WAC, WID, ECAC, CAC, CXC, LAC — VAC?
Abstract: The problems of collaboration take on new shapes and meanings when writing/composing moves into the visual realm. This claim is based on our experiences working with the Bush-Hewlett Grant Project at Spelman College, which was designed to foster collaboration through the building of course-based Web sites. Our study analyzes three sites, and the processes that gave rise to them, using a qualitative case-study approach. Findings indicate that among the key concepts challenged by collaborative projects in visual media are author, audience, and coherence. Further, we argue that collaborative work in visual media must be viewed as an ongoing process which may move outside of the assumed geographies of academic space, such as "course."
Collaboration is fine if that's what it is, but not when it's a lot of independent work and then the people come together.
"What amazes me is how so many of us seem to be trying to use the new technology to do the same old thing with students.
—Beth Baldwin, "Evolving Past the Essay-a-saurus"
For more than 25 years, collaborative work has been advanced as a useful strategy for writers, including students in writing-intensive classes (Beach, 1986; Bruffee, 1978, 1993, 2003; Ede & Lunsford, 1990; Tompkins, 1990). Collaboration offers potential benefits including maximizing knowledge available; incorporating multiple perspectives; fostering critical thinking; energizing participants; and enabling a richer, more expert final product (Bruffee, 1993; Ede & Lunsford, 1990). But for almost as long, the problems of collaboration have also been discussed. Harvey Wiener (1986), for example, points out that the "effectiveness" of students' collaborative work relies more heavily than most evaluations credit upon how a teacher chooses to design and manage students' collaborative processes. John Trimbur (1989), while arguing in favor of collaborative work, names two further potential problems: that collaboration "enforces conformity" (p. 602) and/or over-emphasizes the impact of a small discourse community while de-emphasizing larger social discourses (p. 603). Andrea Lunsford and Lisa Ede (1994, 2001) have explored the ways that collaborative work may subvert our notions of authorship. Finally, Darsie Bowden (1996) adds that students' fears of being plagiarized may cause resistance when students come together to share their writing in peer-response groups. Compounding the complexity of such discussions, the term collaboration is often used without clear definition. The notion of collaboration, then, often operates as a floating signifier. As teachers we know we're supposed to encourage it, but often we aren't quite sure what "it" is, nor what complications may ensue when we put "it" into practice.
Continuing this conversation on the problems of collaboration, we wish to add this point: when "writing" moves into the visual realm, these problems are not merely transposed, but transmogrified. In other words, teachers attending to the recent surge of interest in visual literacies will encounter new, and probably unexpected, challenges and opportunities as a result of their attempts to encourage collaboration in visual spaces such as the World Wide Web. We do not claim to address all potential challenges and opportunities; instead, we offer claims based on our own experiences in working with the Bush-Hewlett Grant Project at Spelman College, a project specifically designed to foster students' and teachers' collaboration through Web-site building. This study analyzes three course-based web sites, and the processes that gave rise to those sites, using a qualitative case-study approach (Merriam, 1998). Through our analysis, we demonstrate some of the ways that the problems—and indeed, the definition(s)—of "collaboration" may change shape as writing projects move into visual media. We also offer some pedagogical suggestions, aimed both at avoiding pitfalls and also at using collaboration productively, for teachers who incorporate visual literacies into their writing-intensive classes.
Karen Naimool is a student assistant on the Bush-Hewlett Grant Project. Her work-study job involves providing technical support on the project, such as Web design and assistance with programs including Dreamweaver, Photoshop, and iMovie. On April 26, 2005, she and I sat together in front of a Macintosh computer, toiling over the Web site that had been produced—supposedly—by two of my composition classes and by another class, an upper-level literature class. In theory, the project proposed by myself and the literature professor had engaged our students in critical explorations of the notions of disability and normalcy, and the Web site provided both an overview of those explorations as well as resources for other members of the Spelman College community interested in disability. In practice, however, the classes had ended four months before, and students had moved on, leaving a large folder of unidentified Dreamweaver files and a skeletal Web site, half plan, half realization. So Karen and I sat alone, "finishing" the site—cleaning out folders, adding images, re-organizing content and design—because the other professor and I had to present our work to the college in a few weeks, and I wanted to have something to show for it. To put it bluntly, I wanted the site to look as if we (or do I mean I?) had accomplished something.
As Karen and I worked, I remarked that I'd been wondering why, in a project that was supposed to be collaborative and student-centered, I was spending so many hours alone in front of the computer, revising the web site to make it—in my mind—presentable. With characteristic acumen, Karen replied, "I see this on these projects all the time. A lot of the time they aren't really collaborative. Collaboration is fine if that's what it is, but not when it's a lot of individual work and then the people come together." As the student assistant for numerous Bush-Hewlett projects, Karen had often seen students and professors engaged in the same activity in which I found myself absorbed: attempting to piece together content and design assembled haphazardly by a large group of others who had since disappeared from the project, and struggling to form this collection of words and images into some sort of coherent text.
Karen's comment awakened us to a fundamental and yet easily overlooked fact: collaboration is often invoked but rarely defined. In this way it operates like the related term "community," which as Joseph Harris (1989) points out has "extraordinary rhetorical power" (p. 13) despite (or perhaps because of) its tendency to float. Labels proliferate—collaborative learning, cooperative learning, and group learning are a few of the more common—and meanings proliferate even more. Ede and Lunsford (1990) found, for example, that much of the research on writing collaboration to that point involved the assumption that a text is first composed alone, then worked on by a group during the revision process. Where definitions are attempted, Ede and Lunsford argue, they often conflict, and may range from group planning of a text to peer editing of a text. Singular Texts, Plural Authors was published more than a decade ago, but many of its observations remain true today. For example, Kenneth A. Bruffee (2003), a leader in the theory and practice of collaborative learning since the 1970s, argues that when we refer to collaboration we often mean "interdependence," which is a "larger, more complex concept" than merely "working in a group" (p. 18).
