws of Cool—is the complex ethos and aesthetic of contemporary information that may well be the rag-and-bone shop of the heart, where literature must now put up its ladder. If one loves literature, I think, one now has to be willing to go speculatively where the language of passionate life goes, especially among the young, who will carry on the cool literary adventure.17

27 Leave a comment on paragraph 27 1 Now let me turn from the why? to the how? question, which will require some scholarly and not just cool adventuring. How can literary scholarship engage with social computing? My best guess is that the scholarly adventure will compass at least two research portfolios.

Social Computing as an Object of Literary Study

28 Leave a comment on paragraph 28 0 First, literary scholars will need to do considerable homework to make social computing a relevant object of study, something worth looking into in its own right. This will be a challenge because it is a matter not just of borrowing concepts from social computing with relevance to literary study but of properly framing social computing and literature together as parts of an integral object of study. That object is a broad notion of expressive sociality in which social computing and literary activity are both aspects of a single communicational phenomenon: the contemporary form of the human need to say something well (memorably, persuasively, movingly, beautifully, wittily, and so on) to someone else.

29 Leave a comment on paragraph 29 0 Conceiving of such a unified field of literary and social communicational study will require significant methodological work. Fortunately, two component methodologies already exist, each well advanced and waiting to be coupled.

30 Leave a comment on paragraph 30 0 One set of methodologies comes from the social sciences. Sociologists, communication scholars, information-studies researchers, organizational theorists, and others have applied a combination of empirical, ethnographic, and documentary-investigative methods—for example, surveys, interviews, observation, discourse analysis, and experiment—not just generally to ICT (information and communication technologies) but specifically to social-computing technologies such as blogs, Wikipedia, MMOGs (massively multiplayer online games), Second Life, and online social or political-action sites. In the process, they have adapted key social science research paradigms developed before the digital age. One key paradigm is social-network analysis, which originated in gestalt theory and in sociometric analysis at the beginning of the twentieth century, evolved through the Harvard-centered “clique” research of the 1930s and 1940s, added the Manchester-school focus on social conflict in the 1950s and 1960s, and then came together in contemporary social-network analysis through the work of the so-called Harvard structuralists with mathematical set theory and graph theory in the 1960s and 1970s (later computationally extended through such commonly used software as GRADAP or UCINET).18 The premise of social-network analysis is that it is not individuals or groups but the pattern of relations between individuals or groups that is socially significant. Such an approach commonly produces analyses in the form of social-network graphs composed of nodes and connecting edges (also called ties; fig. 7) accompanied by metrics of degree, distance, density, betweenness, centrality, clustering, and so on. The goal is to describe a topology of social relations that allows researchers to understand, for instance, which nodes are pivotal to connections within communities. Today, social-network analysis is increasingly used to map online social networks. Digital social-network analysis and graphing tools such as PieSpy, Uberlink, NetDraw, UCINET, Gephi, yEd, and Circos (and even such artistic tools as Warren Sacks’s Agonistics) facilitate mapping the social networks that emerge from—and (one of the crucial research questions) are inflected by—digital networks.

31 Leave a comment on paragraph 31 0
Figure 7. Social-network diagrams comparing online and face-to-face relations in the gateway seminar for the technology and society PhD emphasis at the University of California, Santa Barbara, fall 2005. (Created by Clayton Childress.)

32 Leave a comment on paragraph 32 0 Related to social-network analysis are methodologies that allow researchers to model not social structure but discourse patterns conveying social representations of social structure—a kind of social fiction or ideology of society not too different in kind from a literary work about society. Thus, in a stunning example, John Mohr and Vincent Duquenne apply computational tools to large textual data sets to analyze the way that New York City charity organizations between 1888 and 1917 variously classified indigent women as “poor,” “fallen,” “deserving,” “destitute,” “homeless,” “needy,” “distressed,” “stranger,” “misfortunate,” “worthy,” and so on. Mohr and Duquenne produce diagrams of linguistic networks that reveal the meaningful ties and, sometimes even more telling, lack of ties between nodal terms—that is, the aggregate (ideo)logical syntax of what could and could not be said about poor women without circumlocution (fig. 8). Such discourse analysis yields a revealing map of a time’s own social-network theory.

33 Leave a comment on paragraph 33 0
Figure 8. Diagram from John W. Mohr and Vincent Duquenne, “The Duality of Culture and Practice: Poverty Relief in New York City, 1888–1917.”

34 Leave a comment on paragraph 34 0 That is the first set of methodologies needed to take up social computing as an object of study: social science methods that increasingly accommodate discourse analysis. The second set of methodologies I indicate then comes reciprocally from the humanities—specifically, from discourse-analysis methods in the digital humanities that have now begun to accommodate social analysis. I refer to methods of text encoding, text analysis, data mining, and other pattern-recognition analytics that are either implicitly consonant with or explicitly related to social-network analysis (as in the case of the “author centrality” and “author degree distribution” tools offered by the SEASR initiative that are analogous to social-network analysis [see SEASR; “SEASR Analytics for Zotero”]). For those not familiar with what contemporary text analysis can do, the TAPoR portal site of online text-analysis tools is recommended as a first stop for hands-on experimentation. Choose, for instance, the implementation in TAPoR of the most traditional text-analysis tool, the digital concordance (e.g., TaPoR’s Concordance–Plain Text tool), which descends genealogically from the very first digitally facilitated project of text analysis, Roberto Busa’s IBM-assisted concordance to the works of Thomas Aquinas. Applied to a single text, this tool supplies for any given word a concordance that appears as a vertically stacked set of keywords in context (KWIC) aligned around the chosen common word. Applying the tool to large texts such as novels or plays (or, scaling up further, applying families of similar tools to corpora-scale textual aggregates in the mode of what Moretti calls “distant reading” [Graphs] or the Software Studies Initiative at the University of California, San Diego, calls “cultural analytics”) would produce not just larger concordances but, in many cases, a sense of the way whole societies of speakers share words. A recommended second stop might then be WordHoard, an online text-analysis tool for a select group of “highly canonical literary texts”—all of early Greek epic (in original and translation) and Chaucer, Shakespeare, and Spenser.19 Applying text analysis to these works, which have been “deep tagged” (text-encoded at a fine-grained level), WordHoard allows users to constrain their searching, collocating, concordance building, time charting, and other text analyses not just to word forms but to specific works, speakers, gender, publication year, prose or verse, metrical shape, parts of speech, and others. For example, one can produce a concordance of specific words used by women versus men in particular sets of Shakespeare’s plays. Computationally assisted text analysis, we realize, is a way to experiment with literature to bring out, among other features, its latent social network and that of the characters in its imaginative worlds. In the last analysis, after all, a concordance represents how even disjunct speakers share a sense of a word and so conjoin in a discursive structure that images a larger social structure.

