Page 14 →Chapter 1 Distributed
Curatorial Practices
To get a sense of how definitions of this sort of work were constructed early on in the Western encyclopedic tradition, we begin with our eighteenth-century source: the primary curator of the Cyclopædia. In the preface that appeared in the first two editions, Ephraim Chambers wrote at length about the problems of curating a reference text that aspired to cover the whole of human knowledge. Devoting more than a decade of his life to gathering and filtering previously published scientific, legal, and religious texts before recomposing the results into what became the first edition of this groundbreaking encyclopedia had given him plenty of time to think deeply about the philosophical, ethical, and practical problems the project posed.
The communal nature of his project was absolutely central, he argued: “The work is what it ought to be, a collection—not the produce of a single brain, for that would go but a little way, but of a whole Commonwealth.”1 The pervasive textual borrowing necessary for producing what he called “a work so disproportionate to a single person’s experience, and which might have employed an academy”2 required debts to a multitude of scholarly resources. No available author was spared, regardless of era, creed, nationality, or area of expertise, and no resource was left unrifled, whatever its genre. Consequently he saw himself as only the most recent in a long line of scholars, “an heir to a large patrimony, gradually raised by the industry and endeavors of a long Race of Ancestors.”3 In effect each of these prior authors and texts functions as part of an asynchronous collective, contributing across time to the project of expanding and collecting information on the known world. The resulting Cyclopædia “is not mine, ’tis everybody’s: the mixed issue of a thousand loins,”4 he wrote, adding that there are very few pages in the final text that do not include several instances of this type of borrowing—so few, in fact, that he would not attempt to list the pages that might meet modern Page 15 →standards of originality.5 He went so far as to invite the reader to assume that any given article in the work was drawn from at least one other text, and indeed he describes the Cyclopædia as “derived” from these materials. His use of this term was commensurate with our contemporary North American legal definition of a derivative work, a work that recasts or transforms one or more preexisting works. By openly acknowledging this extensive borrowing, he effectively circumvented potential charges of plagiarism.
This practice of large-scale borrowing from other texts for the purposes of recomposition is the privilege of the encyclopedist, Chambers argued. Common knowledge belonged to and in the commons. Similarly thoughts that were committed to circulation in print became properly available for the use of others, in his estimation: “’Tis idle to pretend anything of property in things of this nature. To offer a thing to the public, and yet pretend a right reserved therein to oneself, if it not be absurd, yet it is sordid. The words we speak—nay, the breath we emit—are not more vague and common than our thoughts, when divulged in print. You may as well prohibit people to use the light that shines in their eyes because it comes from your candle.”6 Positioning knowledge as a nonrivalrous good in the preface accomplishes several important rhetorical goals. First, it directly counters the pervasive arguments that prominent denizens of London’s publishing community were making concerning intellectual property’s equivalence with physical property. In doing so Chambers made an ethical appeal for his reuse of prior texts, claiming that it honored the essential nature of information as a social good produced by and for the community. If free circulation and use are in fact the highest purpose of information, then his own reuse and his curated product were by extension also both ethical and legal.7
What is most important, he established his contention that knowledge is ecological, produced by collectives who constantly collect, refine, and circulate it. The labor of curation is distributed across both time and space, carried out recursively by scholars, publishers, and booksellers as they curate and distribute texts, reusing and expanding the same information base. The growth of knowledge is evolutionary, evolving as each iteration is introduced into the ecology. This point in turn has ramifications for his conceptualization of rhetorical invention and the agency available to writers in the invention process. His explication of the pervasiveness of creative influence reaches beyond descriptions of encyclopedic authorship, claiming that writers in all genres work to construct texts from a patchwork of prior influences. This assertion bolstered his argument about the basic ethics of authorship and, in the process, prefigured prominent twentieth-century theories of influence and originality. The essential nature of invention is dependent on the information ecologies in which the writer lives and moves, he argued:
Page 16 →Even at most, what we are said to invent, is only what results or arises from something already in us. . . . There is no more real invention in the poet than in the tapestry or mosaic worker, who ranges and combines the various colored materials furnished to his hand, so as to make an assemblage or picture, which before had no existence. . . . In effect, the inspiration of the poet amounts to little more than relating things that are naturally incongruous. He presents new objects, new worlds, but ’tis only by differently combining the parts of the old one. He does not make any thing, he only patches. He does not invent, he only transposes. Nor has he the least power to move, other than what he derives from the novelty and strangeness of his combinations. (xii)
All writers, then, create compositions based on prior influences, contributing by developing novel arrangements and presentations rather than inventing original work through solitary processes. If all writing is always social, Chambers argued, then there can indeed be real merit in recomposed texts. He positioned his own foundational contributions as dependent on improving these prior resources by combining information found in individual texts as well as adding the latest information on each topic, thus transforming it into a richer, more finely detailed product. As someone who was intimately connected to the natural philosophy community that coalesced around the Royal Society and the United Grand Lodge of England, he was well positioned to be aware of and incorporate the latest scientific thought in his editions.
He had arranged this information into a tome that purported to form a complete, systematic course of study for any literate individuals who wanted to become educated and could get their hands on the two volumes that made up the encyclopedia. These books, which introduced the interface innovation of cross-indexing and helped popularize alphabetization in early modern English reference texts, would help the reader learn to think critically, “forming a sound mind, i.e. a system of perceptions, and notions agreeing to the systems of things.”8 By working through the systematic arrangement of this text, then, the reader would also come to see the fundamental truths in Newton’s then-controversial view of a systematic, clockwork universe. Further, through the curator’s labor of collecting, recomposing, and recirculating knowledge, contemporary knowledge would continue to be improved and extended. In a period that saw tremendous scientific discovery and the slow build-up to evolutionary theory, Chambers obliquely proposed that encyclopedic texts were best built for the sort of incremental development that could be facilitated by “the advantages of a continued discourse.”9 This view depended on distributed contributions not just to the creation of textual content, but also to its circulation through reading, lending, Page 17 →and discussion in private societies as well as public venues such as coffeehouses.