While bearing in mind the importance of operationalizing a term enough to study it meaningfully, we also wish to make use of Ede and Lunsford's (1990) contention that the definition's mobility can provide "perspectives by incongruity" (p. 16). In other words, a strictly operationalized definition in a study such as ours might cause us to overlook conflicting perspectives among teachers and students which could positively inform our findings. This is particularly true since "collaboration" was not explicitly operationalized for the teachers taking part in the Bush-Hewlett Grant Project, and the teachers in turn tended to operationalize "collaboration" by example rather than by stated definition.
Our decision, therefore, is to conduct our study with a working definition drawn from Deborah Bosley (as cited in Ede & Lunsford, 1990, p. 15): "two or more people working together to produce one written document in a situation in which a group takes responsibility for having produced the document." This provides an accurate reflection of the expectations set forth by the terms of each Bush-Hewlett project while still allowing room for mobility. Indeed, the purpose of our study is to question our own definition—What does it mean to "work together" in a medium of composition that is visual as well as verbal? What is a "written document" in a visual medium? How does the document's appearance in the public space of the Web affect the issue of the group's "responsibility" for the document?—rather than to use it as a means of testing a pre-determined hypothesis.
Interest in visual composition has recently exploded across the curriculum, particularly in the area of writing. A glance at the publishers' tables at any writing-related conference reveals prominent displays of visual-rhetoric textbooks, in print or on disk or both: Seeing & Writing (2000, 2003), Picturing Texts (2004), Ways of Reading Words and Images (2003), and 9 Visual Exercises (2004), among others. Kathleen Blake Yancey's (2004b) address at the Conference on College Composition and Communication, "Made Not Only in Words: Composition in a New Key," itself a visual/verbal text, argued that teachers of writing across the curriculum must attend to the ways that "writing" now means composition in various visual and digital media (p. 299). Programs in visual rhetoric, visual literacy, and/or visual culture have sprung up across the country, including a minor in Film and Visual Culture at Spelman College. Anthologies, articles, and special issues of journals (such as this one) have proliferated. This shift, often referred to as "the visual turn," has dramatic implications for our understanding of many elements of composition, including collaboration. Gunther Kress (1999) argues "that written language is being displaced from its hitherto unchallenged central position in the semiotic landscape, and that the visual is taking over many of the functions of written language" (p. 68). However, the exact nature of these implications—just how this shift will play out—is only now being realized.
We theorize visual composition by beginning with the assumption that any composed text—words, images, both, other—is a palimpsest. In other words, it is underlaid and overwritten with earlier drafts, ideas, words from other authors, feedback from other readers, and so forth. Bronwyn Davies (1993) explains palimpsest as
a term to describe the way in which new writings on a parchment were written over or around old writings that were not fully erased. One writing interrupts the other; the old writing influences the interpretation of the imposed new writing and the new influences the interpretation of the old. But both still stand, albeit partially erased and interrupted. New discourses do not simply replace the old as on a clean sheet. They generally interrupt one another, though they may also exist in parallel, remaining separate, undermining each other perhaps, but in an unexamined way. (p. 11)
As Davies makes clear, the important feature of palimpsest as a metaphor for teachers of writing is its ability to capture the many-voiced nature of a supposedly single-authored composition. Even if a student writes only one draft, consults no outside readers, and cites no other authors, her text is nevertheless a palimpsest in the sense that all texts are, in Mikhail Bakhtin's (1986) term, heteroglossic.
What does this mean when composition moves into the visual realm? Yancey (2004a) argues that a student's self-representation in the visual medium of the Web-based portfolio differs from a print portfolio in important ways. While a print portfolio encourages the author to arrange her self-representation in linear terms, with an eye to "always-forward processes of development" (p. 743), a portfolio designed for the Web encourages the author to demonstrate that she "can make multiple connections [through] multiplicity and elaboration ... [and] can offer a reader multiple narratives extending ever outward" (p. 751). Yancey, following Michael Davidson, offers the alternative term palimtext to describe a composition such as a portfolio, which is read "in its own developmental context" (p. 741). The visual/verbal medium of the Web can allow for greater transparency of the process of adding-on, doubling back, and dialogue that gives rise to any composition. In theory, then, the building of a Web site is an ideal project not only for students and teachers to engage in collaboration, but also to make visible that process of collaboration.
However, as Yancey makes clear, not all Web sites are "Web-sensible" (p. 745). A Web site may be simply "print uploaded" (p. 745), that is, a text governed by the limitations of print text but transmitted in a digital medium. In fact, as our study of the Bush-Hewlett Grant Project demonstrates, visual texts can occlude the processes of collaboration that underlie them at least as thoroughly as can print texts. The old word-processing acronym WYSIWYG (What You See Is What You Get) has taken on an ironic note. Even in the most transparent interface, even in the baldest visual representation, what you see, we argue, is never quite what you get. It is always a map, an echo, a combined presence and erasure—in short, a palimtext.
Theorizing the visual turn in general offers us only part of the story; another important part is what happens when the visual turn plays out in particular contexts. For example, Stephen Knadler's (2001) study of electronic portfolios at our own institution, Spelman College, argues that most literature about women composing on the Web overlooks compositions by Black women, and hence tends to construct women's Web-based self-representations as a tacitly white female author. Therefore, we turn now to the context of our own study, the Bush-Hewlett Grant Project at Spelman College, with these questions in mind: How and why did this project ask its participants to collaborate? In what ways did the participants carry out their collaborative (and non-collaborative work)? What can we take from our study that will be of assistance to other teachers working with collaborative projects in visual space?