35 Leave a comment on paragraph 35 0 Coupling together the social science and digital humanities methods I mention would give us a good start in studying social computing and literature as entangled parts of an integral field of social expression. Social science methodology focuses on social structure but increasingly needs to analyze online discourse to do so. Reciprocally, the digital humanities focus on discourse (including online discourse) but increasingly require social modeling to understand what is happening. The obvious solution is to create a symbiosis of the two methods.

36 Leave a comment on paragraph 36 0 A powerful example from the social science side is Peter S. Bearman and Katherine Stovel’s article, “Becoming Nazi: A Model for Narrative Networks.”20 Bearman and Stovel are researchers in the field of historical sociology who studied first-person “how I became Nazi” narratives dating from 1934. Their hypothesis is not just that narratives can be studied in relation to social networks but that the internal structure of narrative consists of networks of agents, actions, and objects of action that can be analyzed and graphed as if part of a social network. Treating narrative elements as “nodes which are connected by narrative clauses, represented by arcs,” Bearman and Stovel show that “narrative, historical and network data” are all similar in being “locally dense, often cyclic, knotted, and characterized by a redundancy of ties.” Their goal in thus “representing complex event sequences as networks” is “to observe and measure structural features of narratives that may otherwise be difficult to see” (71). Similarly, Roberto Franzosi, in his Quantitative Narrative Analysis, sets forth an impressive method for studying newspapers (in his example, partisan Italian publications of 1919–22) by analyzing narrative discourse as computationally tractable semantic “triplets” (essentially: subject-predicate-object triplets of the sort, “Participant: [Actor: fascists] . . . Process: [Verb: break] Participant: [Object: (Physical object: furniture)]”; 30).21 Statistical operations on these triplets allow him to pattern-mine the story of violence of the times—for example, who struck out at whom? where? with what? when? how often? and in what patterns? The result is a set of social-network, time-series, geographic, and other models of social agon.

37 Leave a comment on paragraph 37 0 Advanced examples from the digital humanities side may be seen in the work of the Stanford Literary Lab, whose participants marry text analysis to social-network analysis to study, for example, plot and character interactions in literary works (Moretti, “Network Theory”) or recommendation networks in the reception history of authors (Finn). The Stanford Literary Lab is also inventing new techniques of literary social-network analysis for studying the impact of particular language on social networks (whereas social-network theory normally treats “edges” in a social graph in a manner insensitive to the fact that they are actually communications with specific semantic or stylistic features) and for walking a computational “window” through plays so as to isolate elemental units of social interaction (one to one, two to one, two and two, etc.) that can be visualized as sequential patterns.22 Other examples include the impressive Mapping the Republic of Letters project, which visualizes networks of written correspondence and intellectual community in the Renaissance and Enlightenment, and Andrew Piper and Mark Algee-Hewitt’s Werther Effect project. Varying Piper’s earlier social-network concept of Goethe’s corpus as a “circulatory process” of authorial editions, adaptations, translations, and spin-off works with “a networked identity” (Piper, “Topologies of Literature”), the Werther Effect and the related Literary Topologies project map Goethe’s work as an evolving, propagating, and differentiating network of linguistic relations (not just bibliographic entities), thus demonstrating, as Piper and Algee-Hewitt put it,

38 Leave a comment on paragraph 38 0 how the patterns of lexical repetition within texts produce meanings that are not localized in, or inherent to, those patterns. Meaning is not a function of signification in a topology, but organization. . . . Topologies reframe our understanding of literary works not as static, discrete objects, but as socially imbedded, circulatory processes—as linguistic events that can be mapped. In so doing, topologies place us in a critical relationship to the network as one of the dominant figures of contemporary thought. (“Werther Effect 1”)23

39 Leave a comment on paragraph 39 0 Of course, considerable technical and other challenges still stand in the way of the coupled social science and humanistic approach to social computing I recommend.24 For literary scholarship, there is a foundational conceptual challenge: most modern literary study has been what may be called documentcentric, to the extreme that in New Critical close reading the document itself is supposed to contain all we need to know about the relevant social context. Even when post–May 1968 literary criticism moved to what I called the margins by recovering the contexts of texts (the intertext, historical texts, marginalized texts), it did so in a manner that closely read those marginal texts. For instance, one reads closely a Derrida footnote or a new historicist anecdote. But social computing today only sometimes manifests in texts of the sort that reward close reading. Instead, the full measure of social computing is taken only through digital versions of what Moretti, as mentioned above, calls “distant reading”—that is, statistical or pattern-based data mining operations outputting (as he puts it in the now-famous title of his book) “graphs, maps, trees.” In the context of social computing, the output is likely to be social-network diagrams and other approximations of the holy grail of Web 2.0 social computing: the social graph. Whatever stunning literary documents may be discovered that are native to Web 2.0, their close reading will be meaningful only in the context of broad patterns of discourse to be understood through new hermeneutical paradigms of distant reading for which visualizations are the placeholders—that is, through what Willard McCarty theorizes as paradigms of modeling in preference to interpreting.

40 Leave a comment on paragraph 40 0 But social computing as an object of literary study, as I have put it, is only the first research portfolio that literary scholarship must open to come to a rapprochement with social computing. There is also a second research portfolio to explore.

Social Computing in the Practice of Literary Study

41 Leave a comment on paragraph 41 0 It is not enough to approach social computing as a conceptual theme or experimental model for literature. It will also be necessary to channel research methods now being prototyped at the research level (as described above) into practical means of reading, interpreting, and writing about literature—first by adapting off-the-shelf social-computing tools and platforms created for other purposes until capable technologies better adapted to literary study can be created. The full potential of social computing, in other words, will only be realized when it is enacted in the routine ways in which researchers and students read, analyze, discuss, and publish about both primary literature and scholarly secondary literature.

42 Leave a comment on paragraph 42 0 Consider, to start with, the way that social-computing technologies are beginning to be used to experience—that is, to read, perform, and communicate (overlapping with “analyze” and “interpret”)—primary literature. The potential is indicated by literary applications of ordinary social-computing technologies and the tools used to analyze such vernacular technologies. One instance is the literary-critical use of social-network graphing tools originally designed to visualize online chat. For instance, although Paul Mutton created his PieSpy tool to “infer and visualize social networks on Internet Relay Chat (IRC),” he soon also used it to perform the same function on Shakespeare’s plays, producing snapshots and animated sequences of shifting social-network character relations in the works (Mutton; see also PieSpy). Using these PieSpy analyses to read The Merchant of Venice, a team of undergraduate students in the Literature+ digital-project course I taught in 2009 studied animated diagrams of the play’s social networks, which, supplemented by text-analysis and other computational methods for grappling with the problems of Shakespeare’s characters, ultimately yielded what I thought was one of the finest examples I have encountered of undergraduates coming to a deep understanding of a literary work. 25 The final interpretive essays that the students wrote seemed to me far richer and wiser than would otherwise have been the case.