Rather than aligning himself primarily with the first rhetorical canon, invention, Chambers defined his work as more closely integrated with the second canon, arrangement. In the opening passage of the preface, he described his primary compositional task as one of filtering and organizing materials, then developing a usable information architecture that would be navigated intuitively by the reader: “Such are the sources from whence the materials of the present work were derived, which, it must be allowed, were rich enough not only to afford plenty, but even profusion. . . . The chief difficulty lay in the form, in the order, and [the] economy of the work: to dispose such a variety of materials in such manner as not to make a confused heap of incongruous parts, but one confident whole.”10 This focus on composing not just narrative text but also the larger information structure is but one element that distinguished these reference volumes from others, which frequently relied on topical groupings rather than alphabetization and cross-indexing as their organizational schema. While he listed himself as the Cyclopædia’s sole formal author, Chambers explicitly disavowed ownership of the public knowledge he had collected from myriad sources and curated as a new encyclopedia. Authorship and ownership of this sort of text were not intertwined, he argued.
In his estimation, identifying the legal author of his editions was not at all difficult because there existed a valid claim to sole authorship based on his own considerable curatorial skill and labor. His initial work was largely undertaken alone in the back rooms of a globe maker’s shop, where he composed the first edition instead of attending to his apprentice map-making duties. His employer supported him, but it was lonely work that required focused thinking about how to compose and structure the text and then creative labor to implement it: “and here it must be confessed there was not assistance to be had,” he remembered, “but I was forced to stand wholly on my own bottom.”11 His approval of his name as the authorial signature that was prominently printed in large, red type on the title page suggests that he both accepted responsibility for the text and expected recognition for what he understood as valid compositional work that was based heavily on arrangement and recomposition. He got both in the form of monetary reward from his publisher, his reception at the court of George II, and his formal election to the Royal Society. His reflection on these seemingly conflicting claims reveals the nuances of textual curation as authorial labor and as a considered stance on the ethics of knowledge production and distribution.
Throughout the process of developing the first edition, Chambers served as the only identifiable and credited curator of the project. As he prepared the second edition, he sought to shift his role to an extent by inviting contributions from Page 18 →the public. He would still serve as the central curator, but by enrolling contributors from as many walks of life as possible, he would coordinate a collective project that brought together multiple social networks that had coalesced around not just class strata but also skill sets. By 1734 he was no longer quite so concerned about justifying the social process of building an encyclopedia. Rather he intended to make it even more explicitly and productively social, and so he placed magazine advertisements soliciting article submissions from the reading public. He also published a pamphlet entitled “Some Considerations Offered to the Public, Preparatory to a Second Edition.”
The concept of public in England then typically meant something very different from our twenty-first-century American notion of a broad, egalitarian public. Reference materials and codified knowledge were most often accessed through extensive home libraries, religious institutions, universities, and societies. They were typically accessible to the landed gentry and to some extent members of the professional class who had the funds, social connections, and leisure time to access these texts. Formal education of women was limited, infrequent, and not considered a good use of family financial resources. Consequently it was most often well-connected men who were understood as having acquired real, substantive knowledge, and the aristocracy occupied many (although by no means all) of the formal roles within the Royal Society. The society also inducted a number of professional men during the time that Chambers was nominated, perhaps as a gesture toward a limited intellectual democracy.12
Chambers sought more egalitarian contributions to his second edition, and when he addressed the public, he had the broadest possible audience in mind. After opening with a consideration of what it meant to undertake revision and expansion of a comprehensive encyclopedia, he clarified his understanding of the project’s audience and actively invited contributions from the public, much like an analog version of our modern Wikipedia. The Cyclopædia was, he wrote, for “all of those concerned in the acquisition of learning, that is of all persons in general, for I know of no rank, condition, or even sex, that is dispensed from the necessity of cultivating and improving their own minds” (2). By encouraging collaborators from as many walks of life as possible, he opened up the possibility of significantly widening the collective that developed this project. This was his invitation, which included the rather radically defined public he imagined:
In this invitation are included persons of every rank, profession, and degree of knowledge: men of letters, of business, and of pleasure; the Universities, the Court, Country, Army, and Navy. Not a college, a chapter, a mercantile company, a ship, scarce a house, or even a man, but may contribute his quota to the public instruction. . . . The less learned may Page 19 →here lay aside their apprehensions of appearing in a work of literature; being masters of the subject, they need not be solicitous as to the style and manner. Many even among the illiterate may here find place, and be of use to men of the profoundest learning. They will find an amanuensis in me, who shall even think it an honor to be dictated to by some who can neither write nor read. Numerous things are wanted from the last quarter—and the more so, as they are not extant in books, libraries, and cabinets of curiosity, but hid in shops, garrets, cellars, mines, and other obscure places, where men of learning rarely penetrate. Rich fields of science lie thus neglected underground: trades, crafts, mysteries, practices, short ways, with the whole vast apparatus of unwritten philosophy. (3)
Here we see Chambers advancing a radical vision of intellectual democracy, one that placed the advancement of knowledge at the center of all human concerns while issuing an open invitation to contributors regardless of class, gender, profession, or literacy levels. In this definition everyone is a potential worker with valid knowledge to contribute to the collective project. The crowd, he argued, can cover the breadth of modern knowledge that a single scholar simply cannot, given the exigencies of information overload.13 After quickly acknowledging the more obvious resources of the aristocracy and distinguished institutions, he included also the professional middle class he worked and lived among. But he devoted a significant amount of print space to the working classes and their unwritten craft knowledge. The knowledge of these groups is just as essential as the knowledge of anyone else, he argued, and represents a central contribution to the Enlightenment project. In making this call, he acknowledged the communal, hive-like nature of encyclopedia building, ratifying it with his signature and the brand of his successful, sought-after encyclopedia. By working to broaden the contributing community and resources, he also subtly highlighted the transformative labor of the encyclopedic author, who filters gathered material and recomposes it into a new (and theoretically improved) text.
