Is There a Hypertext in this Class?
Teaching Victorian Literature in the Electronic Age


Jonathan Smith

Humanities Department
University of Michigan-Dearborn
Dearborn, MI 48128-1491


The essay describes and analyzes the use of hypertexts, in particular George P. Landow's The Dickens Web and The In Memoriam Web, in two undergraduate Victorian literature classes. Hypertext is shown to encourage active student engagement, especially with contextual material; to lead to more focused research topics; and to facilitate student collaboration. The potential of hypertext is best realized, however, when it is extensively integrated into a course. Focus is thus given to two practical questions: 1) How must classroom management and writing assignments be reconceptualized? 2) How are students to be taught to read hypertextually? Landow's claim about the ease with which the latter occurs is questioned.


In Hypertext: The Convergence of Contemporary Critical Theory and Technology, George P. Landow argues that the use of hypertexts in teaching has the potential to revolutionize what occurs both inside and outside the classroom. Like other proponents of the virtues of electronic reading and writing such as Jay David Bolter and Richard Lanham, Landow contends that hypertexts "intrinsically promote a new kind of academic freedom and empowerment" for students (177). Indeed, says Landow, the scariest feature of hypertext for teachers may be the way it actually "answers [their] . . . prayers for active, independent-minded students who take more responsibility for their education and are not afraid to challenge and disagree" (163).

Landow's claims for hypermedia are not merely theoretical but are based on the development and use of hypertexts in his own teaching at Brown University. Two of these hypertexts, The Dickens Web and The In Memoriam Web, are available commercially from Eastgate Systems, and a related web, The Victorian Web, is now accessible on the World Wide Web. Coupled with the presence of other Web sites devoted to nineteenth-century writers--including hypermedia archives for Dante Gabriel Rossetti, William Morris, and Robert Browning (see McGann, William Morris Society, Everett)--it is now possible for teachers of Victorian literature to incorporate such resources into their courses, or even to build their courses around them, adapting existing materials rather than investing in the time-consuming process of developing them from scratch. Since these Web sites explicitly present themselves as provisional and fluid, utilizing them can reinforce a course's hypertextual nature and provide models for the collaborative generation of interrelated knowledge about, and interpretation of, works of nineteenth-century literature.

This paper will describe and critique my own use of hypertextual resources in two upper-level undergraduate courses on Victorian literature. One class was taught fairly traditionally, with hypertexts and the World Wide Web used outside of class for discussion questions and assignments. The other, however, was conducted in a computer classroom, and students were actively involved in modifying existing hypertexts and constructing new ones. Because I believe my experiences are generalizable to any course employing hypertext, I will focus my analysis on what seem to me to be two important practical difficulties. One is the need to reconceptualize classroom management and the nature of writing assignments if the potential of hypertext is to be exploited. The other is the problem of how to teach students to read hypertextually. "Empowerment" and "active, independent-minded" reading are certainly the promise of hypertext, but they do not occur automatically--they must be taught.

Description of Courses

I teach at the University of Michigan-Dearborn, a regional commuter campus whose primary mission is undergraduate education. Its student population, a significant percentage of which is non-traditional, is drawn almost exclusively from the middle-class suburbs of Detroit. I point this out because, for the time being at least, the success of computer instruction is heavily dependent on material and cultural factors that vary widely from institution to institution. Obviously, the ability even to consider using hypertext in a literature class hinges on the state of the institution's existing computer resources. But other factors, often of limited concern at elite residential universities like Landow's, can diminish the effectiveness of hypertextual instruction even where computing infrastructure and support make it possible. Students who have not attended high schools where computers are common, who do not have access to a computer where they live, and/or whose schedules leave little time for extended visits to the computer lab are at a disadvantage (see DeLoughry). The high level of computer anxiety for many literature students, moreover, can exacerbate these problems.

