Dissertation:  An Investigation Into the Hypertext Composing Processes of Middle School Students

The purpose of this dissertation is to question how the act of composing hypertext, the discursive form that allows readers and writers to link to other pieces of text, affects students’ senses of textuality and agency as writers. Its purpose is also to question how hypertext affects students engagement levels in their composing tasks. In this opening chapter I will define hypertext and provide an explanation of hypertext and the theory that surrounds it. I will also provide a brief history of hypertext and discuss some of the research that has so far been conducted on hypertext composing.

Understand that little research has been done in this area. Some teachers like Dene Grigar, in her article “What Is Seen Depends on How Everybody Is Doing Everything: Using Hypertext to Teach Gertrude Stein's Tender Buttons" (27-43) and Gretchen Lee’s 1999 presentation at NCTE’s fall convention, have reflected on the ways in which hypertext composing has impacted classroom practice, and Bradford Barry in his 1998 dissertation "Writer Motivation, Rhetorical Purpose and Classroom Web Publication Projects," has researched student motivation when composing hypertext, but to date no studies have examined the process students use as they work within a hypertext environment. Although research on hypertext composing is limited, there has been research conducted regarding the impact of computers on student writing as well as theories regarding how hypertext may alter their perceptions of text.

Definition of Hypertext

Hypertext can be defined in a number of ways.Perhaps the best explanation is that of Joseph A. Feustle, Jr. who compares hypertext to a piece of woven cloth (299).The threads of the cloth allow us to reach out “by means of a computer program and connect original works, critical studies, bibliographies, and historical backgrounds” (299).

Hypertext has also been simply defined as the electronic linking of text (3).Jay Bolter likens hypertext to a printed book that the author has attacked with scissors.He points out that it is an interactive format that can challenge our concept of how text should be organized (24).Hypertext links, of course, are those words, phrases, or images, often underlined or boxed, that, when clicked, allow us to jump to another document or page.Many hypertext theorists refer to those pages as “lexias,” (Hypertext, 4) a term that George Landow borrowed from Roland Barthes, a French literary theorist who died in 1980.Rather than use the word “page” which suggests a place in a linear series, Landow uses the term “lexia” so that we can shake loose our notions of textual organization, our “book bound” idea that one page has to follow another, that one idea has to build upon another.The term “hypertext” was coined in the 1960’s by computer visionary Theodore Nelson who described it as “non-sequential writing” (3).

Flickers of Light

William Costanzo regards the computer screen as one of the changing sites of literacy today. Costanza explains, “Anyone who has written with a computer knows that language on the screen seems different from language on the page.It seems more flexible, more fluid, more akin to the flickering of light than to the fixity of print” (11). Costanza refers to the fact that electronic text can be changed so easily, and that the text is really nothing more than electronic flickers of light and dark.And this flicker, this fluid text that can be effortlessly changed may alter forever our concepts of text, textuality, and our explanations of how we construct meaning from and with it. Text is no longer something that belongs just on a page.It occupies new spaces.And because it occupies new spaces, and those new spaces allow text to remain in flux, to remain fluid and plastic, Ilana Snyder (1998) believes that it alters how we communicate.She is not alone.

Cynthia Selfe and Susan Hilligoss believe that technology “both physically and intellectually disrupts the ways in which we make meaning—the ways in which we communicate”(1).And though we are living through a communication revolution and tend to think that our experiences in the midst of these new communication spaces are unique, we must remember that humans have initiated such shifts before.Writing is technology.And when a new technology develops, according to Snyder the new technology does not signal the immediate death of the older technology (2).We still use hand written notes, for example.But we no longer use it in book production.In that sense, Snyder points out, new technologies disrupt older ones, relegating the old to specific uses, and challenging readers and writers to redefine what it means to read and write text. (xxi)Aronowitz adds that writing and technology are “ineluctably intertwined,” (133) and as such, affect and redefine literacy.