43 Leave a comment on paragraph 43 0 Similarly, blog and social-network platforms can be used to engage with literary works. For instance, one team of students in the 2008 version of my Literature+ course used the LiveJournal blogging platform to create profile pages for the characters in the General Prologue to The Canterbury Tales. They themed each profile page with images and language they thought appropriate to the character and then wrote comments “in character” on other characters’ pages or on a shared community page (e.g., the Miller commenting rudely on the Knight’s or Wife of Bath’s pages, much as Chaucer’s original Miller was wont to do). They also created a LiveJournal “community” group for all the characters that simulated the spirit of the overall shared pilgrimage.26 The insight of this project was that the framework of social interaction and discourse that Chaucer created has a remarkable affinity to today’s blogosphere—even to the point that the rudeness, “flames,” baitings from “trolls,” and other apparent debasements and provocations of language typifying the extremes of the blogosphere can be understood to be Chaucerian in ethos. The blogosphere has precisely the group ethos that Chaucer represented in his company of intimate strangers. Another student team in the 2008 Literature+ course used Facebook to model Shakespeare’s Romeo and Juliet. After setting up pseudonymous Facebook accounts for twenty-one characters in the play (Romeo, for instance, was Ro Montagues), the team fleshed out each character’s profile page, created “friend” relations between characters, used the Facebook Friend Wheel application to produce a social-network circular graph of the bifurcated Montague-versus-Capulet community, and then brought the whole social structure to life by “acting” the play. They created a Facebook “group” called “The Streets of Verona,” whose message-board forum staged a large fight between the Capulets and Montagues; had characters speak directly to one another on their “walls”; used “photos” to let characters comment on pictures (a way to refer to events in the play that could not be transferred into Facebook); and mounted a party “event” to represent the Capulet masquerade banquet in act I (which also facilitated the first meeting of Romeo and Juliet through photos posted on the page).27

44 Leave a comment on paragraph 44 0 In essence, Facebook became a platform for character or role-playing. It allowed the students to study the play as if they were directors staging it in an alternative medium. All the world, as it were, is not a “wooden O” but a digital Friend Wheel.

45 Leave a comment on paragraph 45 0 In the future, we might expect digital humanities researchers to adapt such off-the-shelf social-computing technologies or innovate new ones to allow for other ways to experience and communicate—that is, to read, interpret, and perform—primary literature. Perhaps custom-designed reading, interpreting, and performing applications will be created by the robust creative and scholarly community of the Electronic Literature Organization to make literature not just what Noah Wardrip-Fruin calls “playable media” but, specifically, socially playable media akin to two of the interactive social modeling examples that Wardrip-Fruin studies in depth in his Expressive Processing: The Sims, a computer game, and Façade, an interactive computer drama. Or perhaps Ivanhoe, the game of interactive, role-playing literary interpretation created by Jerome McGann, Johanna Drucker, and others, will set the mold for socially computable literary experience.

46 Leave a comment on paragraph 46 0 Ultimately, it may be that experiencing and communicating literature through social-computing technologies will do more than supplement older reading, interpreting, and performing practices. The payoff will be an evolution in our understanding of the nature of reading, interpreting, and performing. When we “read” a Shakespeare play as a social-network diagram or a Chaucer poem as a blog, for instance, we expand or reconfigure the nature of reading. Reading overlaps with the actions of modeling, gaming, role-playing, adapting, translating, rendering, and simulating. With reference to Lisa Samuels and Jerome McGann’s important essay “Deformance and Interpretation,” socially computed reading may even experimentally deform literature to discover new truths about the significance of literature.

47 Leave a comment on paragraph 47 0 Equally intriguing in the second research portfolio I have designated “social computing in the practice of literary study” is the way we can use social-computing technologies to evolve the analytic and interpretive practices of secondary literature—that is, literary scholarship and criticism.

48 Leave a comment on paragraph 48 0 Thus, one notable feature of some of the most advanced online scholarly reading or research environments I am aware of is that, while most continue to be documentcentric (focusing their main interface on documents, annotations of documents, or search-result listings of documents), they are also bolting on borrowed social-computing technologies. For example, NINES’s Collex research environment allows users to “discuss” documents, “tag” them for sharing with the community of folksonomic tags, and participate in “groups.” In a similar spirit, the Zotero and Mendeley bibliographic research environments accommodate shared online collections of resources. Other online research environments such as Academia.edu move even further in the direction of social computing by creating what amounts to scholarly social networks in which papers can be posted to individual- or group-centered networks of departments, faculty members, graduate students, and research interests.28 Similarly, RoSE: Research-Oriented Social Environment, developed by a team under my direction, is joining together historical and contemporary authors and works in a combined social-document graph characterized by an extensible repertory of social, intellectual, and professional relations between entities (including, e.g., apprentice of, publisher of, bookseller of, sister of, enemy of, imitator of, chief of, priest of, mentor of, etc.) (fig. 9).29 The goal—shared by SNAC: Social Networks and Archival Context project, the ConceptVISTA tool in the geosciences, the Knalij visualization tool in the medical sciences, and other projects—is to expose to view and make tractable the relations that link people to people, people to documents, and documents to documents across spatially or temporally broad discursive networks.

49 Leave a comment on paragraph 49 0
Figure 9. RoSE: Research-Oriented Social Environment project, University of California, Santa Barbara. Sample screens showing social graph and packed radial visualizations of author-document networks.

