Collaborative Literary Creation and Control:A Socio-Historic, Technological and Legal Analysis

UNDER DEVELOPMENT!!!!!!!

Webster's Revised Unabridged Dictionary (1913)
By the mid-1980's Webster's 1913 edition of their Revised Unabridged Dictionary was nearly forgotten. Its publishers, now Merriam-Webster, Inc., publish small revisions of the dictionary each year (changes are made to only about 50 words) and larger revisions every decade. 1913 had served its time and been pushed aside. The people I spoke to at Merriam-Webster knew nothing about 1913 and assured me that little, if anything, from 1913 remained in their current Collegiate dictionary. They described how the relationship between 1913 and the modern Merriam-Webster's Collegiate dictionaries, while still somewhat unclear, was tenuous at best.

In the first seventy-five years of its life, Webster's 1913 edition was revised several times and then abandoned according to company policy. Copies that escaped the pulping machines survived to gather dust in libraries and on bookshelves. Merriam-Webster retained full control over the text and, if they followed their normal patten of revision, did absolutely nothing with it after a decade. Barred from action by Merriam-Webster's exclusive rights to the text under copyright, neither did anybody else.

As Webster's 1913's quietly celebrated its seventy-fifth birthday in 1988, few noticed that it had finally caught up with the numerous extensions, revisions and rewrites of the United States Copyright Act that the text had seen in its lifetime. As Merriam-Webster's copyright expired, the 1913 version of their Unabridged Dictionary passed into the public domain and became common property. To dictionary manufacturers with their own up-to-date dictionaries, the availability of Webster's 1913's thousands of now antiquated definitions to the public domain meant little.

By 1996, the Internet had become a fixture of the lives of millions around the world. In this year, Project Gutenberg, a project digitizing and distributing public domain texts, released an electronic version of Webster's 1913. As the only large digital dictionary in the public domain, Webster's 1913 was quickly adopted by those unwilling or unable to purchase commercial dictionaries. The dictionary, ignored over generations, was the center of attention once again.

Since then, the dictionary has taken up numerous electronic formats to facilitate different methods of searching and browsing. It has been integrated into text editors, web sites, and word processors. Consequently, 1913 is cited increasingly often—last year in a petition to the United States Supreme Court ([Morrill2002]). The digital versions have been used as a seed for, or an addition to, various online knowledge bases including WordNet, Wikitionary, Wikipedia, Everything2 and others.

More strikingly, Webster's 1913 has grown. With knowledge of the dictionary's dated quality, several collaborative projects have sprung up to augment, fix, or build on the text of dictionary. Foremost among these is the GNU Collaborative International Dictionary of English (GCIDE) which, through an entirely volunteer-based collaborative effort, has revised, updated and added to the dictionary to create one of the most quickly growing dictionary projects in existence.

Webster's 1913's new life as GCIDE is a story about the power and effectiveness of collaboration. With its amazing resurrection, it is also a story about the effect of control on the life of a text. The new life breathed into Webster's 1913 can be tied to the elimination of the centralized way that the dictionary had been controlled. It can be tied to the elimination of Merriam-Webster's legal ability to control the use of the text. It can be tied to a series of technological shifts including the explosion of the Internet and collaborative authoring tools that made distribution and collaborative manipulation of the dictionary possible in new and different ways. In shifting away from highly centralized and individualized systems of control, meaningful collaborative work on the dictionary became possible. Through its history of abandonment and revitalization, 1913 acts as a powerful example of the way that highly individualized control limits the growth of a text—and of the great things that can happen when control is relinquished.

What is Collaborative Writing?
Almost every book and article on collaborative writing begins by asking, "what is collaboration?" In most cases, the authors proceed to tear apart the reader's preconceived notions and to leave the question more confused than when they began. While often impractical and unproductive, this approach is understandable and usually justified; collaborative writing is a slippery concept. It is clear that collaborative writing refers to writing in groups but there are as many ways to write in groups as there are possible combinations of individuals. Where does "a little help" and editorial assistance end and collaboration begin? There are no definitive answers.

Additionally, left to operate in an individual work/collaborative work dichotomy, defining collaboration involves defining what is not collaboration. Can individual writing involve borrowing, citing, appropriation and synthesis? How much? Where does one draw the line. There are no definitive answers. Ongoing academic discussions on the theory, definitions, and virtues of authorship and collaboration begun decades ago show no sign of resolution and continue to grow in size and scope. They demonstrate that there are no definitive answers.

While from one academic perspective, these questions are pleasantly unresolvable, an analysis of collaboration without a definition to frame it remains problematic. For a limited but piratical working definition of collaboration, one can turn to technologists who define collaboration in more mechanical terms. In an article on the technology and processes of collaborative writing, David Farkas offers four possible definitions useful in approaching collaboration through an analysis of processes. For his purposes, collaboration is:


 * 1) two or more people jointly composing the complete text of a document;
 * 2) two or more people contributing components to a document;
 * 3) one or more person modifying, by editing and/or reviewing, the document of one or more persons; and
 * 4) one person working interactively with one or more person and drafting a document based on the ideas of the person or persons. ([Farkas1991] p. 14)

By breaking the common-sensical concept of group-based writing into a four distinct types of work, Farkas' definition paints a picture of what is, and is not collaboration; it provides a useful place to begin.

However, in introducing the concept of "collaborative literature," one must also define "literature." Partially in an attempt to avoid this definition—defining literature can be as perilous as defining art—many who study collaborative textual production simply choose the term "collaborative writing." However, collaborative writing tends only to imply synchronous and fully consensual group work. Literature, on the other hand, is more than just the act of putting pen to paper. It is defined by the Oxford English Dictionary as, a "body of writing" or a "culture" of letters. In the following analysis, I use the term "literature" in a inclusive sense. For my purposes, it is almost synonymous with writing but implies connections between, and unity among, different written works over time and between authors in a way that "writing" does not. These connections may range from traditions and conventions to subtle allusions to quoting and, in their most extreme form, to plagiarism. While not always defensible, these connective acts are always literary. Literature is always collaborative.

Referring to this networked approach literature, Peter Jaszi extends Farkas' definitions in describing a fifth type of collaboration he calls "serial collaboration," a process he defines as borrowing, synthesis and appropriation. Serial collaboration flows from the manipulation of existing knowledge and can be widely asynchronous. For example, through revision and a close relationship to his texts, I might be able to "collaborate" with Charles Dickens in a serial manner by fixing what I felt was an error, elaborating on a set of descriptions, changing an ending, or rewriting an entire story. In the following chapters, I try to afford each of these models of collaboration a place.

In discussing collaborative writing in today's literary world where the dominant paradigm is a single author theory, many models describe collaborations as groups of individual authors working in an micro-economy model. Other models present collaborations as a group of writers occupying the role and space of a single corporate or collective individuality. Yet other models present collaborations as complex organizational entities and aggregations of individuals. By providing a more nuanced and complex model of collaboration and reducing the impact of systemic control, these models occupy an increasingly privileged and "meaningful" place in the following analysis.

Why is Collaboration Important?
Underestimated and ignored for over a century, as I will describe in more detail in Chapter 2, a great deal of attention began being focused on collaborative writing in the early 1970's when English and composition professor Kenneth Bruffee began arguing that by having students write essays and fiction in groups, students produced better work than when they worked alone. He argued that they learned more through group work than when they interacted only with their teacher ([Bruffee1973a]). Bruffee's has continued writing on the subject for several decades, has become involved vibrant academic dialogs, and has defended the classroom use of collaborative writing against criticism from the previous generation of writing professors.[1] Bruffee argued that collaborative writing and extensive peer work was reflective both of the business world and the academic fields in which students studied ([Bruffee1984] 643). Responses to his model collaborative learning have been, for the most part, extremely positive. Collaborative writings' effectiveness in the classroom has been repeatedly confirmed in what has become a large collaborative writing and collaborative learning discourse ([Gebhardt1980] [Bruffee1981] [Gebhardt1981]).

Bruffee's ideas stand upon a strong foundation of theoretical research into group work and collaboration. In their important book on group psychology, Barry E. Collins and Harold Guetzkow introduce a concept they call the "assemblage" effect, which describes the way that a group's final product is usually superior to that of even the best member's individual efforts. Karen Burke LeFevre, writing in 1987 argued convincingly that each aspect of the writing process—including invention, writing, and editing—are inherently social acts that benefit from and thrive in a collaborative environment ([Lefevre1987]). Collins, Guetzkow, LeFevre and other social psychologists use scientific research to give credibility to the power of collaborative writing that continues to be downplayed in the dominant literary environment. These researchers have demonstrated that collaborative writing could, at least in ways that can be tested empirically, produce better work and teach people quantitatively more than in situations where the same individuals write alone.