Popular lore among writing teachers has it that "our students" know more about "technology" than "we" do. The image associated with this claim is that of the Net-savvy, instant-messaging student, an ultra-cool but unspecific creature like the dancing silhouettes featured in a current iPod television advertisement. This student sends text messages under the table during class by feel, carries her flash drive on her keychain, and (to take our collective fantasy to its dystopic extreme) not only has ready access to essays on the Web, but doesn't really understand why downloading chunks of them is plagiarism. By contrast, we, the imagined group of professors, are older, stuffier, and always a few steps behind the technological times. We not only remember when people used answering machines; we remember what it was like to dial a number, listen to eight or ten rings, and then hang up, saying idly, "I guess nobody's home."
Of course, this lore misrepresents students, ourselves, and technology itself. Not all of our students are as media-savvy as we imagine (or fear) they are. Identity categories such as race, class and gender affect how eagerly students adopt and exploit digital technologies (Edwards, 2005; Hawisher & Sullivan, 1998; Knadler, 2001). Moreover, although students may be familiar with various new technologies, their ability to read and exploit these technologies critically requires as much care, as much "paying attention" (Selfe 1999), as any archived photograph or Shakespeare folio. We ourselves are not necessarily the old duffers we fear we are; in fact, our access to digital technologies and to the Internet, because of our more privileged institutional positions, is often better than our students'. And finally, we must remember that "technology" itself does not necessarily mean "digital technology."
Along with the fearful lore about the schisms created by technology between teachers and students comes optimistic lore about ways that new technologies will refigure our classrooms, our curricula, and in general solve many of our most intractable pedagogical and administrative difficulties as writing teachers (Hawisher & Selfe, 1997, p. 306). Collaboration was one of the processes that technology was supposed to invigorate. In their essay analyzing the use of "computers, video, and telecommunications technologies" to implement the AAHE's seven principles of good practice, Arthur Chickering and Stephen Ehrmann (1996) argue that "the extent to which computer-based tools encourage spontaneous student collaboration was one of the earliest surprises about computers." However, as our own experience with the Bush-Hewlett Project demonstrates, "computer-based tools" may be primarily verbal, primarily visual, or a mixture; and the shift from verbal to visual media may be accompanied by unexpected challenges to pedagogy.
From the beginning, the Bush-Hewlett grant proposal was a collaborative undertaking. In the spring of 2002, a small group of faculty and staff in Spelman College's Comprehensive Writing Program and English Department shaped a proposal titled "Building Virtual Learning Communities" with input from faculty across the disciplines, including art, sociology, music, mathematics, history, and political science. The finished proposal's broad goals were to blend writing, research, and technology in pedagogy; to de-center the classroom, hence fostering student engagement; to monitor student responses to the "real audience" implicated in Web publication; and, most important, to ask students to develop their projects through collaboration with each other, with faculty and staff, and potentially with a larger community.
Each semester since January 2003, this grant has funded selected proposals from faculty for digital-based projects that are implemented through individual courses. Proposals can come from one faculty member teaching a single course or from pairs/groups of faculty teaching a range of courses. Each project culminates in the production of a Web site composed by students in the class(es). Faculty fellows and students are supported by grant faculty and staff, including instructional-technology coordinator Daniele Bascelli; by external consultants; and by trained student assistants. Fellows receive a one-week orientation at the project's outset and, along with their students, attend regular workshops on topics such as Web design; use of Web tools including Dreamweaver and iMovie; and intellectual property and fair use on the Web. Bush-Hewlett project topics have varied. Examples include "Environmental Politics and Policies," "The Black Presence in American Dance," and "Disability Discourses."
The implementation of this project has included various challenges and unexpected turns. Collaborative learning, as Harvey Wiener (1986) notes, "is messier in practice than in theory" (p. 60). Among the most surprising challenges has been the difficulty faced by project fellows (faculty members) and students as they attempted to engage in collaborative composition within the visual medium of the Web. The considerable disconnect between the ease of e-mail and the rigor of Dreamweaver was not on our horizon when the project began. Similarly, we were ready to echo Chickering and Ehrmann's (1996) exhortation for "team effort" and "apprentice-like activities" without full comprehension of the groundwork that must be laid before such phenomena become probable, or even possible. Communication among students and faculty requires nurture and had to be far more structured than was anticipated. The complex sequence of demands for Web site construction, including the need for clean design, meaningful images, accessibility, and effective editing—in short, demands stemming from the unfamiliar territory of visual literacy—made each project more difficult than anticipated. Moreover, the Web's public audience made these features dramatically important to faculty, but not always to students. The differences in the way these constituencies en-visioned their projects and audiences made results less predictable.
The visual medium of the Web, and the collaborative nature of the projects, are two important factors that enabled the variety and richness of project topics and approaches. However, this variety posed problems for project staff, since the training in visual literacy and media had to change constantly in order to meet the evolving needs of specific classes. The three brief case studies in the next section provide three views of the way that "collaboration" played out in the construction of their respective Web sites. Through this analysis, we hope to offer other teachers who assign visual compositions insight into the problems of creating such assignments, as well as the promise that may be achieved when we understand our students' potential literacies to be visual as well as verbal.