He asked not just for contributions of new information, but for contributions of all types. He invited subject-matter experts to offer corrections to existing articles and all readers to submit editorial corrections to the wording of articles so that “any point [might be] set in a better light than unusual, or brought into a shorter compass, or reduced to a juster principle, or disposed of in a more convenient method, or pursued to a greater length, to communicate the same by this means to the public.”14 He also asked that other readers take note “as any thing occurs in the course either of their reading or speculation” and contribute information on any topic, term, references, or other knowledge that were both already extant in the text or that should be included.
Page 20 →Chambers served as the central curator of all these submissions, a job made necessary by the technological constraints of a paper-based submission and manuscript preparation system. No archival records of the submissions remain, but there is evidence that he managed a tremendous amount of information in preparation for the second edition. He employed two amanuenses, Mr. Ayrey (from 1728 to 1733) and Mr. Macbean,15 whose duties may have included assisting in research and expansion of the manuscript. Ayrey recalled that during his time of service “he copied near 20 folio volumes, which, Mr. Chambers used to say, comprehended materials for more than 30 volumes of that size, though he at the same time added, they would neither be sold nor read if printed.”16 Macbean wrote that Chambers had written to him from Islington, where he was convalescing, that “I want all the apparatus I used in correcting the new edition of my book, to be brought to Cambury-house . . . particularly a number of books, I believe ten or twelve, and an index wrapped in thick brown paper; the first volume of the Dictionary too, I was at work upon, should be sent.”17 At his death Chambers left behind enough material for seven additional volumes.
By inviting readers to assist with the expansion of information included in the Cyclopædia, curators like Chambers partially dissolved the barrier between author and reader. But a barrier of sorts did necessarily remain, since in each instance the central project curator retained responsibility for soliciting submissions as well as curating the final textual product. The contributors were hardly a full-fledged author, in terms of either agency or responsibility: curators still retained the necessary primary editorial responsibilities, vetting the submissions before moving to more curatorial concerns such as dealing with the arrangement and composition of the final text.18
These practices functioned this way partly because of traditional vetting processes, but also because of the practical limitations of print production technologies. At bottom it is an interface and platform issue: there is no technological means for thousands of writers to simultaneously contribute to and build a printed text. It was easy enough for individual or small teams of readers to peruse the text, determine what was missing in their area of expertise, write the piece(s) they’d like to contribute, and then submit via mail or courier. Recall also that Chambers did not require literacy of his contributors; it was similarly easy for them to complete an interview and for their knowledge to be transcribed. But there is no practical way for an analog crowd to do the close work of composing a usable architecture or handling production and distribution. Some central person or group must perform the labor of filtering the submissions, integrating chosen contributions with the central architecture, managing the navigability of the expanded structure, and then submitting that completed work to the publisher, who in turn makes arrangements for the material labor of printing. Along the way the Page 21 →original submitter’s performance of agency may be significantly altered—that is, rejected or altered through the communal and participatory processes demonstrated in the craft of curation.
Chambers was hardly a pioneer in crowdsourcing through reader contributions. The sixteenth-century Danish astronomer Tycho Brahe was one of the first to solicit crowdsourced observational data in scientific publications, and the laboratory tradition of scientific work has been frequently, if not heavily, collaborative.19 Chambers also referenced the crowdsourced contributions to many magazines of the day, such as the Spectator and Tatler, which he described as “carried on with surprising Spirit and Success.”20 Other reference texts in production during the same period, such as Bayle’s Dictionnaire Historique et Critique and Zedler’s Universal Lexicon, made a similar practice of incorporating reader comments in subsequent editions.21 The innovation of these eighteenth century reference authors was to make space for participation by the broader reading public, rather than hand-selecting authors from a credentialed group or closed community. This practice remained customary in later reference projects: Diderot’s Encyclopédie would incorporate a considerable community of contributors, as did the 1884 Oxford English Dictionary.22 The production of the Encyclopædia Britannica has long relied on a network of invited, credentialed authors, as do most encyclopedic products that aim for an authoritative reputation.
The strategic inclusion of distributed contributions in the Cyclopædia demonstrates an astonishing similarity to the work that has more recently been invited and undertaken in Wikipedia. The open structure of the wiki platform and Wikipedia’s policies explicitly facilitate the sort of open contribution system that Chambers and his publishers worked toward in their requests for analog contributions from the public. Since read-write web platforms enable contributions directly from the public without the mediating agency of a central curator, performances of agency necessarily shift, as does the potential rate of production and type of contributions. A theoretically unlimited number of individuals can directly contribute to the project, subject to access and server capacities. In the print production process, the number of contributors is constrained by the amount of centralized human labor available to read, edit, incorporate submissions, proofread, set type, and create print runs. If ten thousand contributors send articles to a staff of three, either they will be processed long after they are received or some contributions will not be processed at all. If other factors demand that the edition must go to press before the filtering can be fully completed, then some submissions will necessarily either be saved for the next edition or discarded. Since wikis remove the intermediary submission, filtering, and incorporation stages by facilitating direct edits by readers, the potential for successful outside contributions increases.
Page 22 →Part of the reason for Wikipedia’s successful project life, then, is the innovative ease of the MediaWiki interface, which functions as an important actor itself: with the click of a mouse and a few keystrokes, readers can easily become writers and then, after completing their contributions, return to reading. The readers are transformed into curators, adding their own contributions directly to the central text and changing prior text as they see fit. Project policies about formal vetting processes also muddy the traditional divide between writer and reader: because there is no single, unified vetting process, each event of reading is a potential vetting. Readers are invited to simultaneously perform as reviewers because the act of “many eyes” crossing the pages and locating errors constitutes one of the fundamental WikiPhilosophies of eventualism, the idea that enough readers over enough time will improve the quality of a page. Sometimes, though, more eyes on a page increase the potential for destructive contributions in the form of vandalism.
Another formerly necessary intermediary stage is removed as well: that of the printer. Digital text does not require the labor of typesetting, inking, printing, and binding. There is no direct correlation between printer and coder here, because of the WYSIWYG editing interface of the platform. As the reader-turned-curator clicks the “save” button, the edited text immediately appears and is accessible to the world, “published” by the same person who was, just moments ago, reading the text. In the world of Wikipedia, it is impossible to draw clear distinctions between the formerly distinct roles of reader, writer, editor, and publisher, since a single rhetor can perform all these roles and exercise the agency that was formerly distinct to each.