Read alone, the experiences of theorists like Bruffee and the research of Collins, Guetzkow and LeFevre form a strong argument in support of collaborative writing as a more effective mode of literary production: individuals produce better quality work, as evaluated along most sets of empirically evaluable criteria, by working with others. These theories have been affirmed in a number of empirical studies of collaborative learning and composition. John Clifford produced a study of college freshman that, using rigorous control groups, demonstrated that students who wrote collaboratively learned more from each other and, at the end of the study, had produced better work than students who had worked individually ([Clifford1981]). Another study by Collette Daiute confirmed the same phenomenon in fourth and fifth graders demonstrating that "students who collaborate made several types of significant improvements over students who wrote individually" and noted that work by groups of students was better than the best work of any single group member ([Daiute1986] 389).

In Copyrights and Copywrongs, Siva Vaidhyanathan relates the histories of literary, film, and musical copyright to emphasize how copyright is often ill-suited for the type of creativity at the root of American literary and screenplay writing and composing. He highlights the way that individualized authorship runs counter to the tradition of open sharing, borrowing and cross-pollination in American blues and the transgressive borrowing and sampling of modern rap music. He describes the way that even Mark Twain, who devoted much of his life to advocating stronger copyright law, borrowed and stole from African-American storytellers and the African-American storytelling tradition ([Vaidhyanathan2001]). Vaidhyanathan's historical analysis demonstrates that largely irrespective of authors' attitudes toward copyright, American literary history is a history of collaboration articulated as everything from editing to rampant and unabashed plagiarism. Vaidhyanathan's history of copyright shows a legal mechanism pushed in one direction by copyright holders trying to solidify control of their work in a way that legally undercuts the collaborative processes that made their work possible.

As a result, it is unsurprising that in the context of a long history and tradition of persistent and prevalent group work and its increasingly apparent effectiveness, collaborative writing remains prevalent. In a survey of six major professions, Lisa Ede and Andrea Lunsford found that eighty-seven percent of respondents wrote collaboratively in their work at least "sometimes" ([Ede1985]). In just the portion of the book dealing with non-academic settings, the editors of Collaborative Writing: An Annotated Bibliography list hundreds of articles establishing the prevalence of collaborative writing in corporate, industrial and academic reviewing, storyboarding, translation, usability testing and the production conference papers, documentation, policies and procedures, proposals, and technical reports as well as more traditional forms of literature like novels, plays and poems ([Speck1999]). This bibliography reflects an explosion of academic literature around collaborative writing over the past three decades; it covers nearly 1,000 sources written during the seventies, eighties, and nineties. In turn, this discourse reflects the growing popularization of explicitly collaborative writing. It reflects a shift in attention toward collaboration rather than a change in the prevalence of collaborative writing itself. Since academic communities have developed a discourse around collaborative writing and have shifted their gaze away from individualized writing processes, collaborative writing's effects, importance, and ubiquitous nature are being recognized at an unprecedented degree.

The Idea of Control / The Control of Ideas
While increasingly apparent, collaboration and a free interchange of ideas remain extremely difficult. Constitutional law professor Lawrence Lessig has spent the last decade writing about the way that ideas are controlled. In his book Code and Other Laws of Cyberspace, Lessig builds on the technical and legal definitions of code in an attempt to collapse the distinction between the design and implementation of computer programs that facilitate communication—and as a result define it—and the regulatory role that law has traditionally played. He argues that computer code needs to be subject to the same kind of scrutiny, accessibility, and malleability that we demand of our laws. In his second book, The Future of Ideas, Lessig's discussion centers more around the idea "control" defined similarly.

By adding to Lessig's conception of "code" the regulatory role of social and socio-historic forces, this project advances a revised concept of "control" in an attempt to tie Lessig's discussion of code into a larger analysis of the way that we manipulate and control ideas and their expression in text. It attempts to use this analysis to gain insight into the nature of the effect of control on collaborative writing in a broad and interdisciplinary fashion. The analysis alludes to the interconnectedness and underlying similarities between significantly different and apparently disparate articulations of systemic control.

I use control to refer to the systematic limitation of the collaborative manipulation, use, growth, development and distribution of text. Lessig presents the regulatory power of code and then warns his readers that in choosing code, we would best served by flexible and "open" systems. Like code, control is a tool but, unlike code, control can not be easily dismissed as just a tool. The term alludes to the fact that through our ideas, we too can be controlled; humans and societies are controlled, not coded. Control demonstrates how openness and flexibility are a step in the right direction but that they are only one step. As such, the limitations of systemic individualized control create a hostile environment for collaborative writing. For the purposes of my essay, I describe control as it is articulated in three interconnected ways: as conceptions of authorship, as technology, and as systems of law.

Control as Conceptions of Authorship
Social conceptions act as one piece in the production and articulation of control. The modern discourse around authorship began in 1969 when Michel Foucault asked, What is an Author? Foucault drew attention to a shift in the definition of an "author's" role that represented a "privileged moment of individualization in the history of ideas, knowledge, literature, philosophy, and the sciences" ([Foucault2002]). Scholars and theorists of various academic disciplines have spent the last thirty years responding to Foucault in what has grown into a vibrant intellectual discourse on authorship. While a clear picture is no closer than it was in 1969, much has been said about the fact that the importance, role, and definition of authorship has undergone major changes since the beginning of the eighteenth century.

One of the foremost participants in this discussion is Martha Woodmansee, who explains, in one of numerous essays she has written on the subject, that the notion that the author is the only participant in the production of a book worthy of attention and special rights—as opposed to just another craftsman—is rooted in the Romantic notion that significant writers, "break altogether with tradition to create something utterly new, unique—in a word, 'original'" (16). This highly individualized conception of authorship has, to one degree or another, dominated Western society's popular consciousness for the past three centuries and paved the way for copyright and a highly centralized publishing industry.

This popular belief in an author's primary, even exclusive, role in the creation of a text creates a social system that defines the way that texts are written, read, and understood. Through its highly individualized slant, Romantic authorship limits collaborative writing in important ways. Groups of authors do not sit down to write novels in part because most authors believe that this is simply not the way that novels are written. By eliminating the desire, social acceptability, or social space for deviant methods of textual production, conceptions of authorship play a meaningful role in controlling the production of text.

Control as the Technology of Writing
In a different but highly interconnected manner, texts are controlled by their technological and material context. A popular example illustrating this point is the Hebrew Torah and the Christan Old Testament. While both books share the same words, they are read and understood in very different ways. This is in no small part due to the technology through which the words on the page are accessed. To this day, the Torah exists in temples and synagogues in scroll form. As a result, the congregation's relationship to the text is one that progresses linearly, or perhaps cyclically, over time. Christianity coincided in its early growth with the rise of the more familiar codex form—a form that the religion helped popularize. As a result, Christianity is rooted in a less linear and more "random-access" method of interaction with its own Holy Book. The result is two religions with radically different interactions and interpretations of the same text. While the material itself is only one factor in the parallel development of the religions, the effect of material forms should not be underrated.

Nowhere is textual materiality more evident than in the invention of the printing press in the fifteenth century. Gutenberg's invention of movable type represented a radical departure from existing systems of literary production and distribution. This played out both in the use and interpretation of existing work and in the creation of new texts. It is clear that people produced radically different documents in print than they did in scriptoriums. The triple-decker novel is impossible without printing just as pulp fiction is impossible without the cheap paper, cheap printing, and cheap and extensive distribution.

As computer technology appears poised to redefine literary production again, the technology itself is no longer "hardware" like printing presses and movable type but computer source code. As such, our ability to manipulate the terms on which we can communicate and collaborate, as long as we have access to source code, is instantaneously and almost infinitely flexible. We can add a line here, subtract a line here, change a line here and we create a different system and a different environment to shape and control the creation, distribution, or manipulation of literature.

Control as Copyright
Mentioned previously, Lessig's concept of "code" compares computer software's malleability and its regulatory effects with law. By regulating in a similar fashion, technology and law control the production of text. Copyright, the primary legal mechanism regulating the production and distribution of written expression, can stand in for law in the following discussion discussions.[2]

The idea of authors, the technology of reading and writing, and the legal mechanism of copyright are heavily intertwined. The myth of Romantic creation provides the backbone and justification for current regimes of strong copyright. Intellectual property, articulated as parallel to other forms of property, must be owned by an individual.[3] The rise and evolution of copyright as we know it today can be read in relationship to the decline and devolution of collaborative models of authorship and technology facilitating a more interconnected, person-to-person mode of production and distribution.