To take a closer look at the dynamics of collaboration in the Bush-Hewlett projects, we selected three of its Web sites: a Web-based gallery of paintings by iona rozeal brown, with student commentary in English and Japanese ("The Afro-Asiatic Allegory"); an online poetry journal ("L-I-N-K-E-D"); and a virtual museum combined with video interviews of visiting artists ("Issues in Women's Art"). A primary criterion for selection of these sites is that each is self-consciously visual, so that a desire for sophisticated and accessible design figured in the development of each. Unlike other Bush-Hewlett project sites, which tend to be more like what Yancey (2004a) calls "print uploaded," these three sites are highly "Web sensible" (p. 745). Yancey's distinction is made in a discussion of electronic portfolios, but is useful for the Bush-Hewlett web sites as well:
[A Web-sensible] model may include print texts, but it will include as well images and visuals, internal links from one text to another, external links that provide multiple contexts, and commentary and connections to the world outside the immediate portfolio. (p. 746)
Each of the three sites in this study, then, uses the Web not simply as a platform for distribution, but as a medium to be exploited. A secondary consideration was that the production of the three sites involved a wide range of levels and types of collaboration.
The first site, "The Afro-Asiatic Allegory," presents a gallery of paintings by iona rozeal brown which explore ganguro, young Japanese women who darken their skin, dress in styles associated with hip-hop culture, and adopt conventionally African American hairstyles as a means to express their own cultural rebellion. While she was in Japan for several years, brown began a sequence of paintings which conflated these ganguro figures with the genre of seventeenth-century Japanese woodcuts of geishas. The Web site therefore presents a critique by an African American artist who uses a traditional Japanese artistic style to represent the ganguro, who are themselves representations of hip-hop culture. Male and female students from the Atlanta University Center, enrolled in Advanced Japanese, in turn commented on brown's paintings by writing responses in English and Japanese, which accompany the digital images of brown's paintings.
According to surveys
and interviews conducted after the course ended, "The Afro-Asiatic Allegory"
inspired Liu to expand her cultural-studies approach to teaching language, and
increased students' awareness of the complexities of translation; however,
student/student collaboration was at a minimum. For example, Dr. Barnwell
contributed the professional design the museum had commissioned for advertising
the exhibit to the Web site. Students viewed the artist's materials and worked
with articles, art history, and an illustrated study of ganguro culture,
before producing their commentaries, which were selected by faculty and
returned to students for editing and translation. This group of selected
students worked with Liu to develop accurate translations of their informal
writing, including hip-hop terminology, into Japanese. Liu assessed the
students' increase in their level of interest and presented this
self-evaluation in her end-of-the-year presentation. In sum, collaboration
between Liu and her students could be seen as high, but the process of
site-building was very much a faculty-centered, -directed, and
The irony of this project is that, while it entailed the least student/student collaboration and maintained the most traditional pedagogy, its culminating Web site features an elegant and sophisticated design, as well as an array of perspectives, languages, and creative critiques. Its presence serves as an emblem of the difficulty faced by Bush-Hewlett staff and faculty fellows. We want elegance of design and intellectual perspective, a self-conscious yet polished rendering of the palimpsest in a blend of contemporary cultures and traditional forms. However, such elegance may come at the price of eliminating, or at least refiguring, student/student collaboration.
By contrast, the online literary journal, "L-I-N-K-E-D," proposed by faculty fellows Sharan Strange and Opal Moore, both English professors, went through multiple iterations and was developed by an extensive network of student groups performing a variety of tasks. Including interviews, poetry, hypertext commentaries on poems, fiction, and videotaped readings, "L-I-N-K-E-D" exemplifies some of the rich opportunities available through digital and visual composition. Most of Strange's creative writing classes and several of Moore's over the last two years have contributed to the project. Strange arranged teams for Web design and development, editorial standards and selection, interviews or videotaping, and releases and permissions. Each group received training and information from Bush-Hewlett staff and had support available from student assistants. Student commentaries from the first year of collaboration on the Web site indicated some real pleasure in their acquisition of new skills and power to make decisions. However, these comments also indicate a good deal of displeasure in being required to participate in the site's construction as a component of their coursework, "extra" work for which they did not always see the relevance.
In its first iteration, the site had a somewhat industrial look, with stark images and large white spaces. Strange guided the next semester's Web-design team toward a softer style, with student-produced photographs of landscapes and people. After these two iterations, the two classes left behind a great deal of collected material—videotapes of interviews, audiotapes, submitted creative work—in digital folders without clear identification. Strange paid a student assistant to identify and organize these files. For the third semester of work on "L-I-N-K-E-D," Strange devised a much more structured framework of collaborative assignments. Students were assigned a sequence of activities: first, to select a theme for the assignments to come; second, to determine an agreed-upon set of standards for material; third, to post their own work and receive feedback from their peers. Class discussion followed these postings as well. A significant portion of the students resisted giving and receiving peer feedback, insisting on a response from the instructor or delaying their submissions to their peers. In a not unusual lull during the classroom workshop, one student asked the instructor, "Well, what do you want from us?" In this case also, the assumption that student engagement will increase with collaborative projects did not hold true.
"L-I-N-K-E-D," like "The Afro-Asiatic Allegory," challenges the simple descriptor "collaborative." A few students with special interest in particular skills (such as using Web editors or creating video files) led their respective groups or, in some cases, accomplished the work of the group single-handedly. Other students resisted, failed to keep appointments, and proved unprepared to organize and achieve their tasks. Strange and Director Warner arranged several meetings with editorial groups, but attendance was poor. Early expectations that students would take the technology and run with it were not realized. Nevertheless, the Web site ultimately became a coherent, well-designed composition for two reasons: first, Strange continually encouraged students to engage in projects that involved visual elements, and then to incorporate these projects into the site; and second, the project unfolded over three semesters, allowing the site to go through several iterations.
Now, after the third class's participation, "L-I-N-K-E-D" is a successful online literary journal, with work by student contributors as well as by established poets and other creative artists. Accessible from the English Department's home page as well as the Bush-Hewlett Web site, it includes a space for reader submissions. The visual presentation of "L-I-N-K-E-D" indicates that it was created collaboratively and continues to value collaboration through its dynamic, audience-participatory form. However, continuity from course to course is sustained, not by the students who take part in the project, but by Strange.