Many hands working simultaneously on the same project leads to chaos, but it also can lead to exponential productivity. This potential is the basis of what Yochai Benkler calls commons-based peer production. This production method offers two distinct advantages, as Benkler has argued in The Wealth of Networks. First, it distributes decision making, which renders robust peer review an inherent part of the project. Second, it allows more individuals to scour more resources than any single firm or market would accomplish through traditional methods. Each contributor can contribute a small, manageable module that, when added to the whole, moves the entire project significantly forward. Wikipedia, of course, is built primarily on these sorts of incremental public contributions.23
Textual Curation and Authorship
Encyclopedic composition challenges commonly held cultural notions of what constitutes both authorship and writing. Far from the conventions of original creativity we expect in poetry and prose, which demand that writers compose as Page 23 →inventively as possible within the constraints of their chosen genre, encyclopedias require the creation of text that is factual, pragmatic, and arranged according to a culturally acceptable order. While the creative writer and essayist’s claims to originality are culturally accepted and expected (even as we acknowledge the pervasiveness of influence and remix), the reference writer is instead explicitly commissioned to produce derivative work. Encyclopedic articles are expected to rigorously report factual information gleaned from expert sources, not opinion, interpretation, or original research. In both print and digital encyclopedias, working as a textual curator involves engaging fundamental questions of genre, architecture, and meaning making.
While building encyclopedias curators collaborate in multiple temporalities, particularly in large, commons-based collaborative projects such as Wikipedia. They may collaborate after the fact with authors of the texts they are filtering and recomposing; asynchronously or in real time with contemporaneous community members who manipulate their text by not just adding words but manipulating it through code; and then with still other contributors who may alter the text years in the future. Because of the sheer mass of information that must be managed, recomposed, and arranged in order to successfully develop a large digital text, curators are frequently working in collaboration with prior contributors over extended periods of time, as seen in Wikipedia’s development. Collaboration occurs recursively throughout production stages as well as after the fact when the readers begin to write their own experience of nonnarrative texts by finding their own paths through it via cross-indexing, links, or individual choices.24 The textual curator’s contributions, then, come through honed research skills, astute arrangement, clear recomposition, and curatorial information management.
This sort of distributed collaboration can be understood as dialogic, following Ede and Lunsford’s landmark 1990 study of collaborative writing. They understand dialogic collaboration as a loosely structured, fluid collaborative practice in which “one person may occupy multiple and shifting roles as a project progresses.”25 The intensely dialogic nature of curatorial collaboration proceeds in exactly this way, especially in large collaborative projects where collaborators may wear multiple hats in a single work session, let alone over the life of the project. A curator may add primary text to a page, then edit another person’s contributions to another text, audit and clarify metadata, then switch to conducting research in order to gather broad information for filtration and recomposition. In another session the contributor may make sure that navigational links are up to date; all the while other contributors intersect with the first curator’s work, both asynchronously and in real time. The work may occur with little direct communication, discussion may happen behind the scenes, or arguments may break out and then move to arbitration. The Internet’s inherent affordances of speed and reach, Page 24 →which Gurak has written about at length, further enhance both the collaborative potential in digital environments and the potential for fluid shifts in curatorial roles.26 These affordances have fostered the increased deployment of distributed collaboration in both organizations and individual business practices.27 Similarly they have given rise to the sort of distributed collaboration found in open digital projects like Wikipedia that depend on many contributors from all over the world working through a nonhierarchical collaborative process to develop projects that are frequently based on creative works that are in the public domain.
How, then, should we understand the question of authorship in this sort of textual situation? Throughout this study I use the term authorship not to refer to compositional processes, but rather who gets to claim the status of “author” and the ways that we theorize, reward, and punish authors and their creative products. The historically and materially situated ways that we understand authorship have shifted considerably over the past fifty years or so since Barthes and Foucault began to destabilize them with their famous debate over the nature of the Author. Reproductive technologies that simultaneously afford circulation continue to be a central factor in this shift, which is not yet complete. Our own cultural moment is approaching a critical juncture when it comes to these issues. Considering the technological actors that are essential for facilitating this kind of collaboration and production poses further complications. In the quarter century since AOL entered American homes, we have become intensely familiar with networked, social forms of writing. We now carry them everywhere in our pockets, using smartphones to broadcast short-form texts across social media platforms and longer texts through e-mail. We share memes, gaining small bits of social capital by circulating the creative work of others. We launch Google Drive and peck away at shared documents. We quickly look something up on Wikipedia and perhaps make small edits to a page. When hundreds or thousands of collaborators work together asynchronously in networked environments, making contributions that can consist of recomposed texts or one metadata term or a line of template code, our conceptualizations of authorial agency, credit, and property are deeply challenged, along with our notions of what constitutes writing.
This networked information economy has imposed a seismic shift on traditional media production ecosystems and understandings of how authors are positioned within them.28 Thousands of professional photographers have increasingly found themselves out of work over the past decade as prominent publications have come to rely on Creative Commons–licensed photos on Flickr for image content. Citizen journalism contributes to and often drives media coverage, changing the economics of the news and magazine journalism. Twitter streams provide crowdsourced, real-time commentary on emerging news stories that is often repurposed by mainstream media outlets or simply followed by users who now ignore Page 25 →traditional media entirely. And in 2012, after a long public war with Wikipedia, the Encyclopædia Britannica ceased print production and began to extend its ongoing project through an entirely digital edition with limited social contributions.