Copyright's inception can be traced to the invention of the printing press, although its first articulation followed Gutenberg's invention by more than a century and a half. As technology is redefining literary distribution and eliminating the need for strong centralized distribution systems, society is faced with a discrepancy between the social systems of control that govern the way we feel about the creation and transmissions of text, the technological systems that defines the way that we create, give, get and borrow, and the legal system that define the ownership of these works.

Collaborative Literary Creation and Control
The following analysis is divided into three distinct pieces. The next section, Chapter 2, attempts to serve two important purposes. First, it offers what I call a meta-history of collaboration and control. The chapter begins with an introduction to the history of control as defined through social conceptions of authorship, changes in technology, and the closely related legal systems of intellectual property and copyright. Continuing in this context, the essay presents a broad history of collaborative writing. Through these parallel histories, the essay attempts to offer a glimpse into the troubled history and hostile relationship between the two. The chapter's "meta-history" lies in this intersection.

Second, the chapter attemps to provide a historical foundation and context for the two following chapters. Through this historical approach, it aims to provide historical context on the way control is articulated through popular conceptions of authorship, technology, and law. Through its survey of a diverse range of examples over a long period of time, the essay demonstrates the persistence and power of collaborative work before, during, and after the rise of powerful systems of control that include Romantic authorship, centralized publishing and copyright. This meta-history is one of conflict. The essay aims to help define control as a force that creates environments hostile to meaningful and flourishing collaborative writing.

Chapter 3 attempts to employ the concept of control in the development of a methodology for evaluating different computer-supported collaborative writing (CSCW) technologies. By focusing on the way that technical design choices are reflected in the creation of environments facilitating particular types of control, the essay presents a description of important points of consideration in the evaluation of collaborative writing software, a description of the way that control is articulated at each of these points, and suggestions for an environment promoting more "meaningful" collaboration. The paper demonstrates the usefulness of this methodology through application in a handful of case studies. In the context of the larger project, Chapter 3 uses an analysis of computer code to both gain insight into the concept of control and to apply it to the evaluation of computer software. The methodology for this evaluation outlined in the chapter has already been put into active use by one company researching, writing, and supportive collaborative and participatory computer technologies.

Relying heavily on the foundational work in the previous chapters, Chapter 4 levels a critique at contemporary copyright. The chapter introduces a snapshot of contemporary copyright and describes the way that copyright, as a system of highly individualized control, is poorly suited to the promotion of collaborative work.

Chapter 3 and Chapter 4 are designed to be largely self contained and serve similar purposes within the larger piece. Building upon the introduction of concepts in this chapter, and the contextualization of these concepts in Chapter 2, the final chapters aim to take the idea of control presented in the introductory pieces and apply it toward in fully articulated critiques of existing and important examples of systemic control.

Introduction
A well-known English professor once asked me if I honestly believed that a committee could write a great book. My answer to his question is a resounding "yes." Many, perhaps most, of the greatest works of literature, across time, across culture, and across language, are explicitly attributed to groups. As collaborative writing has gained scholarly attention in the last thirty years, many texts long-considered to be the product of single authorship have been revealed to be the product of collaborations.

In fact, my research into collaborative writing has demonstrated that collaboration is so persistent, so important, and so dynamic that this professor's question prompts several questions in response. How can prominent academics overlook collaborative literature in such a blatant manner? As the academic world founds itself on a tradition of synthesis, knowledge sharing, peer review, and editing, what set of political, social, and philosophical structures makes overlooking the importance of group work so easy? What is so important to and so intertwined with our fundamental understanding of creativity that it convinces even the most astute literary scholars to deprecate, discredit, and ignore one of the most historically effective methods of literary creation?

My answer to these questions, discussed in some detail in Chapter 1, is the intertwined group of sociological, technical, and legal concepts I call control. Control systems including social conceptions of authorship, technological methods of distribution, and codes of law, are fluid and dynamic entities. As control grows, as I will argue it has, it creates an environment hostile to collaborative writing. As this environment takes hold, collaborative writing is practiced less, ignored more, and driven underground until it effectively disappears.

In the case of this scholar's question, this process is made abundantly clear. In posing his question, he easily ignored the fact that, to one degree or another, almost every major novel, play, or large-scale poem written before the end of the Renaissance is the product of multiple hands. The little scholars know about literature from the millennia before the Renaissance tells us that early texts were the projects of communities, not individuals. We know that these ideas and texts were the property (if the term is even applicable) of God or mankind; they formed a sort of intellectual commons in which all new knowledge was based and into which all knowledge flowed ([Bollier2002]).[4]

These striking examples are possible to ignore because, during the eighteenth and nineteenth centuries, this community-based concept of authorship and the mechanisms for literary ownership, production, and control were overhauled. At that transitional point, massive shifts in the way literature was produced and distributed took hold. At that transitional point, legal changes saw literature become the property of individuals. It is at that moment, and others like it, that the following history is centered. Through major transitions in the nature of the mechanisms of control, collaborative writing has evolved and persisted.

This essay will present a historical overview of the way that control is defined. Its emphasis will be on the way that conceptions of authorship are articulated. "Authorship" is a complex and dynamic concept. As a result, this essay will consider copyright as partially reflective of attitudes toward authorship. Additionally, this relationship acts as an example of the interconnected nature of control systems as I have defined them.

With this contextual backdrop, the essay will examine a handful of particularly well documented historical examples of collaborative writing as representative of trends in the constantly evolving nature of collaborative writing. For most literary works—especially older texts—drafts, manuscripts and journals documenting the nature of collaboration are unavailable. With a lack of documentation to refute the claim, modern scholars assume singular authorship. Hopefully, my argument, supported by others who have influenced and inspired it, will help shed doubt on the wisdom of this assumption.

Brief History of Authorship
Before the rise and eventual dominance of the Romantic notion of authorship, new writing gained value from its creative affiliation with existing works, or what Martha Woodmansee describes as "its derivation rather than its deviation from prior texts" ([Woodmansee1994] 17). Before this important shift, the authorial role was often compared to that of a commentator, compiler, or transcriber. Contextualized in such a way, it is unsurprising that authors' actions in this period were intensely collaborative.

Prompted by shifts in the nature of book production and distribution ushered in by the printing revolution, authors began to take a more central role in the production of texts. Especially prompted by the rise of copyright in Britain in 1709, the eighteenth century introduced a new concept of individualized authorship based on the idea of a creative genius working alone. This idea—one at odds with collaborative, collective, or corporate creation—has remained widely influential despite powerful arguments made by theorists like Foucault and Woodmansee and a growing body of evidence that collaborative and collective creation is more effective than individual work. Peter Jaszi and a growing numbers of legal and literary theorists argue that it is copyright, a system designed to allow economic and political control of literary knowledge and expression, that has enshrined Romantic creativity in ways that have been difficult to challenge.

Brief History of Printing Technology
As a major factor behind authorship and copyright, social systems for literary production went from dynamic and ad-hoc collaborations among elite and highly interconnected literary circles to highly centralized systems similar to the contemporary publishing industry. Before Gutenberg's invention of movable type, books were written, by hand, by individuals or in scriptorium. Books, which were extremely valuable, were made of high quality materials like velum, and were passed between owners over generations. Often, each owner or reader of a book would make marginal annotations. One medieval form, the gloss, consisted largely of blank space to facilitate the addition of marginalia by readers. As books were copied by hand, changes and corrections were made; histories were extended to include more recent events. Books were designed, written, caligraphed, rubricated, illustrated, illuminated, bound, and decorated by large groups of individuals. Every book was a collaboration and no two books were alike.

The invention of printing revolutionized this system. While Gutenberg conceived as his invention as an "automated scriptorium," its use as something much more became quickly apparent. Not only were more copies produced, but different types of books as well. As one book was produced for many, the personalized nature of literature shrank. As publishing became industrialized, new technologies prompted new economic models which in turn had effects on the types of texts printed.

Brief History of Copyright
Technological changes played an imporant role in shifts in the social context of literary production and were reflected in legal shifts toward systems of centralized control through publishing laws that culminated in the creation of the Stationers company and ultimately in the articulation of copyright in the Statute of Anne. Radical technological and social shifts were intimately connected with the creation of radically different methods for the control of literature.