In an interview, Strange (2005) offered several insights stemming from her experience: First, the one-semester structure of a traditional course is a mismatch for the intricate process of collaboratively developing a Web site, at least by students who are just learning new technologies and visual literacies. Second, students tend to compartmentalize their work and their disciplines; therefore, most students enrolled in her poetry-writing workshop were not "thinking technology." Third, student dynamics vary greatly from class to class; the small, student-run editorial groups which worked successfully in her introductory creative writing class were much less successful with advanced students. To her insights, we would add a fourth: that the collaborative nature of "L-I-N-K-E-D"" is paradoxical. It continues to exist, in its collaborative form, because it possesses a single primary director.
The Web site for "Issues in Women's Art" was created during the Fall 2004 and Spring 2005 semesters by two classes (also titled "Issues in Women's Art") taught by Professor Akua McDaniel. Constructed as a video tour, the site invites the audience into Spelman College, with a virtual visit to the Cosby Academic Center and thence into the Museum of Fine Art. A central photograph from a 1996 exhibit, "Bearing Witness," uses Flash to show a range of contributing artists appearing one by one along a staircase in the Cosby atrium (see Figure 1). Future classes of McDaniel's will expand the site further; at present, it features biographies of and interviews with three African American women artists, and examples of their work superimposed on a streaming video of the museum's interior (see Figure 2).
Figure 1. Contributing artists, rendered in a Flash image
Figure 3. The Web site for "Issues in Women's Art"
Because McDaniel's project was carried out two years after the inception of the Bush-Hewlett Project, staff and faculty had already revised their methods of support several times. Support included regular Friday workshops aimed at specific skills, such as Web design and editing; classroom visits to help students understand the process of Web-based research and intellectual property; and meetings with student editorial groups. McDaniel set aside each Friday meeting of her Monday-Wednesday-Friday class schedule for work on the Web site, so that students developed a pattern of participation. This structure allowed students to function within the course structure, but to also develop a coherent focus on the project and cohesion within the group. As Director Warner met with members of the editorial group outside of class time, she found the participants on task and committed to producing effective biographies. The editorial group had taken all submitted student essays, selected the best, and combined those into a short, dense, well-documented piece. The interviewing teams also functioned well, each presenting a well-researched set of questions to the three visiting artists. The regular meeting time for groups, especially when compatible with the course schedules and deadlines, strengthened group participation.
Immediately in the Fall semester, McDaniel identified a student in the class who had the skills and interest to pursue a distinctive presentation. He participated as a for-credit student during the first semester and was retained as a paid assistant during the second. On his own, he developed the site design and then submitted it for suggestions and review by the class. The end result is a beautiful, well-designed, and inviting Web site. During a presentation of the site to Spelman faculty, McDaniel pointed out that she had contributed very little to the site's design. However, despite the highly collaborative processes that underlie its content, the site's concept and design had the benefit of a single shaper.
The foregoing analysis tempts us to offer a pithy and pessimistic conclusion, to wit: "The better a visual composition looks, the less collaboration it involved." However, we resist this conclusion, both because we are inclined more toward optimism and because it doesn't tell the whole story. Indeed, it is the urge to define "collaboration" as a single thing, or to assume that the outcome of collaboration is a finished composition, that has led to problems with collaborative pedagogies in the past. In this final section, then, we wish to offer a collection of insights drawn from our analysis, along with suggestions for pedagogies which engage students in collaborative visual composition.
Collaborative work is often credited with fostering student engagement, but in our study, the collaborative Web-building projects seemed at least as likely to discourage it. This may have been due in part to the students' unfamiliarity with visual literacies and Web-based technologies. It may also have been due in part to the study's location: attended almost exclusively by young, high-achieving Black women, Spelman College can be a culture that fosters competition more readily than collaboration. And finally, it should be remembered that engagement is a slippery concept and may be perceived differently by different project participants. For example, despite the perception by Strange and Bush-Hewlett staff that one of her classes was especially resistant, students in that class (like almost all students participating in the Bush-Hewlett project) reported an increase in their own sense of engagement with the class as a result of the Web-building project.
While acknowledging these complicating factors, we suggest that the two most significant factors affecting student engagement in this study are authorship and audience. For each of the three Web sites studied, the patterns of authorship that emerged profoundly affected the project's process and its outcome. In "The Afro-Asiatic Allegory," for example, control of the project was generally hierarchical:the museum director and Bush-Hewlett staff took responsibility for the site's design, and Liu for its written content. Spaces provided for student work, such as their writing of commentaries on paintings which they then translated into Japanese, were carefully overseen and circumscribed. As a result, the final product, while visually compelling, cannot be said to have been "authored" by students except in a limited way. "Issues in Women's Art," by contrast, is almost entirely student-authored, from design to content. The works of art displayed were created by visiting artists, but their selection and arrangement was entirely student-driven. Similarly, the interviews with artists, which at present make up a large part of the site's content, are shaped by student-written question protocols. However, the design of "Issues in Women's Art" was not collaboratively authored, except in the qualified sense that the single student designer presented his design to his classmates for feedback. Like "The Afro-Asiatic Allegory," "Issues in Women's Art" seemed to increase student engagement by establishing a structure within which they could work. Collaborative, in these cases, did not mean undirected; it meant selectively directed, or what we might instead call managed.