During roughly the same period, the inclusion of highly usable media production software on laptops alongside the introduction of faster broadband connections that facilitated easier access to digital media artifacts drove the rise of remix culture. With Apple’s 2001 “Rip, Mix, Burn” campaign launched in 2001, the term “remix,” which loosely refers to recomposing found multimodal artifacts into new texts, entered the common parlance. Writing studies scholars investigated the various implications it posed for theorizing composing practices and pedagogy, incorporating playful repurposing and ethical reuse into writing assignments that complicated originality and authorial credit.29 Popular discussions about authorship and intellectual property have also evolved commensurately as public intellectuals have taken the debate outside of the courtroom and academy through public talks and popular publications.30 The popularization of remix culture through sampling, fan fiction, and video production subcultures such as machinima brought visceral demonstrations of the pervasiveness of influence.31 As a result contemporary writing is commonly (if not always) considered to be networked and social, always marshaling a plethora of influences in its invention and production. Alongside these developments postmodern critiques of authorship32 and originality33 have been naturalized in the humanities over the past half century.
Despite this received construction of a destabilized, networked Author, our systems of credit and reward remain resolutely dependent on the idea of an identifiable, discrete genius (either human or corporation) who creates a product that can receive individual credit in the form of copyright, grades, or tenure. We expect this product to meet standards of originality that render it defensible against charges of plagiarism or copyright violation.34 If prior texts are incorporated, we demand that they not just be folded in, but awarded either rigorous citation or credits that are often accompanied by royalty payments.
Consequently the capital-A Author still looms for many of us who study theories of writing and circulation, regardless of our disciplinary affiliations. This Author is a mutable colossus, alternately dead, reborn, deconstructed, distributed, reduced to a signature, anonymous, pseudonymous, feminized, contemplated. And yet, in spite of this plethora of theoretical manifestations, it remains strangely unified within popular cultural depictions. When we consider the Author, the creature we are still most often referring to is the Poetic Author, creator of various forms of literature. Its value lies in the original genius of the novels, plays, poetry, or essays produced. The notion of originality is central: it is considered vital that the work be fresh, inventive, and unusual within its cultural context, lest it lack cultural value or be vilified as plagiarism. Martha Woodmansee has deftly Page 26 →described this Romantic Author as a “unique individual uniquely responsible for a unique product.”35 Lockean sweat-of-the-brow doctrine indicates that with originality comes ownership, in both senses of the word: responsibility for the legal and cultural consequences of the text, as well as the right to compensation for labor. Western intellectual property theory and law are founded on the idea that the original genius demonstrated in these works carries with it these inherent property rights.36 The existence of this proper-noun Author, then, becomes an incontrovertible cultural fact—even an essential attribute of a Real Writer—and copyright law represents the institutional embodiment of that fact.37
Constructions of authorship are intimately tied to historical and material moments of both textual production and critical analysis. The emergent contemporary popular understanding of authorship as social is not new; it existed in a different form in manuscript cultures but has experienced a substantial period of flux that is still with us to some extent. As Margaret J. M. Ezell has argued in her important study of authorship in manuscript texts, there has been a tendency to impose our contemporary understanding of authorship as a lens for examining texts in earlier eras, assuming that what was true now was true then and has been true during any intervening period as well.38 Authorship and composing processes are increasingly figured as social and intimately connected with contemporary technologies, but Ezell established the sociality of authorship in rural areas of Interregnum and Restoration England. The handwritten, mostly single-copy poetry and prose that she examined was, by virtue of its material form, produced for very localized audiences and generally not intended for publication or wider circulation. These manuscripts were often contributed to or transcribed by close family members, with a focus on the creative process and communal readings rather than on publication. “The network of writer and reader that in my view characterizes manuscript literary culture and social authorship is created by the process of being an author rather than by the production of a single text, in Eisenstein’s terms, one capable of being fixed, attributed, and catalogued,” Ezell wrote. Integrating machinery into the equation to set this sort of handwritten manuscript in print and reproduce it for circulation would transform the figure of the author and the author’s social standing in ways that were often considered undesirable by the subjects of her study. For this reason many composers of the collaborative manuscripts in Ezell’s study actively avoided publication.
The rise of a substantial print economy in London across the seventeenth century drove a pragmatic reconceptualization of authorship that was largely predicated on economic interests. Printing technologies represented significant capital expenditures by printing houses, and the market for printed texts offered substantial rewards. The Stationer’s Guild consolidated tremendous power, functioning with the cooperation of the Crown as the central regulatory body that determined Page 27 →who might print and what materials were printable (that is, not blasphemous or seditious). Officers also carefully regulated printing technologies, destroying unauthorized presses or materials and levying harsh punishments. Despite this, a robust underground printing community continued to produce radical pamphlets and broadsides for wide circulation. The mainstream printing house and bookseller communities of Fleet Street and Paternoster Row also flourished, partly because they held full rights to any texts they produced.39 By the final decades of the century, prominent writers in London were deeply concerned with the ways that ownership of their work belonged exclusively to the publishers, and a hotly contested conversation took place in the larger creative community over a number of years.
As Mark Rose and others have established, the construction of the modern Author is contemporaneous with the rise of print. Creative works are transformed into material literary property by print and its potential for broad circulation and economic reward. The idea that an author (rather than a muse, a patron, or a publisher) might own his or her work and thus benefit as the owner of an intellectual property is a distinctly modern concept, and it is one that is intimately intertwined with the affordances of print technologies. The network of nonhuman actors that facilitate the material production of print is as essential as all the humans who take part in the technological system of print production and circulation. And yet the end result—the bound book on the desk—renders invisible the collective of humans and nonhumans that produced it and instead reveals the Author, who in order to logically be awarded ownership must be understood as the sole originator of the text. Print texts consequently drove a reconceptualization of the social author as originary and solitary in order to be proprietary, and this construction did not begin to significantly shift in the popular consciousness until the affordances of desktop computers and cheap, fast broadband created the conditions for a rebirth of the social author.
Curatorial Authorship
The encyclopedic form demands creative compositional responses to communication problems. That is, the text itself becomes an actor in the collective of agents who create encyclopedias. “Agency is textual or, put differently, texts have agency,” Karlyn Kohrs Campbell has reminded us.40 This agency is demonstrated in part through familiar features that guide the audience through processing the text. Form has “a power to separate a text from its nominal author and from its original moment of performance,” she noted, and successful form leads to imitation. Continued imitation in turn leads to prevalent audience and genre expectations. Contemporary audiences now expect encyclopedias to be divided into discrete, topical articles. Print encyclopedias are expected to present these articles Page 28 →in alphabetic order, while digital encyclopedias must include a robust search function, sufficient metadata to be easily searchable, and a hyperlinked information structure. Illustrations are welcome and expected to be clear and targeted. If an encyclopedic project fails to meet these expectations of form, its authority will be called into question regardless of the quality of the textual curation and composition demonstrated by its writer(s).