These systems of copyright replaced a tradition of "privilege" where monarchs would grant exclusive monopoly rights for the production of a particular text to a particular printer—usually without consultation of the text's author or authors. In England, this widespread practice eventually led to the creation of the Stationers' Company, a coalition of English printers that was granted—through an interesting combination of royal privilege and censorship policy—a monopoly on all printed material in England in return for an agreement to not print seditious or heretical material. While antecedent to copyright, the system was fundamentally different; while individual printers were granted exclusive rights, authors were neither mentioned nor consulted.

An important legal shift came in 1710 when this system was replaced with the "Act for the Encouragement of Learning and the Securing the Property of Copies of Books to the Rightful Owners Thereof," commonly referred to as the "1710 Copyright Act" or the "Statute of Queen Anne." The statute provided authors with exclusive rights to their works for fourteen years, extensible by another fourteen years if the author was still alive and cared to renew. In reality, and by design, authors' rights were immediately transfered to publishers—the legislation was lobbied for and supported by publishers alone. In the famous Donaldson v. Beckett (1774) court case, the law recognized authors' natural copyright as common law but decided that the 1710 Copyright Act supplanted this right with a statutory one.[5]

This legal shift paved the way for changing conceptions of authorship. As time went on, authors, perhaps most notably William Wordsworth (whom I will examine in more detail later in this essay) argued for this "natural" right to their artistic productions by connecting the model of Romantic creativity (dependent on the "introduction of a new element into the intellectual universe") to their desire to have more control over the use of their ideas ([Zall1966] 182, [Woodmansee1984] 427, [Jaszi1994] 35). Wordsworth went so far as to argue before Parliament for the institution of these rights. While the legislation in question was never passed, he successfully popularized his conception of authorship, the eventual dominance of which led to a major shift in popular conceptions.

In a causal reversal, this Romantic concept of authorship with its roots in the publishing industry's concept of copyright began to shape copyright itself. In this way, the Romantic conception of authorship enshrined the concept of individual creation in ways that, in the best situations, decreases the importance of collaborative work, and, in the worst, sits squarely at odds with its widespread application.

Early Models of Collaboration Before the Eighteenth Century
Before the shift that Foucault refers to as the "individualization" of authorship, explicit and deep collaboration was the dominant method of literary production. Martha Woodmansee describes the role of the author before the dominance of Romantic authorship as not dissimilar to other types of literary creators like scribes, compilers, or commentators. Woodmansee references a definition by the thirteenth century St. Bonaventure who describes an author as one who "wrote both with his own work and others' but with his own work in the principle place adding others' for purposes of confirmation" ([Woodmansee1994] 17). This thirteenth-century definition of authorship places literary creation squarely within the context of collaboration.

In another article, Woodmansee explores Europe's pre-copyright methods of compensation for artistic work. Prefacing the discussion, she notes that the concept of the professional writer is a relatively recent innovation. Before this period, writing was completed largely as a part-time occupation ([Woodmansee1984] 431). Most early professional writers were supported by honorarium, or a payment to an author to produce works that was given either by printers or by a king or noble. In this way, the honorarium acted as the backbone for systems of patronage. An honorarium was a mark of esteem and a method for a printer or sovereign who appreciated or benefited from the works to ensure the continued work of the author. It was not payment in exchange for exclusive transfer of work. In fact, an honorarium bore no fixed relation to exchange value or an acknowledgment of the writer's achievements. It was usually a fixed sum that did not fluctuate with the publication of new works ([Woodmansee1984] 434-5).

Through creation by non-professional writers and through the support of authors through honoraria, the constant production of new work was insured without the need for system of intellectual property or ownership. This arrangement was essential as the dominant models of literary creation were fundamentally intertwined with borrowing and collaboration in ways that a system of control, ownership and propriety complicates and hinders.

Imperial Chinese Literature
One example of this model is described in To Steal a Book is an Elegant Offense. The author, William P. Alford, attempts to explain why China had no laws resembling Western intellectual property or copyright until the twentieth century. Alford argues that the Chinese refused to adopt intellectual property policies because they were fundamentally incompatible with Chinese literature's basis in a creative process that elevated and necessitated borrowing, synthesis, and quotation—in a word: collaboration.

It is clear that the production of knowledge and literature in Imperial China was shaped by an intimate engagement with the work of others—especially one's predecessors. Drawing from the work of the noted Chinese literary scholar Stephen Owen, Alford makes generalizations about Imperial Chinese literature, describing how in order to "avail themselves of understanding in order to guide their own behavior, subsequent [Chinese] generations had to interact with the past in a sufficiently thorough manner so as to be able to transmit it" ([Alford1995] 25). Owen compares the importance of this connection to the past in Chinese literature to the attention to meaning or truth in the Western literary tradition—perhaps Western literature's most important goal ([Alford1995] 26).

In arguing this point, Alford quotes passages from influential Chinese thinkers spanning several centuries. A passage in the Analects of Confucius states: "The Master [Confucius] said: I transmit rather than create; I believe in and love the Ancients" (bk 7, ch.1). More than a millennium later, As Wu Li (1631-1718) claimed that, "to paint without taking the Sung and Yuan masters as one's basis is like playing chess on an empty chessboard, without pieces" ([Alford1995] 28). Separated by epochs, both thinkers decry the idea of solitary artistic creation. To each, the organization and creation of new knowledge, literary or otherwise, must reach outward rather than inward.

Alford's examples are intriguing because they are employed not in the context of a discussion of Chinese literature but a discussion of intellectual property and control. Alford argues that Imperial Chinese literature was rooted in a conception of authorship that identified the author as a craftsman and a historian. Authors assembled and connected existing pieces of literature in the creation of new works; no good author, even one secluded in the woods, works alone. Consequently, originality was defined not in the context of a lack of influence but from a context of a rich meaningful interaction with existing knowledge. In the absence of a meaningful collaborative literary process—with authors both living and dead—Chinese authors were doomed to inefficacy and unoriginality. This attitude toward literature is summed up with Isaac Newton's famous phrase, "If I see further, it is because I stand on the shoulders of giants."

Alford's conclusion is that without free and unhindered ability to access and change the works of others, especially one's predecessors, this collaborative model of literary creation is impossible. As a result, the popular Imperial Chinese conception of authorship was incompatible with Western systems of control based on copyright and Romantic authorship. The relatively recent institution of copyright in China gives us a unique opportunity to explore collaborative artistic creation in the recent past and gain insight into the European history of creative models before the widespread adoption of modern systems of control. In simple terms, the Chinese experience demonstrates a model not dissimilar to one that Europeans enjoyed before the widespread adoption of copyright.

The Talmud
To western readers, a more familiar example of a collaboratively created text from antiquity is the Jewish Talmud. In its simplest form, the Talmud is a compilation of ancient Jewish law and lore created by large groups of Palestinian and Babylonian rabbis between the late first and seventh centuries A.D. As such, the text's relevance in the context of a discussion of collaborative creation and control needs no further justification. Still, the Talmud is particularly interesting because, as an important religious text, it's history is extremely well researched. However, unlike most other Holy texts, the collaborative nature of its creation is not downplayed in this research but is highlighted as essential to its form and function.

Pages of the Talmud, called folios, are separated into blocks and pieces. Many folios include the Mishna, or bits of traditional law, transmitted and altered orally for centuries until they was transcribed (into numerous differing copies) at some point before the middle of the sixth century ([Strack1972] 20). Flanking the Mishna on each folio are other texts, the majority of which constitute commentary and criticism. While much of the commentary is on the Mishna, a large portion of the Talmud is commentary on the commentaries.

Detailing the nature of the collaborative process that produced the Talmud is a tedious and confusing process attempted over centuries by historians and Talmudic scholars. Recently, these have included Hermann L. Strack, who published an English-language Introduction to the Talmud and Midrash. His books explain that it is clear that the creation of the Talmud spanned centuries, perhaps millennia, and in its current form represents the intellectual work of hundreds, perhaps thousands, of rabbis, thinkers, and jurists.

As such, the Talmud, cultivated and created over centuries, is not the product of a single collaborative model. Until relatively recently, the Talmud was not even a single text. Strack argues that there was never a uniform text and that differences persisted and multiplied as the Talmud was rarely read or copied but more commonly recited from memory ([Strack1972] 77). Much of Strack's early history is concerned with tracing the oral transmission and growth of the law that eventually became the basis for the Talmud—a process that was inherently collaborative ([Strack1972] 8-25). The Talmudic model is one where, as is still the case today, criticism and commentary of the existing text is encouraged. However, unlike contemporary literary models, the best criticism was incorporated into the text itself. Although its form has now been frozen, the Talmud was designed to be a dynamic document—a written conversation over centuries.