"L-I-N-K-E-D" was the most collaborative of the three projects, in the sense that the developmental pattern of its design, content, and execution was more horizontal than for the other two sites. In other words, despite the large amount of work that Strange put into (and continues to put into) the site, students had more shared responsibility as authors, both in visual and verbal modes, than did the students working on "The Afro-Asiatic Allegory" and "Issues in Women's Art." Notably, "L-I-N-K-E-D" was also the project that involved the most vocal resistance from students, at least in the observation of Bush-Hewlett staff. Its process of completion was, to return to Wiener's (1986) term, messy. We believe that students' frustration stemmed in part from the fact that, by Professor's Strange's pedagogical choice, they did not receive explicit instructions on what the site should look like and include. In short, they were being asked to take on more of the task of management than were Liu's or McDaniel's students.
What does this mean for teachers working with students on collaborative visual projects? An important point to realize, we believe, is that composing (not consuming or editing) in visual spaces may be new to students. Although students are accustomed to viewing visual compositions, they are less accustomed to think of themselves as authors, rather than consumers, of them. Moreover, some projects that appear to call for student authorship of e-spaces—such as highly prescriptive e-Portfolio templates—in fact operate more like fill-in-the-blank forms than opportunities for authorship. Strange's point that students may "compartmentalize" knowledge in certain classes or disciplines (assuming, for instance, that digital technologies don't belong in a poetry class) is salient here, since this compartmentalization may also occur between genres. In what ways might an essay be visual as well as verbal? Can we apply the verb "writing" to the creation of a Web site, or do we need to revise our lexicon? These questions must not be skipped over, but engaged with students. Offering models, asking for student responses to these basic but crucial questions, and being explicit about the meaning of collaborative visual composing—whether the working definition is developed by the teacher, by the students, or both—may be helpful strategies for achieving a productive, rather than discouraging, level of defamiliarization.
Moreover, teachers should examine the assumption that composing a Web site involves a "real" audience and hence will be more exciting to students than composing, say, a word-processed paper. How "real" is this "real" audience of Web surfers? Isn't it likely that we as teachers will care more about the visitors to educationally-based sites than our students will? And if that is true, then what audiences do our students care about, perceive as "real"? Again, this question should be asked directly of students. Authorship and audience tend to work together: if a writer cares about her audience, she is more likely to feel and behave like someone with author-ity. Not every teacher may choose to have students determine the audience of their collaborative Web-based compositions, but neither should she assume that she and her students understand a "real" or "important" audience in the same way.
Returning to Weiner's (1986) point that the form and success of collaboration in the classroom depend greatly upon the teacher's methods of managing that collaboration, we believe that teachers who ask their students to engage in collaborative work on a single document—especially one composed in visual media, and hence involving less conventional tools and literacies—should define the students themselves as a "real" audience for the composition. For example, over the course of one Bush-Hewlett class, the teacher could take responsibility for showing the evolving Web site to the class as a whole and asking them to construct a list of objectives for the next period of the project (e.g. the next two weeks or month), until the next viewing. If a visual project continues to evolve over a period of years, as some Bush-Hewlett Web sites do, former student authors should be invited to view the changing site and offer their suggestions to current authors. One reason for low student engagement in some projects, we believe, is that they themselves were not considered a primary audience for the composition.
Coherence is a familiar value among teachers of academic writing. We've all heard—and most of us have delivered—that classic piece of advice: "Establish a thesis and stick to it!" Scholars including Fox (1994) have pointed out that this is generally a Western, and particularly a U.S. value, but that doesn't make it any less important for the student being urged to write essays that demonstrate "unity." However, when a project is undertaken collaboratively, achieving coherence becomes more difficult. This complexity increases again in the dynamic visual medium of the Web, where even a single-authored, expertly executed Web site may lose coherence due to factors beyond the author's control as the site appears in diverse browsers and processing systems.
Faculty leading Bush-Hewlett projects tended to value coherence highly, an issue that returns us again to the question of audience. No doubt imagining audiences ranging from their own faculty peers at Spelman College to their colleagues across the nation and in other countries, faculty stakes in making the sites (literally) "presentable" were high. However, students' stakes in coherence were much lower, and the appearance of the un/finished sites reflects this. Having studied the emergence of the three sites described in this paper, we suggest that teachers of writing should examine their desire for coherent work from students—ideally, in any context, but especially when assigning collaborative Web-based projects. Part of the value of a Web site is its potential for "extending ever outward" (Yancey, 2004a, p. 751). In order to realize its potential, collaboration in Web space must move beyond the boundaries of a single class or a single semester. Instead, we must be willing to understand collaboration as a larger process, one that infuses the culture of a university and is reinforced at every level (Bruffee, 2003).
This is not news to teachers of writing. Although few of us would fail to acknowledge the importance of polished writing for some purposes, our profession has, by and large, accepted the notion that attention to process—the series of decisions, conversations, efforts, and revisions that underlies a composition—is beneficial for our students' writing. However, visual media present a new challenge precisely because they are new, and therefore have a heightened ability to dazzle. Web sites (and other visual compositions, such as videos or PowerPoint presentations) are palimpsests, no less than paper-and-ink essays. Yet our unfamiliarity with these media may make it difficult for us to perceive the palimpsest which under/overwrites the final composition. For example, what is a "draft" of a video or PowerPoint presentation? How should feedback on such compositions be delivered, received, and perhaps recorded? Our challenge is to find ways to help our students render the layers visible, so that we can offer them guidance as detailed and complex as their processes of composing warrant. This may mean letting go of values such as "coherence," and inviting the potential benefits of mess.