The two classes in which I used Landow's hypertexts and the World Wide Web were both upper-level classes of fifteen to twenty students, almost all of them junior or senior English majors. One class was a survey of Victorian poetry and non-fiction prose, the other an intensive study of two Victorian novels, Dickens's Great Expectations and Eliot's Middlemarch. Most of the students took the courses to satisfy a period requirement in nineteenth-century literature rather than to use electronic resources or to experience a new classroom format. 80% had some familiarity with, and access to, a personal computer, and 67% had used the campus's computer network, but only 20% had used e-mail and only 7% had accessed the World Wide Web. (This was during the 1995-96 academic year. These latter percentages have increased substantially in subsequent classes.)

The survey course was taught in a standard classroom. At the beginning of the semester, however, two class sessions were held in a computer classroom to demonstrate and allow hands-on practice with e-mail, Netscape, and The In Memoriam Web. These resources were then utilized throughout the course in several ways. I posted discussion questions for each week's reading assignments via e-mail to a class distribution list several days before the first class session of the week; students were required to post a certain number of replies over the course of the semester but received extra credit for additional postings. Whenever possible, these questions employed material from The In Memoriam Web, for which the campus obtained a multiple-user license. While this Web is built around Tennyson's poem, it also contains information about other poets and about Victorian culture generally, so on several occasions I assigned readings from the Web as supplementary contextual material. I also used The In Memoriam Web as the basis for the first paper assignment, providing students with several topics that allowed them to take advantage of the ability to navigate hypertextually through the poem. On the World Wide Web, the Rossetti archive was my principal resource. I employed it in our unit on the Pre-Raphaelites in the same ways that I had already used The In Memoriam Web: in discussion questions and as an option for a paper topic.

The course on Dickens and Eliot, on the other hand, was taught in a classroom equipped with about twenty-five PCs connected to the campus network. A ceiling-mounted projection device permitted my computer screen to be displayed at the front of the room. In this case, because the students would be writing as well as reading in the hypertext environment, The Dickens Web and Storyspace, the hypertext authoring system in which Landow's Webs are written, were installed not only on a file server but also on the hard drives of the individual PCs in the classroom and on most of the PCs in the campus's two primary computer labs. Students thus had ready access to the software both inside and outside of class; to avoid inadvertent deletions on the "common" copy on the server, however, they were encouraged to use the local hard drive versions until they were familiar and comfortable with the software.

Conducting class in a computer classroom permitted more ambitious and aggressive use of hypertext. During the first two weeks of the semester I introduced the students to e-mail and the campus network, the World Wide Web, Storyspace, and The Dickens Web itself. Much of this time was spent teaching them how to use Storyspace as an authoring tool to construct and modify hypertexts. Then we spent several weeks on Great Expectations, the novel around which The Dickens Web is constructed. I used the Web to stimulate classroom discussions and for short homework assignments. World Wide Web sites were used occasionally in similar ways. Instead of writing a traditional paper, the students either composed new material to add to the Web or expanded existing material, and they created hypertext links between their work and the rest of the Web. This project could, and in most cases did, involve some research, but the focus was on textual analysis and interpretation. In the second half of the course, the class discussed Middlemarch and then constructed its own Web for that novel. This writing project was longer than the first and had to involve research as well as textual analysis.

Description of Storyspace and Landow's Webs

The basic unit of the hypertext Webs constructed from Storyspace is called a "writing space." In what Storyspace refers to as its "map view," a writing space appears as a small box with a title bar across the top. (Other views display the hypertext in chart, outline, and tree map formats.) Writing spaces in turn can contain both "text spaces," in which actual writing and images may appear, and other writing spaces. In the latter case, the map view looks like collections of Chinese boxes:

Material in the web is connected through both "basic links" and "text links"; basic links connect related writing spaces while text links connect words or phrases in text spaces either to writing spaces or to other bits of text. The user employs a small "toolbar" of icons to create and edit the writing spaces, text spaces, and hypertext links.

The In Memoriam Web and The Dickens Web are Storyspace hypertexts and hence are read in the same ways. The material in both Webs ranges from biographical information to social and cultural history to literary criticism and literary history. It embodies traditional approaches--literary influences, themes, imagery, characterization, setting, etc.--as well as more contemporary ones--feminism, post-structuralism, cultural studies, etc. Some of the material provides factual information (a definition of bildungsroman, a description of Utilitarianism, a brief history of Puritanism in England) while other documents contain interpretive commentary (the use of the Prodigal Son motif, the depiction of female aggression, the characteristics of the Victorian long poem).