The basis of the Talmud is law that belonged to all Jewish people. This law was based on concepts that were borrowed from other groups and cultures. While scholars have attempted to pin authorship for pieces of the Mishna on individual rabbis, they do not deny that it is the articulation of centuries of communal Jewish knowledge. It was able to grow and change with time (either intentionally or unintentionally through errors in memory) because it belonged to all Jews. As rabbis and thinkers wrote commentaries on the text and on commentaries of their predecessors' commentaries, they freely pulled from and added to the existing text. While, as in any discussion, clear attribution played an important role, control and ownership did not. The existence of divergent texts demonstrates that, over time, the book's audience felt free to modify the text to make it more effective and relevant.

In this way, the Talmud represents the early literary model of a text as a conversation. [6] In the case of the Talmud, this concept has persisted, to some degree, up until today. In his Invitation to the Talmud, Jacob Neusner repeatedly describes the Talmud as a discussion and invites his readers to join in the collaborative process by reading, unraveling, reshaping, revising, improving, recontextualizing, and then applying the concepts in the Talmud in their own lives ([Neusner1973] 26). He connects this invitation to conversation with the fact that "every Talmudic tractate ... begins on page 2; there are no page 1's because there is no beginning" ([Neusner1973] 29). With its conversational quality and with no beginning and (one must assume) no end, the Talmud exists as a text that is designed to be the product and material for a continuing collaborative process that ensures its continued organic growth.

While tradition, and perhaps changing conceptions of authorship, have frozen the text of the Talmud in its current state, its organic existence continues. Neusner's own book reproduces pages of the Talmud and engages in commentary and explication—it produces a new work with the existing texts and commentaries as its core. Unfortunately, bound by copyright and a strong system of control, Neusner's own work does not facilitate the same type of intra-textual criticism and commentary that it is based on.

The King James Version of the English Bible
While both Imperial Chinese literature and the Talmudic tradition ground themselves on collaborative processes modeled after a conversation between new and existing texts, these are hardly the only models of pre-copyright collaborative literature. Other models include texts that were explicitly designed to be "created by committee." Foremost among these examples is the King James Version of the English Bible (KJV). As such, if I had to highlight a single text in my response to the English professor mentioned in the introduction, it would be KJV.

The King James Version is a vernacular translation of the Bible, a book which is, humanly speaking, is a text of multiple and composite authorship on an unprecedented scale. The books of the Old and New Testaments are explicitly attributed to over forty men from a diverse range of backgrounds—from kings to laborers—writing from between 1500 B.C.E. through 97 C.E.[7] After only a glimpse of the collaborative processes behind the work, it comes as little surprise that many Christians refer to the product and process as a miraculous example of God's hand at work.

Given this rich collaborative foundation, it should come as little surprise that collaborative efforts have been employed in the most revered translations as well. This is evident in the paradigmatic case of the King James Version of the English Bible: the most popular Bible translation and, by many estimates, the single most influential text in the English literary canon. The collaborative process responsible for the KJV was already centuries underway when the translation was commanded by King James at the Hampton Court Conference in 1604.[8] At the end of a century that gave birth to six separate English Bible translations, King James, prompted by Dr. John Rainolds (also spelled Reynolds), President of Corpus Christi College, Oxford, set the wheels in motion for the creation of yet another translation (Daiches 65). Several months later, King James informed English bishops that he had appointed "four and fifty men," (of whom we know the names of only forty-seven), and had called for suggestions, clarifications, or specific insight on Biblical passages from "learned men" anywhere in England.

The committee assembled was "catholic and intelligent on the whole, including most of the ablest men available, whether High Church or Puritan" (Daiches 67). This ideologically diverse group was divided into six sub-groups which met at Westminster, Oxford, and Cambridge. Each location housed a group translating the Old and New Testaments. The scholar translated the text individually and in small groups. Groups came to consensus on a rendering that was then forwarded to a final committee of revisers. This final committee referred to works in Greek, Latin, Hebrew, French, Spanish, Italian and other languages making use "of ancient and modern translations . . . and consulting the old manuscripts that were available" to arrive the most informed decision possible ([Gaebelien1924] 67-69).

The results of this process of creation by committee, while not an overnight success, were nothing short of astounding. Frank Gaebelein describes KJV as "the crown of our literature," and argues that the translation offers "one of those rare cases where superlatives are not only justified but demanded" ([Gaebelien1924] 22, 72). He goes on to describe it as "immortal poetry, enduring in beauty because it reflects so truly the inspired original" ([Gaebelien1924] 75). KJV is and continues to be the highest selling Bible translation worldwide.

KJV succeeded where Wycliffe's, Tyndale's, Bishops', and Matthew's Bibles failed because it employed more translators, more scholars, and more input from the greater educated community; its success was insured by its unprecedented collaborative creative process. Its position has only been challenged by translations that both incorporate the work of previous translators (including the KJV committee) and the work of large numbers of contemporary scholars.

The scholars producing KJV were funded through the system of honorarium mentioned previously. They borrowed at will from existing Bible translations and from their peers. The product of their work was the property of the entire community—neither they nor anyone else owned the translation. [9] KJV succeeded where singularly (or simply less collaboratively) authored translations failed because it was the product and process of intense collaboration. The freedom to collaborate not only ensured the persistant popularity of KJV over almost four centuries, but provided the foundation for several derivative translations including the popular Revised Version and the American Standard Version (ASV).

Conclusions about Pre-Copyright Authorship and Collaboration
KJV, the Talmud, and Imperial Chinese literature serve as impressive examples of the power of early collaborative processes. It is clear that the production of all three would be impossible by any individual. However, they are also symbols of the power of the unhindered access to information, knowledge, and existing works that facilitated their collaborative creation. Collaboration on the scale necessary to assemble the Talmud or KJV must be executed in an environment where the type of widespread borrowing and textual synthesis employed in the creation these texts is possible and even encouraged. Copyright and systemic control are fundamentally at odds with the type of freedoms necessary to produce such works.

Collaboration During the Birth and Early Life of Copyright
As I mentioned in my introduction, the Statute of Anne in 1710 did not mark an instantaneous shift in attitudes toward authorship and artistic creation. At its birth, copyright was lobbied for and designed to benefit publishers alone. For at least the first century of its institution, authors continued to write in the ways they had before. They borrowed as they had before; they collaborated as they had before; they plagiarized as they had before. Collaboration in the forms popularized before the institution of copyright remained popular.

However, by selling the rights to their ideas, authors were presented with a new system of compensation for their work: a way to "live by their pen." They realized that by solidifying their access to these rights, they might insure their ability to make a living. This coincided, and was intimately connected, with the explosive growth of the publishing industry in Europe. Authors felt they needed to insure compensation for their intellectual productions and saw their copyright, described in the Statute of Anne and similar acts in other countries, as an available method for achieving this goal.

To emphasize the importance of copyright—which was initially created in the service of publishers—authors, led by Romantic poets in the eighteenth and nineteenth centuries, introduced a new conception of authorship. These authors must have been aware that the collaborative nature of their literary production sat in direct opposition to calls for the institutionalization of control systems justified by conceptions of Romantic authorship. Perhaps authorship was defined in terms relative to the previous system of unhindered borrowing and collaboration. In any case, these authors seemed comfortable with the hypocrisy of their position.

John Keats
Keats was one such poet who espoused a Romantic conception of authorship while employing collaborative practices in the creation of his poems. Keats explicitly placed his poetry within a larger social context of its creation, revision, reception and influence. During the eighteenth and nineteenth centuries, the mechanisms for collaborative writing took explicit form in the creation of coterie groups of authors that acted as forums for idea interchange, discussion, manuscript circulation, critique and small-scale publishing.[10]

Steeped in this culture of collaboration, Keats himself was influential in the creation, promotion, and continuation of several such groups. The most famous, and consequently most well documented, is the now famous "Cockney School" that included, at different times in its life, the company of Shelly, Byron, Keats, Hunt, Reynolds, Smith and Hazlitt. Through their letters and correspondence, it is clear that each of these poets turned to associations and interactions within the group as a means of cultural production ([Cox1998] 4). While the nature of the association was unstated, unclear, and inconsistent—some shared, borrowed, or took advantage of the association more than others—we know that this group of artists, writers, and intellectuals conceived of itself as a coherent circle, "something between a manuscript coterie circle . . . and [a] kind of self-consciously avant-grade movement" ([Cox1998] 20-21).