The suggestions we have offered so far have a common theme: as teachers of collaborative visual projects, we must articulate for ourselves our purposes in assigning such projects, and constantly re-check those purposes against projects' trajectories and outcomes. Both collaboration and visual media are elements of pedagogy which are tempting to add on to a course like a garnish. However, neither of these elements alone can simply be appended to a course without re-shaping the purposes of that course (whether implicitly or explicitly), and collaboration in visual media presents yet another set of challenges. As the experiences of the teachers in the Bush-Hewlett Project indicate, our students will alert us—and quickly—if our collaborative assignments seem poorly conceived, unpredictably managed, or lacking in purpose. But in fact, this problem may be due not to poor planning (since each Bush-Hewlett project was meticulously planned) but instead due to the need for ongoing renegotiation of plans in response to evolving circumstances. This conclusion leads us to an important point about collaboration in visual space. If we are willing to understand collaboration, not as a single type of activity, but as a proliferation of activities as diverse as the contexts in which they occur, then working with compositions in visual media will enrich that variety and increase possibilities. In other words, planning—and again, here, we would suggest a more flexible term, such as managing—should take place only before a project's inception, but also during the project, to enhance engagement and outcomes.
Twenty years ago, Kenneth Bruffee (1984) called writing teachers to recognize the profound implications of fostering collaboration in our classrooms:
Organizing collaborative learning effectively requires doing more than throwing students together with their peers with little or no guidance or preparation. To do that is merely to perpetuate, perhaps even aggravate, the many possible negative efforts of peer group influence: conformity, anti-intellectualism, intimidation, and leveling-down of quality. To avoid these pitfalls ... requires us to create and maintain a demanding academic environment that makes collaboration—social engagement in intellectual pursuits—a genuine part of students' educational development. And that in turn requires quite new and perhaps more thorough analyses of the elements of our field than we have yet attempted. (p. 652)
Although Bruffee remains vague on just what "analyses" might result from continued exploration of collaborative work, subsequent studies, including ours, show that among the key concepts challenged by collaborative projects include author, audience, and coherence. Collaboration in visual space brings yet more new questions to the table. As scholars, we are accustomed to a mess like this one: ideas evolve, questions go unanswered, knowledge is revised. As teachers, we are perhaps less accepting of mess, especially when our students are making messes (or what we perceive as messes) in public spaces such as the Web. However, allowing mess is a necessary part of the process of working collaboratively, at least as important in visual space as in any other modality. It is only through such mess—or, as Bruffee (2003) more gracefully puts it, "diverse experience [and] consistent cultivation" (p. 20)—that we can develop and revise pedagogies which serve both students' and teachers' aims.
Bakhtin, M.M. (1986). Speech genres and other late essays. Trans. Vern W. McGee. Ed. Caryl Emerson and Michael Holquist. Austin, TX: University of Texas Press.
Baldwin, Beth. (1996). Evolving past the essay-a-saurus. RhetNet: A Cyberjournal for Rhetoric and Writing. Retrieved May 8, 2005, from http://www.missouri.edu/~rhetnet
Ball, Cheryl E. & Kristin L. Arola. (2004). 9 Visual Exercises. CD-ROM. New York: Bedford/St. Martin's.
Bartholomae, David & Anthony Petrosky. (2003). Ways of Reading Words and Images. New York: Bedford/St. Martin's.
Beach, Richard. (1986). Demonstrating techniques for assessing writing in the writing conference. College Composition and Communication, 37, 56-65.
Bowden, Darsie. (1996). Stolen voices: Plagiarism and authentic voice. Composition Studies, 24, 5-18.
Bruffee, Kenneth A. (1978). The Brooklyn plan: Attaining intellectual growth through peer-group tutoring. Liberal Education, 64, 447-69.
Bruffee, Kenneth A. (1984). Collaborative learning and the "conversation of mankind." College English, 46, 635-652.
Bruffee, Kenneth A. (1993). Collaborative learning: Higher education, interdependence, and the authority of knowledge. Baltimore: Johns Hopkins UP.
Bruffee, Kenneth. (2003). Cultivating the craft of interdependence: Collaborative learning and the college curriculum. About Campus, 7, 17-23.
Building virtual learning communities: The Spelman Bush-Hewlett Grant. (2005). Retrieved February 16, 2005, from http://www.spelman.edu/bush-hewlett.
Chickering, Arthur W. & Stephen C. Ehrmann. (1996). Implementing the seven principles: Technology as lever. AAHE Bulletin. Retrieved May 23, 2005, from http://www.tltgroup.org/programs/seven.html
Davies, Bronwyn. (1993). Shards of glass: Children reading and writing beyond gendered identities. Sydney, Allen and Unwin, and New Jersey, Hampton Press.
Dunn, Patricia & Kathleen Dunn DeMers. (2002). Reversing notions of disability and accommodation: Embracing universal design in writing pedagogy and web space. Kairos, 7. Retrieved February 3, 2003, from http://english.ttu.edu/kairos/7.1/binder2.html?coverweb/dunn_demers/index.html.
Ede, Lisa & Andrea Lunsford. (1990). Singular texts/plural authors: Perspectives on collaborative learning. Carbondale, IL: Southern Illinois UP.
Ede, Lisa & Andrea Lunsford. (2001). Collaboration and concepts of authorship. PMLA, 116, 354-369.
Edwards, Mike. (2005). Vitia: Faults, sins, abuses. Retrieved March 30, 2005, from http://www.vitia.org.
Faigley, Lester, Diana George, Anna Palchik, Cynthia Selfe. (2004). Picturing Texts. New York: Norton.
Fox, Helen. (1994). Listening to the World: Cultural Issues in Academic Writing. Urbana: IL, NCTE.
Harris, Joseph. (1989). The idea of community in the study of writing. College Composition and Communication, 40, 11-22.
Hawisher, Gail E. & Cynthia L. Selfe. (1997). Wedding the technologies of writing portfolios and computers: The challenges of electronic classrooms. In Kathleen Blake Yancey and Irwin Weiser (Eds.), Situating portfolios: Four perspectives (pp. 305-321). Logan: Utah State UP.