Webs constructed from Storyspace in other words, because they are based on the same basic hypertextual principles as the World Wide Web, are navigated in similar ways. Links enable the reader to move quickly between related parts of The Dickens Web or The In Memoriam Web by pointing and clicking on highlighted words or images. Storyspace Webs differ from the World Wide Web, however, in two important ways. They are smaller and self-contained--and hence can be viewed as wholes--and they are designed to be modified and expanded. Users can edit existing text, construct additional links, and incorporate completely new material that in turn can be linked to other parts of a Web.

Teaching Hypertextually

My experiences with these two courses suggest that hypertext can be incorporated into existing courses with little disruption. Using it more ambitiously, however, requires a considerable amount of reconceptualization of assignments, grading, and classroom management.

In my Victorian survey course, I was able simply to adapt my practices to the new medium. I use ungraded discussion questions in my other classes, for example, to encourage interactive reading and to identify and provide feedback to those students with weaker reading skills. The electronic format afforded the opportunity, in effect, to initiate discussion prior to class, so I tried to pose questions that could stimulate replies from the other students. Weaker students had the advantage of seeing how their more sophisticated colleagues read and thought, yet I could still provide personal feedback by replying to an individual rather than to the whole list. For contextual reading, I sometimes assigned parts of The In Memoriam Web rather than sections from Robin Gilmour's The Victorian Period: The Intellectual and Cultural Context of English Literature, 1830-1890, one of the course's required texts. On the first paper assignment, I provided students with three options based on material in The In Memoriam Web. Landow argues that In Memoriam, composed in a non-linear fashion and frequently re-arranged, is itself a proto-hypertext, so I encouraged the students to approach the poem on those terms. One of the paper options involved reading the Web's material on the various structural schemes proposed for the poem and then explaining what was "natural" about Tennyson's own division of In Memoriam into "nine natural groups." Another option invited the students to select one of the poem's image patterns and explain how its use changes over the course of the poem and how this is related to an overarching interpretation of the work. Since the text of the poem is part of The In Memoriam Web, readers can follow links from the list of citations for an image directly to the text of the sections in which each citation appears, and often to annotations and/or commentary on the section:

A third option asked them to explore the Web's material on the pastoral elegy and then argue to what extent Tennyson draws on or departs from that tradition. For the second paper, students had the option of writing on the significance for the depiction of gender in the differences among the three published versions and the oil painting of Dante Rossetti's "The Blessed Damozel," all of which were available in the Rossetti Archive; they were free to employ other information--biographical, critical, etc.--contained in the Archive as well.

Within the classroom, I attempted to emphasize the reading and thinking skills that hypertext foregrounds. Students struggle to make connections when reading in a print environment, so hypertext's atomization of information into seemingly random bits can add to students' difficulties. To give them more practice in drawing connections and exploring realtionships, I used a number of classroom activities and exercises. In most cases this again meant simply modifying what I already do or incorporating hypertext into it. To provide a model for hypertextual reading, during one of the early classes held in the computer classroom I had the students examine the material in The In Memoriam Web on realism and empiricism and on the Victorian crisis of faith, then asked them to explain what this material had to do with the poem. Although quicker at seeing the individual connections than at how the three fit together, they were ultimately able to suggest Tennyson's purpose in offering his experiences as both personal and representative, abstract but rooted in a real emotional crisis that could not be solved by an appeal to traditional arguments about faith. I conducted similar exercises throughout the course. For example, I asked the students to explain the relationship between Ruskin's Unto This Lastand the chapter on "The Nature of Gothic" from The Stones of Venice, and I put them in groups to describe the depiction of women in a selection of Christina Rossetti poems. In a discussion of Arnold's "Dover Beach," I provided the text of Wordsworth's "It is a Beauteous Evening," to which Arnold's poem alludes, and asked them to develop an explanation about the relationship between the two poems and about the effect of Arnold's allusion on our interpretation of "Dover Beach."