We also know that Leigh Hunt provided the nexus around which the group was organized. Hunt published several journals, most notable of which was The Examiner, organized coterie meetings, played host to poets, artists and intellectuals and ran his much maligned sonnet contests to encourage the creation and critique of new works. In each of these ways, Hunt provided a public space for the discussion and exchange of idea necessary for his ideal literary process, one that took part in a social sphere ([Cox1998] 7). The group's collective work included the production of commonplace books, collaborative projects, and "contest" poems as well as major individually attributed efforts which were executed in the context, and with the assistance, of the members of the group through a system of manuscript circulation and revision.[11]

The influence of members of this group on each other is described in detail by Jeffrey N. Cox in Poetry and Politics in the Cockney School in a thorough analysis of themes, images, characters and plots shared between members of the group and through explicit affiliation in dedications and inter-textual references. For example, Cox describes connections from every poem in Keats' 1817 Poems to at least one other member of the group. He draws parallels along ideas that occurred in different members poems, especially political ones, and employs evidence of political, economic, and literary support between group members ([Cox1998] 84).

While Cox's book acts as an analysis of the collaborative relationships that gave birth to the ideas and themes behind Keats' poetry, Jack Stillinger, in Multiple Authorship and the Myth of Solitary Genius, explores the nature of the collaborative system of revision and production at play in the stylistic creation and evolution of Keats' Sonnet to Sleep and Isabella. In both cases, Keats' friends and publishers were instrumental in the creation of the finished product ([Stillinger1991] 17-21, 29-45). Isabella, with a nearly exhaustive set of revisions available to historians, provides a particularly useful example of the collaborative processes at work in Keats' poetry.

Isabella was written and then revised once by Keats with the input from friends and colleagues including J. H. Reynolds. It was then passed to Richard Woodhouse who originally copied the poem into shorthand before re-expanding it and introducing several changes in wording and punctuation in the process. Woodhouse subsequently revised it twice. Woodhouse's second version served as the printer's copy when the poem was first published ([Stillinger1991] 26-29). Handwriting from the poem's printer, John Taylor, is also visible in the final manuscript version ([Stillinger1991] 29). Keats' collaborators made additions, deletions, rewordings and wholesale rewritings while prompting Keats to make other revisions on his own ([Stillinger1991] 34). These revisions were often made independently—and perhaps divergently—of Keats' surmisable intentions ([Stillinger1991] 39). Other alterations appear to have been made by Keats, Woodhouse, and Taylor working together ([Stillinger1991] 44). The contributions were so substantial and the contributors so passionate and involved that Stillinger claims the collaborators demonstrated a proprietary interest in the work.

Stillinger notes that while none of Keats' collaborators' changes deal with theme, character, or plot, it is the stylist nature of the poetry they focused on that makes up the acknowledged, "Keatsian" qualities of the work ([Stillinger1991] 30). Cox argue that Keats' association with the Cockney School reveal that the theme, character, plot and messages of Keats poem are also rooted in collaborative associations. If readers are to believe both Cox and Stillinger, as I believe is warranted, we must approach Keats as having communicated collaboratively conceived messages and themes through collaborative mechanisms.

This is a far cry from the ideal of Romantic authorship usually attributed to Keats and his contemporaries. Many Romantic scholars have attempted to downplay both the importance of the coterie process and the role of the collaborative revisions. In some cases, they've gone so far as to blame Keats' early or weaker poetry on collective processes while elevating his later work as the product of his unrestrained Romantic individualism ([Cox1998] 13). In reality, Keats' collaborative relationships played such an essential role throughout his creative life, that he returned to live with Hunt, his primary collaborator, during his final illness ([Cox1998] 84).

As Keats penned his poems, copyright had already begun to enshrine the concept of Romantic authorship. As a result, Keats, while he worked in massively collaborative processes, took sole responsibility for the texts. At the time, his close literary associations and his collaborative relationships with friends and publishers were unstated because they were assumed; there was an acknowledged discrepancy between the way authors created and the way they drafted their bylines. Authors saw nothing wrong with gaining compensation through false or exaggerated claims of individual authorship; that was simply how things were done. However, this unwritten information has been lost with time or simply ignored by critics and scholars. It is only through recent historical work by researchers like Cox and Stillinger that the collaborative nature of Keats work has been revealed.

As a result of this type of research, Keats' widespread collaboration becomes evident. It is also clear that the collaborative processes were facilitated through Keats' lack of control over the poems. While Keats had a limited copyright to this work, he showed little interest in and demonstrated little control over texts themselves. He "almost certainly" did not read the final printers proofs—with full knowledge that his printer made editorial and stylist changes. This attitude, unsurprising in a literary culture steeped in collaboration, allowed Keats' collaborators unhindered access to the poetry and cultivated the culture of collaboration and critique that made the type of cooperation possible that helped Keats create to his best ability—one greater than Keats working alone.

Samuel Taylor Coleridge and William Wordsworth
This culture of cooperation is mirrored in the literary lives of Samuel Taylor Coleridge and William Wordsworth a generation before. Like Keats, the pair produced many of their literary works through intensely collaborative methods. Also like Keats, criticism steeped in the ideology of Romantic authorship has attempted to dismiss the importance of collaboration in their work. Unlike Keats, Wordsworth and Coleridge played a huge role in the Romanticization of their poetry and processes. Wordsworth promoted the ideal of Romantic authorship emphatically and is largely responsible for the Romantic lens through which critics now view his work.

That said, Lyrical Ballads, with poems by Wordsworth and Coleridge, is probably the most famous coauthored book in the English language. The idea of collaboration by Wordsworth, generally viewed to be one of the most original English authors and often credited with the success and dominance of Romanticism, is difficult for some critics to understand ([Stillinger1991] 96). In fact, they are not alone: Wordsworth himself seemed to have difficulty reconciling his own coauthorship as the book was first published anonymously in 1798 with more than ten references to a single author in the books advertisement. In the three subsequent editions, the byline remained singular and, when it finally mentioned "the assistance of a friend," it did so without mentioning Coleridge's name. Coleridge's name was not connected to Lyrical Ballads's first poem, his own The Rime of the Ancient Mariner, until almost two decades later when it was published in a separate collection ([Stillinger1991] 70).

While Coleridge wrote The Rime of the Ancient Mariner, the story was developed by Wordsworth and Coleridge collaboratively with Wordsworth providing a rough version of a themes and some major plot elements. In a similar way, Coleridge claimed credit for half the preface to Lyrical Ballads ([Stillinger1991] 71). Additional collaborators on the book included Thomas De Quincy and Humphrey Davy, a friend of Coleridge, who each made significant contributions to the work in the role of editors while Wordsworth's willing amanuenses probably played no insignificant role in the shaping of the text ([Stillinger1991] 71). This of course, does not begin to discuss the literary influences on Coleridge and Wordsworth, particularly Milton, or the role these influences played in shaping, inspiring, or directing the text. One additional, and additionally interesting, source of inspiration for Wordsworth was Dorothy Wordsworth's journals. Her Alfoxden Journal of 1709 and the Grasmere Journals of 1800 and 1802 contain passages that many critics believe Wordsworth employed as the basis for all or parts of Beggars, Resolution and Independence, I wandered lonely as a cloud, and a number of other poems and passages ([Stillinger1991] 72). All of these pieces were appropriated and published without acknowledgment.

Coleridge, for his part, was famous, perhaps infamous, for his borrowings from other authors. Convincing arguments have been made that Coleridge's Frost and Midnight attempts to mimic, either for reproductive, synthetic, or critical purposes, a poem from William Cowper's The Task ([Stillinger1991] 101-103). More interestingly, and problematically for Coleridge's scholars and defenders, are Coleridge's unacknowledged borrowings, often in the form word-for-word translations or direct paraphrases, of large sections of writings by German philosophers, most notably Schelling, in Coleridge's Biographia Literaria. Surveys of the Biographia and antecedent literature have shown that up to twenty-five percent of the text is lifted, to one degree or another, without attribution from German sources; the figure reaches as high as thirty to forty percent in several chapters ([Stillinger1991] 104).

While Coleridge was attacked for his plagiarisms by his friends and during his lifetime, his supporters, like G. N. G. Orsinis, have defended Coleridge with the justification that "a genius can be creative even when he is borrowing" ([Stillinger1991] 105, 107). While this exposition of collaboration is in complete agreement, their argument is, after all, one that much of this essay echos and affirms, the defense is uncharacteristic, perhaps even incompatible, with the Romantic notions of authorship that both Coleridge and his defenders embrace and espouse.