Educators have long seen the educational possibilities inherent in computer technology and often resort to almost utopian rhetoric when discussing the power of computer technology in the educational setting.The National Education Technology Standards published by the International Society for Technology Education, state that “Technology is a powerful tool with enormous potential for paving high-speed highways from outdated educational systems to systems capable of providing learning opportunities for all, to better serve the needs of 21st century work, communications, learning, and life”(XI).For this reason schools have embraced computer technology since personal computers became affordable and hence available in the mid 1980s.With the growth of the internet, schools have increasingly made use of web based technology, especially in the area of information retrieval.

More Than Information Retrieval

But some teachers in some schools are also using web based technology, or hypertext, as a means of assisting students in the act of composing and for publishing.And though it does not represent the primary composing experience of most students yet, it is a growing phenomenon.

Before we go any further in this discussion, however, it is important to understand that the hypertext composing in this dissertation is situated within a school environment where, critics have observed, student writing is generally not privileged.In other words, the writing that students produce in a school setting is not particularly valued by teachers or by those outside the school community.Student writing is either used as an example of how schools are failing, or it is used as an example of how much individual students do not know about written language.It is an artifact to be graded, commented on.

Michael Joyce believes that hypertext technology specifically offers the possibility of adapting cognitive skills “that experts routinely, subtly, and self-consciously apply in accomplishing intellectual tasks" (40).He adds that hypertext tools “promise to unlock these skills for novice learners and to empower and enfranchise their learning” (40).

Part of that power and enfranchisement comes from attending to the issues of readers.In other words, hypertext writers must write their documents in such a way as to assist readers so that they may read efficiently.To do this, the writer must think of the reader’s pleasure and comfort within the web, and must give the reader some indication as to what lies beyond a link.Snyder points out that “Effective hypertext writing depends therefore on the tension between regimentation and richness, between predictability and excitement” (35).

Readability is something all writers must consider, of course.And it is one of the functions of writing instruction in schools to help students write text that others can read, to consider the needs not only of themselves as readers, but others. And so the question is, how do student writers go about the process of negotiating Snyder’s “regimentation and richness, ” that “predictability and excitement?”And does that tension between those elements empower and enfranchise their learning?

This deliberation involved in hypertext, of course, involves dialogue.Hypertext is always a dialogue between the reader and the writer, and Bolter points out that both share the responsibility for the outcome of any given web environment (117).The hypertext environment brings the reader and writer into conscious dialogue.The writer invites the reader to choose paths.The reader considers the author’s invitations and follows various paths through the links-as-invitation on that quest.According to Bakhtin, dialogism is the relationship between the self and an other, with the self-maintaining a relative center to the other.But in order for there to be a self, there must be an other.A user of text, then, is always in dialogue with an other, perhaps many others when the text is uttered or put into thought.Douglas Eyman, in a 1999 Kairos article, applies Bakhtin’s concept of the utterance, thought which is given voice, to hypertext and calls the lexia an example of a Bakhtinian utterance.Eyman writes that Bakhtin’s“description of the environment in which a given utterance is produced can be appropriated and used as a description of the function of lexia in a hypertext” (http://english.ttu.edu/kairos/1.2/features/eyman/bakhtin.html).

Bakhtin explains that "the authentic environment of an utterance, the environment in which it lives and takes shape, is dialogized heteroglossia, anonymous and social as language, but simultaneously concrete, filled with specific content and accented as an individual utterance" (272).Eyman reminds us that in order for us to engage in dialogue, we must be able to apprehend, internalize, and recreate the utterances of others.Kristeva believes this is the same experience readers have.Kristeva’s notion of intertextuality is perhaps the closest description we have of what happens when a reader and writer engage in a hypertext.Eyman describes Kristeva’s “coordinates of dialogue” as “the writing subject, the addressee (or ideal reader), and exterior texts” (http://english.ttu.edu/kairos/1.2/features/eyeman/julia.html).These are important considerations when we think of the associational thinking that takes place when a reader dances through a hypertext web, and when a writer choreographs the dance a reader might perform when he or she sees the music of links. And part of that deliberation is dialogue.