It was the course on the two novels, with its more extensive integration of hypertext, that posed the greater challenge for me in both design and implementation. As I constructed my syllabus and developed assignments, I encountered a number of basic problems that underscored how radically hypertext can alter teaching and learning. What should I call the major writing assignments? Were they "papers"? "research papers"? how many "pages" should they be? should I require the construction of links, and if so, how should they be graded? to what extent should I allow and encourage collaboration, and how would I assess individual contributions to a collaborative effort? Similar problems arose for particular class sessions. How could I use class discussions to model the hypertextual reading and thinking I wanted the students to display in their writing?

Ultimately I decided to call the two major writing assignments "Web Writing Projects." The students could work in pairs or groups on a topic as long as their contributions were separated into individual parts rather than collectively authored. The students exchanged and submitted printed copies of their work for the sake of simplicity and familiarity. This enabled me to set page requirements while still enabling them to use Storyspace's "import" function to transfer their files directly into the Webs. Despite the reliance on such traditional trappings of the literature paper as page requirements and works cited lists, I encouraged them to approach their work hypertextually and to present it in whatever way they thought would be most effective in the Web. They could, if they chose, divide their project into multiple text spaces. For example, they could place historical information in one space and critical analysis in another, or lengthy passages from various critics and their response to those passages in separate spaces. For the first project, I required them to construct both basic links and text links to connect their material to the existing parts of the The Dickens Web, and to explain what a future reader of the Web would gain, in terms of understanding Great Expectations, from following those links. Such explanations would not, of course, become part of the Web, but I made it clear that the grade for the overall project would be based in part on the rationale for the links.

The project for The Middlemarch Web had similar requirements, but since this Web was constructed from scratch, there were several differences. The class members first discussed what this new Web should contain, agreeing on its basic parameters. With no existing information on Eliot to guide them, they were more dependent on their peers, and they had to learn about the nature of other people's projects to determine what their own project could and should be linked to. For the group of students interested in Eliot's life and the novel's depiction of women--her relationship with Lewes, her views on "the Woman Question," the assessments of Middlemarch by modern feminist critics, etc.--such interaction was especially fruitful and the identification of potential links easy. Some students, however, like one working on Eliot's chapter epigraphs, had to remain content with constructing links within their own material.

Since the students in this class were expected to write as well as read in a hypertext environment, I devoted much more attention to the Webs early in the term than I had in the survey course. In place of discussion questions, I assigned more extensive homework exercises. Although these exercises were designed to initiate thought about various aspects of Great Expectations, their primary purpose was to provide practice at navigating through the Web and constructing hypertext links. On one occasion I modified an existing assignment in the Web, asking the students to read the material on "Public Health" and "Race and Class" and then to locate a passage from the novel that was illuminated by the Web material, and to explain how. On another occasion I asked them to read the material in the "Religion" writing space and then to construct two links to other parts of the Web, explaining what they had learned from doing so, how their understanding of the novel had changed. In a third assignment I asked them to start anywhere in the Web and follow a series of three links, again explaining what the links had revealed to them.

Teaching in a computer classroom made it easier than in the survey course to reinforce these reading and thinking skills by adapting The Dickens Web for class discussions. Typically, I used the Web to initiate or complicate a discussion. Instead of defining the bildungsroman myself, for example, I had the students read the material in the Web and then discuss in small groups the specific ways in which Great Expectations fulfilled the criteria offered there, particularly in terms of education and vocation. After comparing the groups' conclusions, I invited them to read the brief essay in the Web on Utilitarianism and Dickens's opposition to it. I asked them if and how Dickens's view of Utilitarianism could be reconciled with his emphasis on work and vocation, which in turn facilitated a discussion of the novel's view of the individual in relation to the society.