However awkward, Coleridge's plagiarism seems somehow congruous with his famous, and famously hidden, collaboration with Wordsworth on Lyrical Ballads. Historical research has established that the decision to attribute Lyrical Ballads to a single anonymous author, and then to withhold acknowledgment of Coleridge's contributions for three subsequent editions, was a decision to which both of the books authors agreed ([Stillinger1991] 70). Both authors realized that the need to claim copyright and gain compensation was simplified and strengthened by exaggerated or falsified claims of singular authorship. As much as Wordsworth espoused his ideal of a Romantic genius, perched alone in the wilderness, drawing all creative inspiration from within, it was not a description of a method of creative production that even he employed consistently.

Reconciling the conflict between the Romantics' professed ideologies and their actions can be difficult. By speaking in Parliament for the creation of authors' natural rights, Wordsworth was attempting to manipulate the connection between Romantic authorship and legal mechanisms of textual control and ownership; Romantic authorship was, from its birth, intertwined with the politics of copyright. But there remains a deep irony in that the processes that allowed Wordsworth and Coleridge to demonstrate their greatest achievements as creative or artistic geniuses, easily the strongest evidence for their claims, were collaborative. By dropping Coleridge from the byline of Lyrical Ballads, by failing to acknowledge Dorothy Wordsworth's contributions, and by omitting reference to Scheller's role in Biographia Libraria, Coleridge and Wordsworth make the shortcomings of a system based on their concept of authorship and strong individual control abundantly clear. Modern readers must assume that, raised in a literary culture of assimilation, borrowing and critique, the Romantics considered this attitude a necessary and acceptable hypocrisy for those aiming to "live by their pen."

During the Shift Conclusion
These problems of attribution are representative of the first century of copyright and the first awkward legal steps into the discourse of Romantic authorship. They provide a window to the types of contradictions and clashes between persistent collaboration, Romantic authorship, and systems of ownership. Over time, the popularization of Wordsworth's ideal authorship has strengthened and reinforced copyright to the detriment of collaboration. Contemporary authors must conceive of themselves in Wordsworth's terms but cannot collaborate in the same unapologetic fashion in the context of more rigid technological, social, and legal systems of control. As the publishing industry has been reshaped by these powerful and lucrative systems of control reinforced by the discourse of Romantic authorship, the contradictions that Wordsworth and his contemporaries happily ignored have shaped the dominant systems of literary production.

Contemporary Collaboration and Control
Wordsworth could not have conceived of the effect that his conception of Romantic authorship would exert. Over the nineteenth and twentieth centuries, copyright jurisprudence became intertwined with the Romantic conception of authorship. In his essay on The Author Effect: Contemporary Copyright and Collective Creativity, Peter Jaszi details some of the ways that this has played out at turning points in American copyright jurisprudence, including Burrow-Giles Lithographic Co v. Sarony, Feist Publications, Inc. v. Rural Telephone Service, Rogers v. Koons and Basic Books v. Kinkos's Graphics Corp. He argues that in each case, the court's opinion was, often inaccurately or inappropriately, influenced by a firm adherence to the concept of Romantic authorship.

With the dominance of Romantic authorship and the continued expansion of the scope and term of copyright over the past century, the environment for literary production is controlled in a manner that is increasingly unaccommodating to collaborative models of literary creation. Spurred by the growing dominance of capitalist economics, copyright, originally a privilege, became interpreted as a form of "intellectual property." These systems of strong individualized control helped create an environment that has fostered genres and a publishing industry based on Romantic authorship and strong control to the detriment of preexisting and new collaborative models. While explicit collaboration still occurs, even widely, it is usually in awkward, hidden, or relatively ineffective or dis-empowering forms.

Collaboration between T.S. Eliot and Ezra Pound on Eliot's The Wasteland
One such example can be seen in Eliot and Pound's collaboration on The Wasteland. Their relationship is particularly useful in a study of twentieth century collaboration because the nature of the collaboration between the two great poets is clearly documented in the Eliot's extant manuscripts with Pound's scrawled markings and marginalia. It is also interesting as an example of an extensive collaboration that has tested the limits of the idea of Romantic authorship for many critics.

The details of the editorial changes made to The Wasteland are documented in a facsimile edition of the manuscripts published by Valarie Eliot. They are also summarized concisely by Jack Stillinger in his chapter on Pound's Waste Land. In short, Pound reduced the poem from over 1000 lines to its current 434. In the process, he focused and limited the poem's message and eliminated a sarcastic tone. The critical view, with only the exception of a handful of scholars, is that Pound's edited version is an undeniable improvement. Jack Stillinger concisely sums up the popular critical response:

The majority view is that the 434 lines of The Waste Land were lying hidden from the beginning in the 1000 lines of draft, rather in the manner of one of Michelangelo's slumbering figures were waiting to be rescued from the block of marble. But Michelangelo, in this analogy was both artist and reviser simultaneously. In the case of The Waste Land, it took one poetic genius to create those 434 lines in the first place, and another to get rid of the several hundred inferior lines surrounding and obscuring them ([Stillinger1991] 127-128).

Eliot, who was mentally infirm and hospitalized during the period of writing and revision of the poem, acquiesced to almost all of Pound's revisions and suggestions ([Stillinger1991] 137). Stillinger brings attention not only the extent of Pound's changes but connects the collaboration to an argument that the resulting text constitutes a co-authored work.

There is additional evidence to support this claim. In the first release of the poem, Eliot dedicated the poem to Pound as "il miglior fabbro," an Italian phrase meaning "the greater craftsman." Through his life, Eliot was also upfront about the importance of Pound's additions to the work, describing, quite accurately, the way that Pound had "turned The Waste Land from a jumble of good and bad passages into a poem" ([Stillinger1991] 132). However, the manuscripts were not released by Eliot during his lifetime; they were released by Valarie Eliot, T. S. Eliot's widow, in 1971.

After their release, descriptions of the multiple authorship of The Waste Land, while supported in the textual evidence, faced fierce opposition from many critics and supporters of Eliot. Some critics, a number of whom had published major books on Eliot in the previous years, clung to their image of Eliot as a Romantic genius by making statements that attempted to minimize or trivialize Pound's contributions ([Stillinger1991] 132-134). Their arguments were simply unsupported by the textual evidence. It is impossible to deny that without Pound, The Wasteland would be an extremely different, and substantially less impressive poem.

While Eliot and Pound played different, unquantifiably important, and equally essential roles in the creation of the poem, Pound's role is, typically denigrated, at best, to the role of "an editor." Rather the describing the The Wasteland as a vibrant creative collaboration between two brilliant poets, critics substitute the image of Pound suggesting simple editorial changes to Eliot's poem. This unfortunate configuration is forged in the conceptions of authorship defined and sustained by an discourse of ownership: The Wasteland is Eliot's poem. While I am not confident that I understand exactly what Stillinger desires in his calls for "multiple authorship," I'm not sure that I agree that another name on the byline of The Wasteland is a particularly useful goal. That said, his critique is sound: there is a deficiency in a system of authorship and ownership that cannot acknowledge Pound for the important role he played in the creation of The Wasteland.

The short stories of Raymond Carver and his editor Gordon Lish
In the way suggested by Eliot's relationship with Pound, the role of the editor is important to any understanding of twentieth century collaboration. Max Perkins, the editor for Fitzgerald, Hemingway, and Thomas Wolfe, is one such example. Often, as is the case with Eliot and Pound, the term "editor" is applied with the goal of downplaying the role of the less-authoritative collaborator. However, the roles of "author" and "editor" also serve to act as terms that, when applied in the context of a new literary relationship, place firm limitations on the nature of the collaboration allowed to transpire. Crossing these limits can be disastrous for an author's reputation by depriving her of the sole authorship of her work. Through an unusual attempt to claim responsibility for a text, the increasingly common conflict between author and editor was recently highlighted in the relationship between popular 1980's short story author Raymond Carver and his friend and editor Gordon Lish.

Carver, considered by many to be America's most important short story writer when he died of lung cancer fourteen years ago, pioneered and popularized a dark minimalistic literary style that exploded in popularity during the 1980s. In an 1998 article in the New Yorker, D.T. Max examined many of Carver and Lish's original manuscripts and met with Lish himself in an attempt to investigate Lish's increasingly loud claims claims that "he had changed some of the stories so much that they were more his than Carver's" ([Max1998] 35). Max goes into some detail on the changes marked in the manuscripts which include Carver's 1981 collection, What We Talk About When We Talk About Love, in which Lish "cut about half the original words and rewrote 10 of the 13 endings." Editorial work of this extent was typical in many of Carver's stories, some of which Lish cut by over seventy percent before they were published ([Max1998] 37).