50 Leave a comment on paragraph 50 0 In a different vein, one of the most interesting uses of blogging technology for scholarly research is CommentPress (used for this essay in the MLA Commons), a theme and plug-in for the open-source WordPress blogging and content management system initially developed by the Institute for the Future of the Book and first applied to prepublish McKenzie Wark’s Gamer Theory.30 CommentPress allows users to read a document and comment on specific paragraphs, thus forming communities of discourse around discrete zones of text. It can also be used to migrate the practice of closed, small-group peer review to that of open, large-group social peer review. As Wardrip-Fruin reflects in the afterword to his Expressive Processing: Digital Fictions, Computer Games, and Software Studies, which he prepublished using a version of CommentPress, “the blog-based review form not only brings in more voices, . . . and not only provides some ‘review of the reviews’ (with reviewers weighing in on the issues raised by others), but is also, crucially, a conversation” (432).31

51 Leave a comment on paragraph 51 0 Scholarship, equipped with Web 2.0, becomes a fully social act.

52 Leave a comment on paragraph 52 1 In the end, it may be that social computing will change the whole paradigm of literary reading and research to make central the social environment of literature. In her 2008 paper “Social Networking Tools for Professional Readers in the Humanities,” Cara Leitch observes the following: “Professional readers are becoming increasingly aware of the potential of social networking tools as scholarly research tools. A successful online reading environment would integrate social networking tools in a way that extends readers’ existing strategies.” The crucial issue in this regard is how we understand “readers’ existing strategies.” It might be hypothesized that in the present, just as in the past, a literary reader’s primary strategy is not to seek authors, documents, or scholars and critics separately. Rather, the goal is to seek knowledge and experience wherever it is vested and most easily accessed. Where does the knowledge and experience of literature occur? It occurs distributed through combinations of authors, documents, readers, and scholar-critics—that is, in the social networks of all the above. Wanting to know something about Shakespeare’s world, one might variously start by reading the plays, finding a book of criticism or history, or—we call it teaching and advising—asking questions of an expert. It really doesn’t matter what the initial point of entry is. The purpose of a fully formed social network of literary study in the future will be to conduct the seeker of literary knowledge and experience from the first point of contact to the dense clusters of authors, documents, readers, and scholar-critics whose relationality constructs knowledge and experience. In her Planned Obsolescence: Publishing, Technology, and the Future of the Academy (fittingly prepublished in CommentPress), Kathleen Fitzpatrick imagines it this way: “[scholars] need to have available to them not simply the library model of texts circulating amongst individual readers but also the coffee house model of public reading and debate” (106).32 Or, to cite Leitch again: “An online community that models a community of practice combines content with communication.”

53 Leave a comment on paragraph 53 0 Additional topics that need to be discussed in relation to reading and social computing await a future when the platforms, applications, tools, and practices sketched above are more developed. Most important, the question will need to be asked: what is the differentia specifica of literary social computing? That is, how does engagement with literature or literary communities inflect, extend, or criticize the culturally dominant tools and practices of vernacular social computing? But such questions can wait until we have more experience with literary social computing.


54 Leave a comment on paragraph 54 0 1. I chose the name core circuit for this concept for two reasons. One is that it borrows metaphorically from computing technology, where the “core” of a CPU processor chip—the central integrated circuit in a computer that reads or executes program instructions—is the unit on the chip that actually does the reading or executing. (High-powered computers today have “multicore” processor chips.) The second reason is that the latent semantic contradiction in the phrase (core means “central,” or “at the heart”; circuit means “path around,” or “circulatory system around” the heart) conforms to my thesis about the need to embed highly centralized models of literary activity in more fully socialized models of such activity. For example, normal literary criticism considers only the interaction between authors, readers, and interpreters (with publishers just sometimes entering the picture). Rarely, if ever, does a critic consider the innumerable other nodes in the circuit that I mention in this essay—e.g., retail outlets (whether yesterday’s peddlers, balladmongers, and booksellers or today’s independent booksellers, chain stores, and e-book retailers).

55 Leave a comment on paragraph 55 0 2. See the glossary in this volume for definitions of technical terms mentioned in this essay. See also The Tech Terms Computer Dictionary.

56 Leave a comment on paragraph 56 0 3. When Web 2.0 was first formulated in such influential essays as Tim O’Reilly’s “What Is Web 2.0,” it was a loose concept made up of a variety of protocols, concepts, and ideologies at different levels of abstraction, all tied to exemplary applications or sites that stood in place of a cohesive definition. Today, the same looseness of concept continues. I simplify by calling social computing the hallmark feature of Web 2.0. Such now widely recognized clichés of Web 2.0 as “wisdom of the crowd” and “collective intelligence” embedded in O’Reilly’s essay support my generalization.

57 Leave a comment on paragraph 57 0 4. “[N]o longer just a calculator, control mechanism, or communication device, the computer becomes a media processor” (Manovich 25–26).

58 Leave a comment on paragraph 58 0 5. I refer to the University of California, Santa Barbara, Social Computing Group, which I helped start with colleagues from my campus’s Departments of Communication, Computer Science, and Sociology as well as the Bren School of Environmental Science and Management in affiliation with the Transliteracies Project (a project I directed from 2005 to 2010 that focused on the relation of social computing to online reading). For the definition of “social computing” that I cite, see “What Is Social Computing?” This definition is a simplification of one suggested by Rama Hoetzlein in a grant proposal written by the Social Computing Group: “Social computing can be defined as the deployment of network communication systems for the purpose of allowing communities of people to interact within particular domains of knowledge for one or more shared goals.”

59 Leave a comment on paragraph 59 0 6. For an influential reflection on (and technical critique of) the concept of the social graph as it is implemented in social-networking systems, see Fitzpatrick with Recordon.

60 Leave a comment on paragraph 60 0 7. I have elsewhere discussed the parallel evolutions of information technology (in the era of the personal computer and network) and literary theory / cultural criticism (in the poststructuralist and postmodern era). See, e.g., Liu, Laws of Cool 4–5; Liu, Local Transcendence 6.

61 Leave a comment on paragraph 61 0 8. In one of the two common types of semiconductor transistors (an NPN transistor), the main circuit connects an “emitter” to a “collector.” But a third current source, the “base,” comes in from the side (as it is normally visualized in circuit diagrams) to switch on or off the electrical flow between the emitter and collector. We may say that this third current is the ur-signal that transforms the power flow of the main circuit into an information signal. Much of traditional literary history effectively functions on this model, where an ur-signal granted an originary or foundational status (e.g, influence, history, mind, class, etc.) is introduced into the discourse expressly to switch what would otherwise be a pure power flow (e.g., the mainstream) or lack of power flow (e.g., the marginal) into a directed hermeneutical signal (e.g., Geist, the spirit of the times, subcultural resistance, etc.). Of course, this is a reductive analysis of literary scholarship. But the analysis is not untrue to the innate reductiveness, which is also to say mythic power, of the stories that literary scholarship ordinarily and necessarily tells about literary revolutions, rises, declines, falls, and so on—e.g., in books about literary history or in undergraduate survey courses. Many of the most powerful works of literary scholarship of the past few decades, after all, have operated on the basis of binary on-off positions of the cultural switch between, for instance, dominance/subversion, majority group / minorities, male/female, etc. Reductiveness at the level of such elementary cultural switches does not prevent complex messages from emerging at higher levels in an interpretive system, any more so than myth prevents nuance and intonation in the telling. (For my earlier use of the transistor analogy and a more detailed development of the literary-historical “switching” model, see Liu, Local Transcendence 194.)