Despite the derivative nature of this writing and the many moving structural parts that it includes, the textual curator is an author every bit as deserving of the respect accorded to Authors as an essayist or poet is. The curator’s value is demonstrated through the work of improving the quality of information available, the quality of writing in the articles, or the accessible arrangement of information. And indeed the authoritative reputation and success of any encyclopedia rests primarily on the successful execution of these skills, which we commonly understand as not just invention and revision of narrative text but also linking paths through the text with cross-references or hyperlinks and creating contextually defined metadata through keywords and headings.
I suggest that these essential skills constitute a category of compositional craft called textual curation. The labor and small texts necessary for this work are not less important or diminished writing but carry with them the value attached to writing-as-skilled-craft. This expanded skill set of filtration, recomposition, and composing for findability and navigation includes careful attention to some aspects of writing that we have traditionally emphasized, such as source assessment and citation, but also demands a reconsideration of the canon of arrangement’s vitality. This reconsideration includes a broader conceptualization of textual organization that moves outside of individual texts and into the ecologies they exist within,41 both in their original state and in new ecologies that are constantly collaboratively developed. This recomposition, which most often requires working with multiple informational texts that will be filtered and recomposed into a new text, pushes against traditional expectations of original authorship.
Given its frequently collaborative nature, it also requires becoming comfortable with forms of authorial agency that are explicitly distributed and contextual. Jodie Nicotra has argued that in the case of the crowdsourced taxonomic vocabularies known as folksonomies, “rhetorical agency and intention become much more complicated, because invention is revealed as not simply the product of an individual, isolated mind, but as a distributed process driven by the interaction of a multitude of users” (273). The same is true for other curatorial tasks that require us to reconsider the requirements of original, single-authored writing, such as writing metadata for another’s article, collaborative editing, distributed architecture development, and the like. Addressing this need means teaching the curatorial moves required when writers are managing digital information in Page 29 →both personal and business contexts, whether it be categories on an organizational blog, a basic business website, a citation library on Zotero, tagged images on Flickr for a nonprofit, a Storified archive of tweets collected via hashtags at a convention, or the information architecture of a digital humanities project.
Modern encyclopedias present a unique rhetorical and textual situation within which to analyze the fluctuations of authorship: they are utilitarian rather than creative, real-world rather than fictional, objective rather than subjective, and schematically arranged rather than narratively structured. The pragmatic aspects of dealing with such a broad project necessitate collaborative authorship: since the information explosion of the Enlightenment, it has been impossible for a central reference text to be originally written and mastered by a single author.42 Neither has it been necessary, since vast amounts of previously published dictionaries, lexicons, and handbooks—and eventually encyclopedias—were available for “inspiration.” The curatorial process is almost directly counter to the more usual conceptualization of an author as someone who creates an original text through his or her own individual genius.43 In fact original research has traditionally had no place in the modern Western encyclopedic tradition and is specifically banned in Wikipedia on the grounds that it runs counter to the project’s Neutral Point of View (NPOV) policy.44 As Collison wrote in his discussion of plagiarism allegations against D’Alembert’s Encyclopédie, “The truth is that new encyclopedias are always built on the structure of their forerunners, and thus reflect the process of the accretion of knowledge itself.”45
Curation and Invisibility
Curatorial authorship is not often acknowledged as authorship or even as writing because of its invisibility when done well: a beautifully curated reference text allows users to simply access the topical area they need without undue thought, and the authority bestowed on canonical encyclopedic brands also largely relieves them of needing to consider the sourcing, filtering, and recomposition processes. The textual-curator-as-subject is often similarly obscured. This author departs from our continuing notions of easily identifiable authorship in a variety of ways, chief of which is the ways its inherently collaborative aspect frequently obfuscates easy identification of individual writers. Because of the sheer mass of information that must be managed, recomposed, and arranged in order to develop an encyclopedic text, the encyclopedist is always working in collaboration with prior scholars and writers as well as assistants and coauthors. Since the writer is also very rarely the printer or coder who handles large-scale information architecture and textual production tasks, collaboration occurs at the production stage as well as after the fact, when the readers begin to write their own experience of the text. Retreat to Page 30 →the proverbial garret is simply not an option, and in both our early modern examples and Wikipedia, the individual authorial signature is purposefully banned. The encyclopedist’s contributions, then, come through research skills, astute arrangement, clear recomposition, and the sheer labor of project management.
A number of helpful terms for understanding similar authorship forms have emerged from writing studies’ highly collaborative subfield of professional and technical communication, which offers a natural arena for examining authorship that is primarily collaborative and does not rely on cultural notions of individual original genius in the same ways that poetic and academic genres do.46 Jessica Reyman has argued that “authorship as applied in technical communicator professional practices is a far cry from the concept of authorship purported in the academic environment. In a professional context, technical communication is rarely considered an individualized activity; it is not something to which we assign ownership by an individual writer; and it typically does not produce something valued most for its originality.”47 In contrast to from the conventions of academic work, technical communicators engage in compositional work that is quite different from that taught in traditional composition classrooms: single-sourcing; development and use of boilerplate; repurposing documents for varied internal and external audiences; and producing corporate documents that bear no authorial signature and are produced under work-for-hire agreements that place the corporation in the legal role of author. Consequently the properties of authorship in technical genres look quite different from the properties (autonomy, proprietorship, and originality) that have been central to modern understandings of authorship.