Reading Hypertextually

My experiences with these two courses suggest that hypertext does have the potential to make students more active readers who can connect and synthesize even disparate material. But this potential is more easily reached the more hypertext is integrated into a course, and such integration in turn requires not just pedagogical reconceptualization but also the development of materials that reinforce what it means to read hypertextually. Like any new writing medium, hypertext, as Landow and others have stressed, redefines what it means to read and write. For most students, and certainly at least for the immediate future, this redefinition won't occur automatically or even rapidly upon exposure, as Landow contends. It will have to be taught and learned.

There are, to be sure, some aspects of hypertext, and particularly of Landow's Webs, that, because of the visual model they provide, alter student engagement with literature. I teach In Memoriam in several different courses, but students in the survey course displayed far more enthusiasm for it than any previous group. On the second paper the overwhelming majority chose the hypertextual option that utilized the Rossetti Archive rather than two more traditional assignments. In the course on the novel, The Dickens Web, by making such a range of information available in an unusual format, exposed the students to more contextual material than they are usually interested in reading--a striking confirmation of what Landow regards as one of hypertext's greatest strengths (Hypertext 126, 187). Whereas in my experience most students fail to utilize adequately the editorial and critical apparatus designed to bridge the gap between past and present, in this class almost half specifically remarked on their course evaluations that The Dickens Web had introduced them to a wider array of materials and approaches than they were used to. They saw more clearly why textual details matter and how these details can be connected to a work's broader themes.

The exposure to this range of interconnected material also generated more focused and interesting topics for the writing projects. One of the dilemmas I consistently face in a course that requires a research project is how to provide students the freedom to choose a topic that interests them while avoiding the vague and sweeping proposals that unlimited possibilities often invite. Because students tend to encounter only the results of research, not the process of it, they lack models both for the kinds of research questions they can ask and for ways to answer them. The diversity of material in The Dickens Web, however, provides them with just this kind of model. A number of students chose historical projects, usually linked in some way with Great Expectations's themes: Newgate Prison and confinement, Australia and the function of empire, sanitation and disease in Victorian London as both fact and metaphor. A paper on Dickens and Harriet Beecher Stowe that could have been a rather pedestrian comparison of their sympathy for the downtrodden became instead, thanks to the Web's material on social class, a discussion of the limits of Dickens's sympathies for the working classes in contrast to Stowe's more radical ideology. Another paper on Pip's relationships with Estella, Biddy, and Miss Havisham became, with the prompting of some of the Web's material on women and economics and Susan Walsh's essay on the use of the female body for figuring the circulation of capital, a much more sophisticated analysis of the novel's messages about women's roles. Even a more traditional exercise in close reading like one student's examination of the novel's allusions to Hamlet was enriched by links to information elsewhere in the Web on literary history and the father-son relationship. Again, nearly half of the students said in their evaluations that they found their research projects more interesting than in other courses.

Offsetting such benefits to a considerable degree, however, was resistance to the new medium, some of it rooted in computer anxiety and some in the mundane inconveniences of life at a commuter institution. About 20% of the students in the course on Dickens and Eliot felt that their discomfort with computers and the difficulties associated with working in the computer lab impeded their learning experience. In the survey course, some students avoided the e-mail discussion questions for the same reasons, and those who contributed regularly only got into the spirit of thoughtful but casual and conversational exchange on relatively rare occasions.

This failure of the discussion list underscores the value of the thorough integration, and extensive modelling, of hypertextual reading and writing. While I discussed the conventions of discussion lists and tried to model such conventions in my own postings, the students tended to approach this activity as they would in a traditional format. They wrote fairly formal replies clearly intended for my eyes rather than for a group discussion, and they responded to the comments of others very infrequently. (In one case, a student posted a plausible but deliberately provocative comment to the effect that the portrait of the Bishop in Robert Browning's "The Bishop Orders His Tomb" is sympathetic, but no one responded.) There were several causes for this. At the practical level, most of the students, lacking remote access to the campus network, had to go to one of the computer labs to post their comments. Since they tended to find thinking and composing in such an environment uncongenial, they often wrote their answers elsewhere. This translated into longer and more polished writing, with much of the immediacy and conjecture written out. And as commuters, their schedules often made it impossible for them to get to the computer lab until shortly before class. The small flood of isolated postings that invariably arrived in the hour or two before class confirmed that sending their own postings on time, not reading and replying to the postings of others, had the highest priority. But at another level, lacking experience not only with discussion lists but also with classroom discussion other than Socratic dialogue between teacher and student, the students were poorly prepared for and uncomfortable with a multivocal conversation among themselves. Even when I gave them the opportunity to respond to postings and to class discussions after the fact, very few did so.