Lish's claims of responsibility for elements of Carver's stories prompted similar claims from others in Carver's life. Carver's wife, the poet Tess Gallagher, has made claims on elements of Carver's work arguing that several of his plots were originally hers and comparing Carver's actions to "stealing." After Carver's early death, access to his manuscripts and to his living literary partners has made the collaborative processes behind the creation of his stories unusually transparent to the general public. The scene appears to have been one of rich collaboration between his friends, family, and editors—all of which was hidden during this life.

Many of Lish's editorial interventions added touches that were later called "trademark techniques" of Carver's. Max claimed that Lish's "additions gave the story new dimensions, bringing out moments that I was sure Carver might have loved to see" ([Max1998] 38). These edits were so extensive that in a letter to Lish, Carver expressed, "fear [of] being caught" ([Max1998] 40). In 1982, Carver pleaded with Lish, "please help me with this book as a good editor, the best . . . not as my ghost" ([Max1998] 40). Carver was worried because he felt his own contributions to the work, which were, after all, almost the entire texts of the stories, threatened by Lish's subtractions. Moreover, he felt his own originality and creativity threatened by the fact that his work was heavily edited; the fact that he had collaborated made me him feel like a fraud. Because the edits were extensive, Carver felt that Lish's role was more than what an editor's "should" be but was unwilling or incapable of interacting with him as an explicit collaborator or coauthor; there was simply no classification in the dominant system of literary production for their type of collaborative work. The stories are, in most critic's opinions, better as a result of Lish's edits; the minimalist style that Carver became famous for can be almost completely attributed to Lish if the available manuscripts are to be trusted as representative.

Few will argue that Lish's changes were insignificant. However, in attempting to claim ownership of the work, Lish is forced to demonstrate more than mere collaboration; he needs to demonstrate "joint authorship."[12] The type of collaboration necessary for legal "joint authorship," must exceed that of the "normal" editor-author relationship. Lish's predicament highlights the fine line that the contemporary authorial-editorial relationship straddles. Editing is an acknowledged and widely used method for literary collaboration because the product of editing is defined as uncopyrightable and therefore transfers no rights of ownership or authorship from the author to the editor. Both Eliot and Carver demonstrate that authors want, even need, to collaborate. They realize that by working with others their work becomes better. But in order to gain ownership, attribution, and remuneration for their work, they must not collaborate "excessively." To succeed in the contemporary literary world, an author must collaborate as much as possible without losing authorship, and by extension their ability to claim sole attribution and ownership of their text. In the end, writers collaborate less, and the world is left with an inferior texts.

Contemporary Industry Collaboration
Carver and Eliot's secret collaborations, hidden at least during the authors' lifetimes, imply that explicit creative collaboration is rare in our current literary landscape. However, the power of collaborative work is too powerful a model to avoid altogether. Over time, the writing and publishing industry has manipulated both the model of collaboration and the system of copyright to facilitate collaboration in several notable ways. Dominant in this landscape are corporately "authored" texts created under a process known commonly as "works made for hire" (described in more depth in the Section called The Works Made for Hire Doctrine in Chapter 4) and jointly authored works, as represented by the terms, "with," "as-told-to," and "and" in by-lines ([Barbato1986]).

Described here briefly and in more depth in the Section called The Works Made for Hire Doctrine in Chapter 4, the works made for hire doctrine, a common part of copyright jurisprudence, states that authorship (and as a result ownership) for works created within the scope of employment rest with the employer. The implications for the promotion of collaboration are not difficult to trace; if an employer, like Disney, hires five brainstormers, ten script writers, five composers, ten musicians, twenty voice actors, fifty animators, and twenty editors to produce a movie, the product of this work, because it was created within the context of contractual employment with Disney, belongs not to the one hundred and twenty artists who created the film but to the corporation. With Hollywood corporations as a prime examples, this model has been instrumental in facilitating wide-scale corporate collaboration. However, it is aggressively hostile to the type of "serial collaboration" and borrowing from existing texts that was essential to earlier models of collaborative writing.[13] The works made for hire doctrine is not a method for aggregation of ownership or authorship or a method for collective creation or control. In fact, it ensures that the creator receives payment but no rights at all.

This is not to imply that collective authorship with implications of collective control is impossible. In actuality, joint authorship is steadily increasing in popularity and influence. However, joint authorship operates in an environment hostile to collaborative work, and, as a result, is difficult at best. Empirical studies have shown that instances of joint authorship—a measurement taken by tallying books and articles with more than one person on the byline—are becoming increasingly popular and prominent ([Barbato1986] 1986). While these collaborations are important in highlighting the persistent power of collaborative writing, they are hindered by the hostile climate of control and authorship created by copyright.

In an article written for science-fiction authors on How to Collaborate without Getting Your Head Shaved, Keith Laumer, an author and collaborator, ends his short piece with the advice, "if you possibly can, write it yourself. Collaborations, like marriages, should only be undertaken if any alternative is unthinkable" ([Laumer1977] 217). Mark L. Levine wrote an article for Writers' Digest titled Double Trouble where he urges potential collaborators to first sign a complex contract that clearly delineates both the roles that the collaborators will play in the creation of the book and the division of payment ([Levine1985] 34-35). In an article for Writer, Leonard Felder points out that not only should potential collaborators first agree to a division of royalties and payments, but that they must have "a written agreement on . . . the way your names will be listed on the book's cover" ([Levine1985] 22). Unfortunately, this advice is all perfectly sensible in today literary climate. While many of these articles also mention the potential benefits of joint-authorship, they explicitly, and accurately, approach the collaboration as a business relationship; their emphasis is on avoiding the pitfalls of such joint work.

In reality, no literary collaboration can be divided cleanly into portions or dissected on a contract sheet. As evidenced by Eliot and Carver, neither can roles such as "editor," "author" and "coauthor" describe the spectrum of meaningful literary collaboration. However, under current systems of literary production defined by copyright and Romantic conceptions of authorship, writers have few other options. By emphasizing ownership and control as the primary, and in most cases the only, method of compensation for literary work, meaningful collaboration becomes difficult in all cases and impossible in most. Rather than borrow and work together, authors will work alone. Rather than borrow an idea, passage or theme from another novel and risk a copyright suit, authors are more likely to not include the theme or passage at all. The fact that joint-authorship and collaboration can function, and even experience massive growths in popularity, in this hostile environment, is testament to the power and of collaboration. Without a strong system of control shaping the landscape of literary creation, there is no guessing what other works we might enjoy.

Conclusion
There is no denying that collaboration is persistent, but a handful of citations might have demonstrated this point as effectively as this essay. There is no denying that collaboration is effective, but this also would have been well served will less effort and spilled ink. The real conclusion, one that has been echoed throughout this piece by the accumulating evidence I've offered and alluded to in this essay, sits at the interstice of collaborative writing and control: when we give control of literature to individuals, collaboration is less common, less meaningful, and less effective.

Given the evidence in this essay, restating this conclusion seems almost unnecessary. Yet, it remains absolutely essential; in the dominant modes of literary criticism and production, it is almost wholly ignored. History has shown that the importance and power of collaborative creation is one of the most powerful mechanisms for the creation, organization and dissemination of knowledge. The dominance of the Romantic notion of authorship has forced us to ignore both the importance and power of collaborative creation and the effect that this type of ownership has on collaborative models. We need not ignore the power that ownership and individualized control bring to the table, but we should not dismiss collaborative work because it's incompatible with the ideology that lets us control, and amass fortunes, from our ideas and from those of others.

It is established that authorship reflects ownership. In today's literary world, one where collaboration is growing in its use and influence, ownership—at least as defined by copyright—does not rest with the creator or creators of the work but with the name or names in the byline. There is only room for one—maybe two—individuals in this space. In response, we push collaborators into our bibliographies, acknowledgment pages, or out of the book altogether; still, perhaps this is unavoidable; perhaps it is even excusable. It becomes inexcusable when we limit the extent of our collaborative enterprises because we are unable to represent the nature of our collaborations in a way that will ensure attribution or compensation for the work. As the last thirty years of Foucault and his supporters have shown, attacking and deconstructing authorship is not enough, we need to deconstruct the systems of control that have enshrined and are perpetuating these conceptions.