Curators of encyclopedic articles frequently work under similar conditions. During the first one hundred years of the contemporary Western encyclopedia, individual authorship of individual articles was not identified in the final product. It was not until the 1824 edition of the Encyclopædia Britannica that authorial signatures began to appear at the end of articles in the form of contributor initials.48 Today signatures vary by project; the standard byline on Wikipedia articles credits the work to “Wikipedia, The Free Encyclopedia.” Authors are rarely if at all referred to in social references to encyclopedia articles: we refer to “The Britannica” or “Wikipedia,” not to the writers who wrote it. Articles are expected to be composed of previously established and published facts, and to conform to a general template for introductions, standard illustrations, order, and formatting. Careful metadata management is required in the form of cross-indexing or live links.
In recent years scholars of technical communication have described this collaborative, primarily less-original composition in a variety of workplace studies. In his 2005 study on workplace collaboration, Scott Jones described common technical communication tasks such as “document borrowing,” which includes single-sourcing and adapting prior documents for new contexts.49 Through these Page 31 →and other collaborative tasks, all done as work-for-hire, the writers in his study transitioned from “being the sole creators of mainly independent units of texts to being among the creators of highly interconnected parts of a very large unit of text.”50 Other scholars have described this process of recomposing and structuring texts in a variety of ways: “layering,”51 “environment selecting and structuring,”52 and “textual coordination”;53 the resulting assemblages of texts are understood as “genre ecologies” by Freedman and Smart and Spinuzzi and Zachry.54 Both Jones and Slattery have suggested that these changes are at least partly due to changes in technology, which encourages broader collaboration by making the tasks of coordination more efficient. This sort of distributed teamwork has driven the composition of encyclopedic texts for centuries,55 but the technological affordances of wikis certainly facilitate this sort of work in far a more immediate fashion.
While curation shares some aspects in common with each of these terms, it differs fundamentally in its context especially in the case of Wikipedia: the wilds of the Internet rather than controlled corporate environments. Each of the above studies examined the work of recomposition within workplace contexts. While textual curation can certainly be performed within institutional contexts, it is also common in technical communication and knowledge work that are commonly practiced outside of corporations. It is also most frequently practiced by composers who would not consider themselves to be technical communicators and who are not formally employed in this profession. I suggest that this technical, distributed work constitutes technical communication in the wild, not only in the perhaps more obvious case of Wikipedia, but also in games,56 piratical file-sharing communities,57 and forums devoted to technical topics such as high-end audio or woodworking. Studying these and similar sites that exist outside of traditional workplaces has the potential to reveal the extent to which such work is shaped by the contexts of production, availability of technologies, and the constraints of genre.
Rhetorical Agency and Collectives
The text, its curators, its production technologies, and other agents who form part of the collective that produces it all function as points of articulation that drive the curatorial process forward. The vitality of these connections is particularly apparent in examinations of authorship and ownership in the encyclopedic tradition, and they, along with the specific exigencies and demands of what we might loosely term the encyclopedic situation, impose a paradoxical agency. The curator’s authority is partially derived from the particular forms of agency that are shaped by the production process of a comprehensive reference text.
From one perspective this author has very little agency, since he or she works within a tight form with rigid genre expectations, recomposes information Page 32 →gleaned from other resources, and works toward the strict goal of organizing information in ways that provide optimal user experience. Encyclopedic projects are supported and constrained by a number of social as well as material forces: religious and political dictums regarding knowledge, the publisher’s threshold for courting controversy, and the physical problems of printing or navigating such a large textual object. They are also driven by economic alliances: The Cyclopædia was pushed forward by the mores of the Enlightenment project, one material expression of which was the newly formed United Grand Lodge of England’s devotion to advancing and distributing scholarship on Newtonian science. Wikipedia is supported in part by supporters of commons-based and open-access philosophies that have thrived in postmillennial digital cultures. But at the same time this author exercises extraordinary agency in the process of curating a text because the encyclopedist wrestles with and decides what constitutes knowledge and how readers might access it. In the process curators frequently confront the nature of knowledge and the ethics of access.
These performances of agency are also distributed across many agents—more than three hundred thousand in the case of Wikipedia58—working together to expand, improve, produce, or even vandalize a central encyclopedic project. Their performances can involve direct dialogue with each other but just as frequently involve interaction only with the text and interface as they make contributions. Just as these contributions often push against our notions of what constitutes real writing, they also push against typical conceptualizations of agency that are based on full arguments presented to an audience, dialectic between two agents or among a small group, or conversations that drive civic participation. Curatorial performance may consist of robust arguments about whether a topic is notable enough for inclusion, but it may also involve the insertion of a single metadata tag, inserting a link and thus transforming text into to live code, or reverting vandalism. It may be demonstrated by an automated agent inserting census data or an automatically generated map, or even reshaping human performance by patrolling for common curse words. These granular contributions constitute direct action that impacts the form and content of the text, and they require recognition or response from the community, who may let the changes stand, build on them, or remove them.
The curator and his or her collaborators, both human and nonhuman, demonstrate the highly contextual nature of rhetorical agency. In her exceptional essay on aspects of rhetorical agency, Campbell described the ambiguous, polysemic properties of agency, “a term that can refer to invention, strategies, authorship, institutional power, identity, subjectivity, practices, and subject positions, among others” (1). For Campbell agency is fundamentally the capacity to act, and to be “recognized or heeded” by one’s audience. No rhetor succeeds in persuasion Page 33 →by acting entirely alone. Given the fact that persuasion always necessarily involves multiple actors, it is also then essentially social and contextual, she wrote: “Agency is communal, social, cooperative, and participatory and, simultaneously, constituted and constrained by the material and symbolic elements of context and culture.”59 The performance of agency occurs within a complex network of cultural factors, which include institutional and community pressures as well as expectations regarding form.
Both agency and authorship occur not as essential properties of a subject or object but as interactive processes that involve exchange between multiple agents, texts, and influences. Agency itself lies in rhetorical performance, in what Carolyn Miller has described as an exchange of kinetic energy between agents.60 Conceptualized as performance rather than as inherent capacity, agency is not a property that a subject can possess61 or a force that arises from original genius (in the case of the Author), but rather it arises through response to a situation composed of parameters beyond the control of any single actor. The rhetor brings forth and responds, prompted by any number of contextual impetuses: discourse, community constraints, genre constraints, usability requirements, or code. Agency lies in the potential between that response and its reception.