Such discomfort with connecting the text, their own voice, and the voices of others in interpretive discussion and commentary was another version of their difficulties with links, which Landow calls "the essence of hypertext technology" (Hypertext 186). The mechanics of links did not pose significant problems for the students in either The Dickens Web or The In Memoriam Web. Following a link was a simple matter. For those students in the course on the novel, learning how to construct a link was more cumbersome, but with practice they became reasonably proficient at this as well. The difficulty was with what to do with a link once they had followed or constructed it. In obvious cases where the content was clearly related at a general level, they had no trouble. It was easy to see why the "Dickens and Darwin" writing space, which listed thematic similarities between the social world of Dickens's novels and Darwin's vision of nature, was linked to the "Darwin's Origin of Species" writing space, which described the major aspects of Darwin's theory of natural selection, or why clicking on the term "Reform Act" in the "Chartism" text space took them immediately to a text space that described the Reform Act. But what they often could not see, especially at first, was how a piece of information mattered to the novel, how it could affect our reading or understanding of Great Expectations itself. Presented with the knowledge that both Dickens and Darwin are concerned with issues of kinship, inheritance, and competition, for example, the students had difficulty moving from the identification of specific examples of these issues in Great Expectations to the development of an argument about Dickens's view of the "naturalness" of competition in the social realm.

It was evident in the survey course that the visual component of seeing and following links facilitated the students' understanding. They did a better job with the early in-class exercises conducted directly on The In Memoriam Web than they did with the three exercises described above on Ruskin, Christina Rossetti, and Arnold, all of which were conducted much later in the term but not on the Web. In the case of "Dover Beach" and "It is a Beauteous Evening," for example, they could see that Arnold's poem, while indebted to Wordsworth's, was very different in tone and mood, but they could not articulate what they felt Arnold was trying to say. Drawing their attention to passages from Arnold's essay on Wordsworth and "The Function of Criticism" only increased their confusion. And when they worked with the hypertexts independently for their papers, similar difficulties surfaced. They were very good at descriptively cataloguing the various appearances of their image in In Memoriam or the various revisions in "The Blessed Damozel," but their efforts to explain the significance of such individual occurrences and to synthesize them into an argument was no better than what I would receive in a traditional class.

This is, of course, no surprise. The ability to link what is not explicitly related, and especially what appears unrelated, requires considerable cognitive sophistication. Landow acknowledges this: "[p]erceiving possible connections and arguing for their validity," he writes in Hypertext, "is a high-level intellectual skill" (136). As experienced readers and critics, we can see the possibilities for such constructing and justifying connections that hypertext affords, but Landow goes further, arguing that hypertext enables "a novice reader to learn the habit of nonsequential reading" (126), to "quickly and easily learn the culture of a discipline" (128). Although many of my students became better nonsequential readers of literary texts, few achieved this "quickly and easily." The medium itself does not guarantee that students will make sophisticated connections, even with the kinds of exercises Landow describes. This was especially evident in the uneven success of my more extensive efforts in the Dickens and Eliot course to teach the students how to read hypertextually.

On the early homework exercise in which the students were to explain how following a link had affected their understanding of the novel, most provided a tautological answer: "Reading the material about prisons helped me because there are prisons in the novel, and knowing about them helps me understand why they are there." On the exercise in which they were to explain how a series of links had affected their understanding of the novel, only two or three students actually followed a series of links. The rest followed three different links from the same starting point, thus suggesting that their basic concept of a link was limited to a direct, one-step movement. My hunch is that most of the students who did this had initially tried to follow a series of links but saw no obvious way to relate the very different starting and ending points to each other or to the novel. This is, I suspect, another manifestation of their difficulty with making connections that are allusive, thematic, or abstract.