62 Leave a comment on paragraph 62 0 9. For Johns on book piracy, see, e.g., 3–4; for his dispute with Eisenstein’s notion of fixity, see 10–20.

63 Leave a comment on paragraph 63 0 10. For my detailed discussion of subversion and new historicism (as well as such cousins as new cultural history), see Liu, Local Transcendence, chs. 1–2.

64 Leave a comment on paragraph 64 0 11. For a discussion of the many roles involved in the literary circuit (including a similar use of the word circuit), see Darnton’s chapter, “What Is the History of Books?” in his The Case for Books (175–206). In proposing “a general model for analyzing the way books come into being and spread through society,” Darnton describes “a communications circuit that runs from the author to the publisher . . . , the printer, the shipper, the bookseller, and the reader.” He continues, “the reader completes the circuit because he influences the author both before and after the act of composition. Authors are readers themselves [engaged in] reading and associating with other readers and writers. . . . [The author] addresses implicit readers and hears from explicit reviewers. So the circuit runs full cycle” (179–80). See also Darnton’s diagram of the multiple positions in the communications circuit of books (182). (By comparison, fig. 5 is designed only to suggest the concept of a larger field of literary positions, not sketch out the actual logical or historical flows of transmission between positions.) A total census of the roles in the literary circuit would also include not just the domains of production, circulation, consumption, instruction, and scholarship, but—mindful of Prescott’s discussion of “being a curator, an academic, and a librarian”—also curation and heritage.

65 Leave a comment on paragraph 65 0 12. For a discussion of the role of publishers, printers, and booksellers in early modern ballads, see Nebeker’s dissertation, “The Broadside Ballad and English Literary History, 1540–1700.”

66 Leave a comment on paragraph 66 0 13. See also Fitzpatrick’s discussion of “reading and the communications circuit,” which cites similar observations by scholars of the history of reading (Planned Obsolescence 104–09).

67 Leave a comment on paragraph 67 0 14. The whole cluster of articles on “community reading” in this issue of PMLA is of interest in this context: e.g., Levin and Davis; Wiedmer.

68 Leave a comment on paragraph 68 0 15. Nunberg writes about blogs, “That informal style recalls the colloquial voice that Addison and Steele devised when they invented the periodical essay in the early 18th century, even if few blogs come close to that in artfulness. Then too, those essays were written in the guise of fictive personae like Isaac Bickerstaff and Sir Roger de Coverly, who could be the predecessors of pseudonymous bloggers like Wonkette, Atrios, or Skippy the Bush Kangaroo, not to mention the mysterious conservative blogger who goes by the name of Edward Boyd.” In a note quoting an earlier piece of his, Nunberg compares blogs to the novel: “It’s what the novel was trying to achieve when eighteenth-century writers cobbled it together out of subliterary genres like personal letters, journals, and newspapers, with the idea of reproducing the inner and outer experience that makes up daily life.”

69 Leave a comment on paragraph 69 0 16. “Rule of many” and “wisdom of crowds”—like “collective intelligence,” “hive mind,” “crowdsourcing,” and so on—are dominant memes, also known as clichés, of Web 2.0. On “wisdom of crowds,” e.g., see Surowiecki.

70 Leave a comment on paragraph 70 0 17. For an example of interesting new Web 2.0 possibilities for literature, see the Twitter fiction of Bushman (e.g., Good Captain). See also ’“Storytelling 2.0,” an interview with Bushman.

71 Leave a comment on paragraph 71 0 18. See the chapter “The Development of Social Network Analysis” in Scott 7–37. See also Freeman. For my initial introduction to social-network theory, I owe thanks to Clayton Childress for his presentations on the topic in the gateway seminar for the technology and society PhD emphasis at University of California, Santa Barbara, and to the University of California, Santa Barbara, Social Computing Group. I have also benefitted from the recent work of such current or former colleagues conducting advanced research on methods for studying online social networks as Jennifer Earl, Miriam Metzger, and Ben Zhao.

72 Leave a comment on paragraph 72 0 19. My description of WordHoard adapts the summary I wrote for Liu, Toy Chest.

73 Leave a comment on paragraph 73 0 20. My thanks to Amanda Phillips for initially bringing this essay and also the book by Franzosi to my attention. For her description of the Bearman and Stovel article, see the annotated bibliography she contributed to a course I taught in 2010 (Phillips).

74 Leave a comment on paragraph 74 0 21. I have simplified Franzosi’s encoding typography in this example.

75 Leave a comment on paragraph 75 0 22. My gratitude to Franco Moretti, Matthew Jockers, Ryan Heuser, Kathryn VanArendonk, and others at the Stanford Literary Lab who hosted me during a visit to the lab on 21 May 2012, when members of the lab presented their ongoing research. During the presentations that day, Moretti spoke on the need to find ways to make semantic and stylistic analysis matter to the social-network analysis of literature, and Heuser spoke of his “window” method of isolating sequential units of social interaction in plays.

76 Leave a comment on paragraph 76 0 23. My thanks to Piper for an advance look at his and Algee-Hewitt’s Werther Effect and Literary Topologies projects, along with manuscript copies of essays related to the projects.

77 Leave a comment on paragraph 77 0 24. For instance, the following are a few pressing technical and conceptual challenges (some currently insuperable) to social science and humanistic approaches to social computing:

  • 78 Leave a comment on paragraph 78 0
  • Scaling up data mining to handle the extremely large and dynamic corpora of social computing.
  • Creating or improving a standard, extensible metadata scheme for annotating social networks and their behaviors. We need a scheme at once more comprehensive and fine-grained, for example, than the current “prosopography” (informally, “personography”) of the Text Encoding Initiative’s P5 Guidelines, the metadata protocols of Facebook Markup Language (now deprecated), or the metadata-cum-API protocols of recent entrants in the so-called open social networking movement.
  • Reconciling top-down, formal-ontology metadata with the bottom-up, informal tagging systems characteristic of social computing.
  • Creating authority-adjudication systems that allow multiple communities of interest—including, but not limited to, experts—to accredit or filter socially computed information.
  • Developing a new generation of information-visualization and other pattern-recognition methods to manage the contact zone between macroreading and close reading—e.g, by creating a gradient between the two, or by facilitating smarter recursion such that macropatterns suggest directions for close reading even as behaviors of close reading modify the algorithms of macrodiscovery.
  • Incorporating historical people—e.g., Shakespeare—in social-networking systems so as to represent their continuing intellectual agency. For example, might Shakespeare have a “profile page” in a social-networking application or, through creative data mining of his works in interaction with those of later researchers and even contemporary events, be able to blog his views about today’s tragedies of state?
  • Incorporating contemporary “other” people of research interest—e.g., racially, ethnically, nationally, and economically other people (including those not on the Internet)—so as to represent their undeniable “power of identity,” as Castells calls it, in network society. (See Castells’s discussion of the “power of identity” in networked society in 1: 23–25 and vol. 2 [The Power of Identity].) For instance, might there be a computationally tractable method of “fluid ontology,” as Srinivasan and Huang call it, for meshing local cultural knowledge systems with global museum classification systems (see also Srinivasan, Becvar, Boast, and Enote)?
  • Correlating the structure of the social networks constructed or described in newspapers, plays, novels, TV shows, etc. with the “real-world” social networks that produce, consume, or otherwise shape the way such works imagine social networks. For example, what is the relation of the social networks imagined in a Shakespeare play to those of the court, city, and theater world of Shakespeare’s time? Or of the social network imagined in a TV sitcom like Friends to those of the writers, producers, directors, advertisers, and viewers of the show?
  • Evolving new kinds of visualization methods for understanding and analyzing large-scale social networks with a high density of nodes and edges that just become visual noise in standard node-and-link social graphs. See, e.g., work on matrix-adjacency visualizations and hybrid node-link and matrix visualizations (Henry and Fekete; Henry, Fekete, and McGuffin).

79 Leave a comment on paragraph 79 0 25. For the student projects referred to in this essay, see the team projects on the Web sites for the 2008 and 2009 versions of my Literature+ courses. I taught the course in 2009 with my colleague James Donelan. See also my essay “Literature+” for a discussion of the course concept. For the students’ project on The Merchant of Venice that I refer to here, see their explanatory page (“Venetian”).

80 Leave a comment on paragraph 80 0 26. For the student LiveJournal adaptation of Chaucer’s General Prologue referred to here, see the explanation page for their “Canterbury Blogs” (“Timeline Project”).

81 Leave a comment on paragraph 81 0 27. See the explanatory page of the student project, “Romeo and Juliet: A Facebook Tragedy.”

82 Leave a comment on paragraph 82 0 28. For surveys, reviews, and discussion of Academia.edu and other academic social networking sites, see Leeder; Leitch; and Hudson.

83 Leave a comment on paragraph 83 0 29. RoSE was started as part of the University of California system’s Transliteracies Project in 2008–10 and subsequently developed further at the University of California, Santa Barbara, on a NEH Digital Humanities Start-Up Grant in 2011–12 (with me as director and Rama Hoetzlein and Rita Raley as codirectors).

84 Leave a comment on paragraph 84 0 30. In 2006, Wark prepublished in CommentPress a draft of his Gamer Theory under the title GAM3R 7H30RY. On CommentPress, see Hovey and Hudson; Knight.

85 Leave a comment on paragraph 85 0 31. For the version of CommentPress used to prepublish Wardrip-Fruin’s Expressive Processing, see Douglass. The draft of Wardrip-Fruin’s book appeared in CommentPress on the Grand Text Auto blog beginning with his post “Expressive Processing: An Experiment in Blog-Based Peer Review.”

86 Leave a comment on paragraph 86 0 32. I originally read this passage in Fitzpatrick’s post entitled “Reading and the Communications Circuit” from her CommentPress prepublication of Planned Obsolescence. My thanks to Cara Leitch for leading me to this quotation.

Works Cited

Bakhtin, M. M. The Dialogic Imagination: Four Essays. Trans. Caryl Emerson and Michael Holquist. Ed. Holquist. Austin: U of Texas P, 1981. Print.

Barthes, Roland. S/Z. Trans. Richard Miller. New York: Hill, 1974. Print.

Bearman, Peter S., and Katherine Stovel. “Becoming Nazi: A Model for Narrative Networks.” Poetics 27 (2000): 69–90. Print.

Bérubé, Michael, Hester Blum, Christopher Castiglia, and Julia Spicher Kasdorf. “Community Reading and Social Imagination.” PMLA 125.2 (2010): 418–25. Print.

Blair, Ann. “Note Taking as an Art of Transmission.” Critical Inquiry 31.1 (2004): 85–107. Print.

Busa, Roberto, et al. Index Thomisticus. Ed. and trans. Eduardo Bernot and Enrique Alarcón. Società CAEL, 2005. Web. 2 Aug. 2009. <http://www.corpusthomisticum.org/it/index.age>.

Bushman, Jay. The Good Captain. N.p., 1 Mar. 2008. Web. 12 July 2012. <http://jaybushmanportfolio.tumblr.com/post/17144142761/the-good-captain>.

———. “Storytelling 2.0: The Epic Poet of Twitter.” Interview by Richard Fisher. New Scientist. Reed Business Information, 16 Nov. 2010. Web. 18 Dec. 2010. <http://www.newscientist.com/blogs/culturelab/2010/11/storytelling-20-the-epic-poet-of-twitter.html>.

Castells, Manuel. The Information Age: Economy, Society and Culture. 3 vols. Malden: Blackwell, 1996–97. Print.

Certeau, Michel de. The Practice of Everyday Life. Trans. Steven Rendall. Berkeley: U of California P, 1984. Print.

Collex.” NINES (Networked Infrastructure for Nineteenth-Century Electronic Scholarship). NINES.org, n.d. Web. 20 Dec. 2009. <http://www.nines.org/about/software/collex.html>.

Cultural Analytics. Software Studies Initiative, U of California, San Diego, Sept. 2008. Web. 8 Dec. 2010. <http://lab.softwarestudies.com/2008/09/cultural-analytics.html>.

Darnton, Robert. The Case for Books: Past, Present, and Future. New York: Public Affairs, 2009. Print.

Derrida, Jacques. “Living On / Border Lines.” Trans. James Hulbert. Deconstruction and Criticism. By Harold Bloom et al. New York: Seabury, 1979. 75–176. Print.

Douglass, Jeremy. CommentPress 1.4.1. Software Studies Initiative, U of California, San Diego, 1 Jan. 2008. Web. 18 Dec. 2010. <http://lab.softwarestudies.com/2008/04/commentpress-141.html>.

Eisenstein, Elizabeth L. The Printing Revolution in Early Modern Europe. Cambridge: Cambridge UP, 1983. Print.

Facebook Markup Language (FBML).” Facebook. Facebook.com, n.d. Web. 9 July 2012. <http://developers.facebook.com/docs/reference/fbml/>.

Finn, Ed. “Becoming Yourself: The Afterlife of Reception.” Stanford Literary Lab. Pamphlet 3. Stanford Lit. Lab, 15 Sept. 2011. Web. 6 July 2012.