This sense of energy and flux is particularly suited to the encyclopedic form, which is driven by an ever-broadening cultural quest for new knowledge and, consequently, new textual iterations that are shaped by the affordances and constraints of available technologies. This unending process of discovery and development is evident in the Greek origins of the word “encyclopedia,” enkyklios paideia, which roughly translates as “circle of knowledge” and conveys the idea that knowledge is without end. The invention process is also without end. Each of the actors in this study functions as what Campbell has called a point of articulation, negotiating material, linguistic, and cultural constraints in order to function as “inventors, in the rhetorical sense, articulators who link past and present, and find the means to express those strata that connect the psyche, society, and world, the forms of feeling that encapsulate moments in time. In this sense, agency is invention, including the invention, however, temporary, of personae, subject-positions, and collectivities” (5). As we will see, our actors invent not just encyclopedic texts through the process of curation, but also new collectives that have shaped human knowledge for the past three hundred years. In the eighteenth century they helped define natural philosophy (which we have come to know as science). In the twenty-first they have shifted our notions of what the encyclopedic genre can be and what it means for it to function as an open compendium of knowledge.
Conceptualizing agency as performance that is dependent on contextual, social response necessitates abandonment of the idea of an autonomous agent, Page 34 →instead shifting the emphasis to engagement and action. Performance of agency is not necessarily coupled with an agent: as Lundberg and Gunn argued, “Every action, discursive or otherwise, is only born of an engagement with the set of conditions that produced it” (96). Diane Davis further developed this view of response-ability, theorizing that agency is not “heroic”—that is, located in a self-determined individual—but is instead “an ethical relation that precedes identity, intellection, and intentionality. . . . Rhetorical agency, before it can involve symbolic action, requires an extra-symbolic signification, a saying, and the responsibility to respond.”62 For Davis as well as Lundberg and Gunn, response-ability through engagement with an ecology is at the foundation of rhetorical agency. This broader conceptualization of agency makes space for an expanded view of rhetoric, allowing for serious consideration of extra-symbolic, material aspects that are vital in shaping the conditions of rhetorical performance. We engage with other humans, certainly, but focusing exclusively on humans and symbolic communication does not provide a full picture of the real “set of conditions” that produces rhetorical action. We also inevitably engage the contextual, material constraints and affordances that shape rhetorical action: space, sound, images, and, most important to this study, technologies.
Thomas Rickert argued in his award winning book Ambient Rhetoric that acknowledgment of this engagement with the material world enriches our understanding of rhetoric’s possibilities and of its reality, allowing for a far more holistic examination of the properties of suasive force. For Rickert a more comprehensive model of rhetoric accounts for pervasive, ambient factors, “reject[ing] the notion of situations as mere composites of what is subjectively requisite, or the intersection of networking strands, with everything else relegated to epiphenomena, and instead see[ing] all elements as operational and even necessary, albeit at various levels of scale” (7). Dissolving the subject/object dichotomy saves us from the arrogance of presuming a human-centered universe and, at the same time, renders us more human in the best sense, more capable of acknowledging and accounting for the other agents we dwell alongside.
Developing a model of a rhetoric that includes material forces and nonhuman agents has profound implications for our understanding of rhetorical agency. “The change in perspective is crucial,” Rickert wrote. “Not subjective agency in a (necessary) context but a dynamic interchange of powers and actions in complex feedback loops; a multiplication of agencies that in turn transform, to varying degrees, the agents; a distribution of varied powers and agencies. Such an assertion dethrones the idea of mind as the engine of reason and seat of the soul” (10). Once the presumption of a fundamental subject/object divide is discarded, sentience no longer constitutes a necessary or sufficient requirement for the performance of agency. The human, with its underlying privileging of sentience, is no longer Page 35 →the baseline of what we understand as performance. Instead nonhuman actors come to the table and occupy a third position, one in which, as Bruno Latour has theorized, they “escape the strictures of objectivity twice: they are neither objects known by a subject nor objects manipulated by a master (nor, of course, are they masters themselves).”63 Their rhetorical agency thus takes forms that we have not yet quite defined, let alone fully acknowledged in the field of rhetorical studies.
We can locate and explore them within ecologies of actors and events that provide a view of agency as networked, articulated performance. Decentering the human as the absolute locus of rhetorical action, agency, and ethics creates a fundamental expansion of our models of how rhetoric functions, transforming it from an emphasis on actors to an ecological model that is fundamentally dependent on interaction with nonhumans as well as human interlocutors. Consequently it also leads us to a more nuanced understanding of agency that can account for multiple agencies apart from the traditional model of an autonomous human. Clearly neither a simple pencil nor a complex interface exerts the same form of agency as a human writer. But neither can a human write without a writing technology, even a technology as basic as a stick for writing in sand. “Ambience itself is a form of agency,” wrote Rickert. “While it may not be the agency we customarily attribute to human beings—and while we must grant such agencies different weights and values (which is of itself rhetorical work)—nevertheless, it is of a magnitude and scope to challenge more traditional notions of human agency” (16). It also challenges us to acknowledge the ways that we work alongside and with nonhumans at many junctures in rhetorical action, and the ways that affordances and constraints of tools shape our own action.
A full, responsible vision of rhetoric is obligated to account for the potential and implications of multiple agencies. This book does not pretend to develop such a comprehensive vision but rather works to contribute by exploring some of the ways that distributed agency is performed in collectives of humans and nonhumans who are engaged in the work of textual curation. In the process of composing carefully architected information structures like encyclopedias, we are intimately engaged with nonhuman actors such as texts and tools, both as collaborators in rhetorical action and as discrete actors themselves. Understanding agency as performance means accounting for the full range of human and nonhuman performances, correctly placing emphasis on action and exchange between actors rather than presuming a human-centered emphasis on autonomous cognition. Without humans collaboratively curating information and writing the small texts that support information architectures, without lead type and presses and binding, without the Wikimedia interface and Internet infrastructure and bots that write, these encyclopedias would not exist or circulate. These texts are the mixed issue of a thousand actors and just as many agencies.