When we discussed this during our next class session, I asked the students to examine some of the materials in the Web on women's history and feminism. Then we followed a series of links from an essay on "Difficulties in Childbirth" to one on "Sanitation and Disease in Rich and Poor" to one on "Queen Victoria." I invited them to brainstorm about the connections among these writing spaces--what they revealed when taken as a group--and about possible connections to Great Expectations. They simply could not develop any ideas--even when I prompted them with a character like Mrs. Joe on whom to focus. "What connection could there possibly be between Mrs. Joe and Queen Victoria?" asked one student in genuine exasperation. This question led many of the students to complain that very little of the material in the Web seemed to be connected to the novel at all.

The students' response revealed much about their assumptions. For them, The Dickens Web was a reference tool, not a pedagogical one. In Delany and Landow's terminology, they treated the hypertext as a "resource" (information provided by experts for extraction by individual users) rather than as an "environment" (a shared body of knowledge that users continually reshape) (32-33). They wanted it to provide them with answers, or at the very least with the questions. They approached it in a fundamentally passive way and hence were annoyed rather than gratified by the freedom it provided.

Exposing and disrupting such passivity also requires explaining explicitly the role of the hypertext in the course and foregrounding discussion of how hypertexts are to be read. Such self-consciousness demands even more in the way of reinforcing exercises and assignments than I provided. After our class discussion about Queen Victoria and Mrs. Joe, I prepared an exercise involving the World Wide Web that provides an example of how students can be brought to understand the principles behind hypertextual connections. I asked the students to access and explore a Web site devoted to an exhibition at the University of California-Berkeley's art museum on "The New Child: British Art and the Origins of Modern Childhood, 1730-1830." The exhibition argued that our modern conception of childhood as a special, innocent period around which family life revolves, was a construction of the Enlightenment and a radical departure from the earlier paradigm in which children were viewed merely as small versions of adults (see Steward). After reading the exhibition's on-line text and viewing some of its images, the students were to explain what relationship Dickens's depiction of Pip's childhood bore to the exhibition. Virtually every student saw that Pip's childhood departed from that of the "new child," but many also went further. Some recognized the importance of class issues, arguing that since Pip came from a working-class family rather than a bourgeois one, he was less the exception than he might appear. Others were even more perceptive, contending that Dickens exploited this conception of the "new child" as a standard against which his audience would assess Pip's experiences and, thus, sympathize with them.

Why did this exercise produce such impressive responses when similar exercises that also included meaningful but informal writing about the significance of links did not? There are, I think, two reasons. First, the assignment was focused exclusively on the activity of making conceptual links--it was not seen by the students as an ancillary component of a larger activity. Second, it was directed--the students were given two specific things to link, the exhibition's argument and Dickens's depiction of Pip's childhood. Additional exercises of this type early in the term, employing materials either within The Dickens Web or outside it, could gradually yield to more open-ended assignments in which students make their own links and determine their own connections.

Such exercises and meta-discussions about reading did not fully solve the problems. The students' explanations and justifications of the links for their Web projects remained heavily tautological. And the projects themselves, even when most interesting, tended not to link contextual material and novel very successfully. Textual analysis remained at a general level, rarely focusing in detail on, or commenting on the significance of, the passages most relevant to the contextual material. It took two drafts for the student who wrote on the experiences of convicts in Australia to actually analyze Magwitch's account of his experiences. The person who wrote on Hamlet initially did not discuss the production of the play that Pip attends, and he had trouble connecting it to the rest of his thematic analysis. For The Middlemarch Web, a student who researched the competing theories about the treatment of fevers had to be coaxed into examining the passages where Lydgate defends his practices against those of his colleagues.