Fitzpatrick, Brad, with David Recordon. “Thoughts on the Social Graph.” Bradfitz.com. N.p., 17 Aug. 2007. Web. 19 June 2008.

Fitzpatrick, Kathleen. Planned Obsolescence: Publishing, Technology, and the Future of the Academy. New York: New York UP, 2011.

———. “Reading and the Communications Circuit.” MediaCommons Press. MediaCommons, 15 Sept. 2009. Web. 18 Dec. 2010. <http://mediacommons.futureofthebook.org/mcpress/plannedobsolescence/three-texts/reading-and-the-communications-circuit/>.

Foucault, Michel. “What Is an Author?” Trans. Donald F. Bouchard and Sherry Simon. Language, Counter-memory, Practice: Selected Essays and Interviews. Ed. Bouchard. Ithaca: Cornell UP, 1977. 113–38. Print.

Franzosi, Roberto. Quantitative Narrative Analysis. Los Angeles: Sage, 2010. Print.

Freeman, Linton C. The Development of Social Network Analysis: A Study in the Sociology of Science. Vancouver: Empirical, 2004. Print.

Henry, Nathalie, and Jean-Daniel Fekete. “MatLink: Enhanced Matrix Visualization for Analyzing Social Networks.” N.p., n.d. Web. 6 Sept. 2012. <http://research.microsoft.com/en-us/um/people/nath/docs/Henry_interact07.pdf>. Orig. pub. as Lecture Notes in Computer Science (Proceedings of the 13th IFIP TC13 International Conference on Human-Computer Interaction—INTERACT ’07: 13th IFIP TC13 International Conference, Rio de Janeiro, Brazil, September [10–14], 2007) 4663 (2007): 288–302.

Henry, Nathalie, Jean-Daniel Fekete, and Michael J. McGuffin. “NodeTrix: A Hybrid Visualization of Social Networks.” IEEE Transactions on Visualization and Computer Graphics 13.6 (2007): 1302–09. Web. 1 Nov. 2012. <http://www.aviz.fr/~nhenry/docs/Henry-InfoVis2007.pdf>.

Hovey, Pehr, and Renee Hudson. “CommentPress.” Transliteracies Project: Research in the Technological, Social, and Cultural Practices of Online Reading. U of California, 22 Apr. 2009. Web. 7 Aug. 2009. <http://transliteracies.english.ucsb.edu/post/research-project/research-clearinghouse-individual/research-papers/commentpress-research-paper>.

Hudson, Renee. “Academia.edu.” Transliteracies Project: Research in the Technological, Social, and Cultural Practices of Online Reading. U of California, 20 Nov. 2009. Web. 18 Dec. 2010. <http://transliteracies.english.ucsb.edu/post/research-project/research-clearinghouse-individual/research-reports/academia-edu>.

Hunicke, Robin. “Jane on Facebook and LFG.” Gewgaw. 4 Sept. 2007. Web. 21 Dec. 2010. <http://www.cs.northwestern.edu/~hunicke/blog/?p=41750>.

Ivanhoe. By Jerome McGann, Johanna Drucker, et al. N.p., n.d. Web. 16 Oct. 2009. <http://www.ivanhoegame.org/>.

Johns, Adrian. The Nature of the Book: Print and Knowledge in the Making. Chicago: U of Chicago P, 1998. Print.

Knight, Kim. “CommentPress.” Transliteracies Project: Research in the Technological, Social, and Cultural Practices of Online Reading. U of California, 7 May 2008. Web. 7 Aug. 2009. <http://transliteracies.english.ucsb.edu/post/research-project/research-clearinghouse-individual/research-reports/commentpress-2>.

Landow, George P. Hypertext: The Convergence of Contemporary Critical Theory and Technology. Baltimore: Johns Hopkins UP, 1992. Print.

Latour, Bruno. Reassembling the Social: An Introduction to Actor-Network-Theory. Oxford: Oxford UP, 2005. Print.

Leeder, Kim. “Social Networking with a Brain: A Critical Review of Academic Sites.” In the Library with the Lead Pipe. N.p., 10 Dec. 2008. Web. 15 Sept. 2009. <http://www.inthelibrarywiththeleadpipe.org/2008/social-networking-with-a-brain-a-critical-review-of-academic-sites/>.

Leitch, Cara. “Social Networking Tools for Professional Readers in the Humanities.” Social Networking Tools for Professional Readers in the Humanities: SDH-SEMI 2008 (Annual Meeting of the Society for Digital Humanities / Société pour l’étude des médias interactifs). U of British Columbia, Vancouver. 3 June 2008. Manuscript of address.

Levin, Amy, and Phoebe Stein Davis. “‘Good Readers Make Good Doctors’: Community Readings and the Health of the Community.” PMLA 125.2 (2010): 426–36. Print.

Liu, Alan. “The End of the End of the Book: Dead Books, Lively Margins, and Social Computing.” Michigan Quarterly Review 48.4 (2009): 499–520. Print.

———. Laws of Cool: Knowledge Work and the Culture of Information. Chicago: U of Chicago P, 2004. Print.

———. “Literature+.” Currents in Electronic Literacy (2008): n. pag. Web. 19 Jan. 2009. <http://currents.cwrl.utexas.edu/Spring08/Liu>.

———, ed. Literature+: New Media and Cross-Disciplinary Methods of Literary Interpretation. U of California, Santa Barbara, 2008. Web. 20 Dec. 2009. <http://english149-w2008.pbworks.com/>.

———. Local Transcendence: Essays on Postmodern Historicism and the Database. Chicago: U of Chicago P, 2008. Print.

———, ed. Toy Chest (Online or Downloadable Tools for Building Projects). U of California, Santa Barbara, 19 Oct. 2007. Web. 20 Dec. 2009.

———, ed. Voice of the Shuttle. 1994. U of California, Santa Barbara, n.d. Web. 17 Oct. 2009.

Liu, Alan, and James Donelan, eds. “Literature+: New Media and Cross-Disciplinary Methods of Literary Interpretation.” U of California, Santa Barbara, 2009. Web. 20 Dec. 2009. <http://english149-w2009.pbworks.com/>.

Manovich, Lev. The Language of New Media. Cambridge: MIT P, 2001. Print.

McCarty, Willard. Humanities Computing. Basingstoke: Palgrave, 2005. Print.

McKenzie, D. F. Bibliography and the Sociology of Texts. Cambridge: Cambridge UP, 1999. Print.

Mohr, John W., and Vincent Duquenne. “The Duality of Culture and Practice: Poverty Relief in New York City, 1888–1917.” Theory and Society 26.2-3 (1997): 305–56. Print.