Hawisher, Gail E. & Cynthia L. Selfe (Eds.). (1999). Passions, pedagogies, and 21st century technologies. Urbana, IL: NCTE & Logan, Utah: Utah State UP.
Hawisher, Gail E. & Patricia Sullivan. (1998). Women on the networks: Searching for e-spaces of their own. In Susan C. Jarratt and Lynn Worsham (Eds.), Feminism and composition studies: In other words (pp. 172-197). New York: MLA.
Herrington, Anne & Marcia Curtis. (2000). Persons in process: Four stories of writing and personal development in college. Urbana, IL: NCTE.
Kress, Gunther. (1999). "English" at the crossroads: Rethinking curricula of communication in the context of the turn to the visual. In Hawisher & Selfe (pp. 66-88).
Knadler, Stephen. (2001). E-Racing difference in e-space: Black female subjectivity and the web-based portfolio. Computers and Composition, 18, 235-255.
Lunsford, Andrea A. & Lisa Ede. Collaborative authorship and the teaching of writing. In Martha Woodmansee and Peter Jaszi (Eds.), The construction of authorship: Textual appropriation in law and literature (pp. 417-438). Durham and London: Duke UP.
McQuade, Donald & Christine McQuade. (2000). Seeing & Writing. New York: Bedford/St. Martin's.
McQuade, Donald & Christine McQuade. (2003). Seeing & Writing 2. New York: Bedford/St. Martin's.
Merriam, Sharan B. (1998). Qualitative research and case study applications in education. San Francisco: Jossey-Bass.
Naimool, Karen. (2005, April 26). [Personal interview.]
National Center for Technology Innovation. (2005). Universal design. Retrieved June 1, 2005, from http://nationaltechcenter.org/resources/dl/default.asp.
Salvo, Michael. (2002). Deafened to their demands: An ethnographic study of accommodation. Kairos, 7. Retrieved February 3, 2003, from http://english.ttu.edu/kairos/7.1/binder2.html?coverweb/salvo/map.htm.
Strange, Sharan. (2005, April 19). [Personal interview.]
Tompkins, Jane. (1990). Pedagogy of the distressed. College English, 52, 653-660.
Trimbur, John. (1989). Consensus and difference in collaborative learning. College English, 51, 602-616.
Vielstimmig, Myka. (1999). Petals on a wet black bough: Textuality, collaboration and the new essay. In Hawisher & Selfe (pp. 89-114).
Wiener, Harvey S. (1986). Collaborative learning in the classroom: A guide to evaluation. College English, 48, 52-61.
Yancey, Kathleen Blake. (2004). Postmodernism, palimpsest, and portfolios: Theoretical issues in the representation of student work. College Composition and Communication, 55, 738-761.
Yancey, Kathleen Blake. (2004). Made not only in words: Composition in a new key. College Composition and Communication, 56, 297-328.
 Influenced in part by our own argument that collaboration must be understood in complex and various ways, we have structured this collaboratively-written paper as an amalgam of unified and individual voices. Inspired by Vielstimmig (1999) and Herrington and Curtis (2000), we have attempted to make more visible—although imperfectly so—the way that collaborative writing toggles between singularity and plurality. Most of this article represents our collaborative authorial voice. Two sections, however, represent our individual voices as we speak from the respective standpoints of Bush-Hewlett project director (Warner) and Bush-Hewlett teacher and fellow (Price). Individually-authored sections are preceded by our names.
 We wish to thank the many people who have participated in this article's development: all Bush-Hewlett staff and student assistants, including Dan Bascelli, Bonnie Tidwell, Mathew Weed, Maya Forester, and Karen Naimool; participating Bush-Hewlett faculty and fellows, including Andrea Barnwell, Xuexin Liu, Sharan Strange, Opal Moore, and Akua McDaniel; and Bill Condon and iona rozeal brown.
 For a more detailed description of the project, see the section "The Bush-Hewlett Project: Promise and Problems."
 This article is limited to the primary concern of composition in visual spaces. However, we strongly advocate teaching in multiple modes and approaches—aural, tactile, conversational, musical, movement-based, and so on. For more on multi-modal teaching and accessible instructional design, see Dunn & DeMers (2002); Salvo (2002); and the Web site of the National Center for Technology Innovation ("Universal Design").
 The Atlanta University Center (AUC) is a consortium of schools made up of Spelman College, Morehouse College, Clark Atlanta University, the Interdenominational Theological Center, and the Morehouse School of Medicine.
 Museum Director Dr. Andrea Barnwell and Foreign Languages Professor Xuexin Liu worked with students and staff to create this Web site.
 This assertion is based on exit surveys completed by students in WebCT at the end of each course. It should be noted that surveys were usually completed before final grades were submitted, and hence may not be as reliable as data collected after grades were recorded, such as interviews.
Margaret Price is an assistant professor at Spelman College, where she teaches composition, rhetoric, and creative nonfiction. Her work has appeared in publications including College Composition and Communication; Teaching, Writing, Disabilities: A Professional Resource Book; Creative Nonfiction; the Michigan Quarterly Review; and Bitch: Feminist Response to Pop Culture. She is currently at work on a book, Writing From Normal: Critical Thinking and Disability in the Classroom.
Anne Bradford Warner is associate professor, director of the Comprehensive Writing Program, and director of the Bush-Hewlett grant "Building Virtual Learning Communities." At Spelman for almost 25 years, she served for seven years as chair of the English Department. A scholar of nineteenth-century U. S. literature, especially the slave narrative, Professor Warner has presented many papers and published articles on Harriet Jacobs's Incidents in the Life of a Slave Girl, in addition to other work on U.S. autobiography and poetry.