On the other hand, the students increasingly approached their writing projects in ways that the hypertextual medium facilitated. They took advantage of the opportunity to present their work in non-traditional ways and their presentation of contextual material was almost invariably effective, reducing complex arrays of information into clear, concise, and relevant mini-essays, as in one student's essay on nineteenth-century medical education in Britain and France. They also saw their projects as connected to other, sometimes very different, parts of the Web; in several cases they became interested in and made links to the work of other students in the class. A student working on Dickens's depiction of Joe Gargery's working-class dialect, for example, was surprised to find that her project had parallels with the project on Dickens and Stowe as well as another on utilitarianism and the novel's treatment of industrial versus mercantile capitalism. For The Middlemarch Web, a student working on medical education linked her material to another's work on the 1832 Reform Bill to show how the training of the various medical men in Middlemarch signals both their social status and political views. By that point in the term, many of the students were employing the conceptual language of hypertext, and that language had in turn encouraged intellectual interaction and collaboration among them.

The use of hypertext thus really does offer the potential to expose students to a wider range of possibilities in the study of literature, to develop their disciplinary reading skills, to enhance their engagement with the material and with each other, and to develop course materials that can be used and expanded by future classes. (For a discussion, based on my experiences in the Dickens/Eliot course, of the practical and theoretical implications in making student contributions public, see Smith.) But because links are so central to hypertext, the more it is used the more it exposes student weaknesses in reading--especially the inability to read non-linearly and to link text with context, specific detail with general theme, or one facet of the text with another. And since most students have been trained to be passive readers rather than active, independent, and subversive ones, such exposure tends to increase their anxieties, at least initially, rather than to liberate and excite them, as Landow claims. This is probably a necessary step, one that, if brought into the open, can then be incorporated into the process of teaching the students how to read hypertextually. It means, however, if my experiences are any guide, that for hypertext to live up to its pedagogical potential, it should be integrated as much as possible into the course. Instructors should be prepared to revise not just their individual class sessions and writing assignments but the very conception of how they teach. The rewards, however, are worth the effort.

Works Cited

Bolter, Jay David. Writing Space: The Computer, Hypertext, and the History of Writing. Hillsdale, NJ: Erlbaum, 1991.

Delany, Paul, and George P. Landow. "Hypertext, Hypermedia, and Literary Studies: The State of the Art." Hypermedia and Literary Studies. Ed. Delany and Landow. Cambridge and London: MIT P, 1991. 3-50.

DeLoughry, Thomas J. "Students and Colleges With Ample Computer Access Find New Opportunities; Others Lag Behind." Chronicle of Higher Education. 14 June 1996. A16-17.

Everett, Glenn, et al., eds. The Browning Page. N.D. Online. 27 May 1997.

Gilmour, Robin. The Victorian Period: The Intellectual and Cultural Context of English Literature, 1830-1890. London and New York: Longman, 1993.

Landow, George P., ed. The Dickens Web. Diskette. Cambridge, MA: Eastgate, 1992.

---. Hypertext: The Convergence of Contemporary Critical Theory and Technology. Baltimore and London: Johns Hopkins UP, 1992.

---, ed. The Victorian Web. N.D. Online. 27 May 1997.

---, and Jon Lanestedt, eds. The In Memoriam Web. Diskette. Cambridge, MA: Eastgate, 1992.

Lanham, Richard A. The Electronic Word: Democracy, Technology, and the Arts. Chicago and London: U of Chicago P, 1993.

McGann, Jerome J. The Rossetti Archive. 1993. Online. 27 May 1997.

Smith, Jonathan. "What's All This Hype About Hypertext?: Teaching Literature with George P. Landow's The Dickens Web." Computers and the Humanities 30 (1996): 121-29.

Steward, James. The New Child: British Art and the Origins of Modern Childhood, 1730-1830. Berkeley: University Art Museum/Pacific Film Archive, 1995. Online. University of California-Berkeley Art Museum. 27 May 1997.

Storyspace. Vers. 1.0.9 for Windows. Diskette. Cambridge, MA: Eastgate, 1995.

Walsh, Susan. "Bodies of Capital:Great Expectations and the Climacteric Economy." Victorian Studies 37 (1993-94): 73-98.

William Morris Society. William Morris Home Page. N.D. Online. 27 May 1997.