Moretti, Franco. Graphs, Maps, Trees: Abstract Models for a Literary History. London: Verso, 2005. Print.

———. “Network Theory, Plot Analysis.” Stanford Literary Lab. Pamphlet 2. Stanford Lit. Lab, 1 May 2011. Web. 6 July 2012.

Mutton, Paul. “Shakespeare Social Networks.” Jibble.org. Mutton, 2001–10. Web. 21 Dec. 2012.

Nebeker, Eric. “The Broadside Ballad and English Literary History, 1540–1700.” Diss. U of California, Santa Barbara, 2009. Web. 11 July 2012.

Nunberg, Geoffrey. “Blogging in the Global Lunchroom.” Geoffrey Nunberg. School of Information, U of California, Berkeley, 20 Apr. 2004. Web. 22 Aug. 2010. <http://people.ischool.berkeley.edu/~nunberg/lunchroom.html>.

Ong, Walter J. Orality and Literacy: The Technologizing of the Word. London: Methuen, 1982. Print.

O’Reilly, Tim. “What Is Web 2.0: Design Patterns and Business Models for the Next Generation of Software.” O’Reilly Media, 30 Sept. 2005. Web. 8 Sept. 2006.

Phillips, Amanda. “Annotated Bibliography Assignment.” “Literature+: New Media and Cross-Disciplinary Methods of Literary Interpretation.” 11 Feb. 2010. Web. 17 Dec. 2010. <http://english236w2010.pbworks.com/w/page/23055101/Bibliography-by-Amanda-Phillips>.

PieSpy Social Network Bot. By Paul Mutton. Jibble.org. Mutton, 2001–10. Web. 21 Dec. 2012. <http://www.jibble.org/piespy/>.

Piper, Andrew. “Topologies of Literature: Network Theory and Literary Study.” McGill U, n.d. Web. 17 Dec. 2010. <http://piperlab.mcgill.ca/files/Topologies.htm>.

Piper, Andrew, and Mark Algee-Hewitt. Literary Topologies. Literary Topologies, n.d. Web. 9 July 2012. <http://literarytopologies.org/>.

———. “The Werther Effect 1: Goethe Topologically.” Distant Readings / Descriptive Turns: Topologies of German Culture in the Long Nineteenth Century. Ed. Matt Erlin and Lynne Tatlock. Rochester: Camden, forthcoming. N. pag. Manuscript.

Prescott, Andrew. “An Electric Current of the Imagination: What the Digital Humanities Are and What They Might Become.” Journal of Digital Humanities 1.2 (2012) : n. pag. Web. 30 June 2012.

Romeo and Juliet: A Facebook Tragedy.” Literature+: New Media and Cross-Disciplinary Methods of Literary Interpretation. U of California, Santa Barbara, 2008. Web. 20 Nov. 2012. <http://english149-w2008.pbworks.com/w/page/19011402/Romeo%20and%20Juliet%3A%20A%20Facebook%20Tragedy>.

RoSE: Research-Oriented Social Environment. Transliteracies Project, U of California, 2012. Web. 14 July 2012. <http://rose.english.ucsb.edu/>.

Samuels, Lisa, and Jerome J. McGann. “Deformance and Interpretation.” New Literary History 30.1 (1999): 25–56. Print.

Scott, John. Social Network Analysis: A Handbook. 2nd ed. Los Angeles: Sage, 2000. Print.

SEASR. SEASR.org, n.d. Web. 20 Dec. 2009.

———. “SEASR Analytics for Zotero.” SEASR. SEASR.org, n.d. Web. 20 Dec. 2009. <http://seasr.org/documentation/zotero/>.

Sherman, William H. Used Books: Marking Readers in Renaissance England. Philadelphia: U of Pennsylvania P, 2007. Print.

SNAC: The Social Networks and Archival Context Project. Inst. for Advanced Technology in the Humanities, U of Virginia, n.d. Web. 13 July 2012.

Srinivasan, Ramesh, Katherine M. Becvar, Robin Boast, and Jim Enote. “Diverse Knowledges and Contact Zones within the Digital Museum.” Science, Technology, and Human Values 35.5 (2010): 735–68. Print.

Srinivasan, Ramesh, and Jeffrey Huang. “Fluid Ontologies for Digital Museums.” International Journal on Digital Libraries 5.3 (2005): 193–204. Print.

Stallybrass, Peter, Roger Chartier, John Franklin Mowery, and Heather Wolfe. “Hamlet’s Tables and the Technologies of Writing in Renaissance England.” Shakespeare Quarterly 55.4 (2004): 379–419. Print.

Surowiecki, James. The Wisdom of Crowds: Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business, Economies, Societies, and Nations. New York: Doubleday, 2004. Print.

TAPoR. Vers. 2.0. TAPoR, McMaster U, 2012. Web. 6 July 2012.

The Tech Terms Computer Dictionary. TechTerms.com, 2012. Web. 10 July 2012.

“TEI: P5 Guidelines.” Text Encoding Initiative. TEI, n.d. Web. 20 Dec. 2009. <http://www.tei-c.org/Guidelines/P5/>.

“Timeline Project.” Literature+: New Media and Cross-Disciplinary Methods of Literary Interpretation. U of California, Santa Barbara, 2008. Web. 20 Nov. 2012. <http://english149-w2008.pbworks.com/w/page/19011420/Timeline%20Project>.

“The Venetian Project.” Literature+: New Media and Cross-Disciplinary Methods of Literary Interpretation. U of California, Santa Barbara, 2009. Web. 17 Dec. 2012. <https://sites.google.com/site/thevenetianproject2/>.

Wardrip-Fruin, Noah. “Expressive Processing: An Experiment in Blog-Based Peer Review.” Grand Text Auto. Grand Text Auto, 22 Jan. 2008. Web. 18 Dec. 2010. <http://grandtextauto.org/2008/01/22/expressive-processing-an-experiment-in-blog-based-peer-review/>.

———. Expressive Processing: Digital Fictions, Computer Games, and Software Studies. Cambridge: MIT P, 2009. Print.

Wark, McKenzie. GAM3R 7H30RY. Vers. 1.1. Inst. for the Future of the Book, n.d. Web. 18 Dec. 2010. <http://www.futureofthebook.org/gamertheory/>.

———. Gamer Theory. Cambridge: Harvard UP, 2007. Print.

“What Is Social Computing?” Social Computing. UCSB Social Computing Group, 8 Nov. 2008. Web. 12 Sept. 2012.

Wiedmer, Caroline. “The Politics of Reading Street Newspapers.” PMLA 125.2 (2010): 437–42. Print.

page 3

Source: http://commons.mla.org/from-reading-to-social-computing/