In order to determine whether hypertext is able to do this, it is important to investigate the processes students appear to use in the act of composing hypertext or web-based documents. But before doing so, it is important to point out that I am not claiming that the heightened senses of agency, textuality, and engagement are elements that can only be found in hypertext composing.But I am arguing that hypertext does indeed seem to create a textual environment that allows or supports student engagement in the task of composing, that the nature of it seems to heighten students’ sense of textuality, and that it places students in a textual environment where they become agents who can make decisions about their emerging texts and see the results of those decisions.Unfortunately, this does not seem to happen consistently with other composing tasks in a school environment.

Explanation of the Project

This study will investigate that site of hypertext as it is negotiated by middle school students as they write informational hypertext documents.It will pose the question as to whether hypertext impacts the composing processes of middle school writers.Specifically, the study will address the following issues:
How do students approach the process of composing hypertext webs?

What role does agency play in the process?
How do students develop and use their sense of textuality as they compose their webs?
What role does engagement play in students’ hypertext composing processes?
In what way, if any, does a sense of audience affect decisions the student authors make?
How do the student authors decide what their links will be?
How do students keep track of multiple textual threads within a single hypertext web?
How do students maintain control of their composing processes within a multi-linear domain?
Before we can look at student hypertext composing processes, however, we must examine three issues—the composing process in general, the rhetorical situation of school composing, and the impact of computers in the composing processes.
The Writing Process
Writing is a mode of expression that is unique to each individual.No two people compose a piece of writing in quite the same way.And not only is the process unique to each individual, it is unique to each composing task.However, there are some key elements or steps that individuals generally take.Generally, a writer goes through some sort of invention process where he or she decides what to write and begins organizing materials, thoughts, and strategies in order to prepare for the next step, which is drafting.During the drafting process the author writes the piece.But also during the drafting process, an author will often critically read his or her piece, may return to the invention process in order to better organize his or her thoughts and information, and may, indeed venture into the next phase of the process, which is revising.Revising involves revision of the piece.Again, during the revising process, an author may return to the invention process and may return to the drafting process. At some point in the process a writer will also edit.This means the author may go back and correct any surface errors, or he or she may substitute one word for another.During the editing stage, however, a writer may continue to revise, draft, or even invent. Generally, the final stage of the process involves publication, which means that the piece of writing gains a larger audience.However, that does not mean the author stops revising, drafting, inventing, or even editing.The stages of the process are fluid and recursive rather than linear and lockstep (http://www.csuohio.edu/writingcenter/writproc.html).

The Classroom as Rhetorical Situation

It is important to understand that this study takes place in a public school classroom, a middle school classroom.William Bigelow, when talking about his own Portland, Oregon school, points to the “hidden curriculum” of compliance, sexual stereotyping, and a reinforcement of class differences in his school, and by extension, most schools (75).Bigelow adds that students are generally powerless to effect any lasting change (78).David Bloome and Sonia Nieto point out that such literacy staples as basal readers, which two of the subjects in this study experienced in Riverton, and one subject experienced in a neighboring school district he attended before moving back to Riverton, teach students a hidden curriculum that is “less about learning to read and write and more about adapting to a bureaucratic society” (86).And later they add that using basal readers helps young readers learn their standing in a social order and tends to marginalize some students, disconnecting them from ”conditions for learning and intellectual and personal growth and development” (92).Eric Crump, in a 1999 post to an NCTE e-mail discussion list, points out that schools are really institutions of control rather than places of learning.It isn’t that learning doesn’t happen in a school environment, but it is designed to happen in controlled spaces by controlled people, teachers, who are charged with maintaining control of their students. Only certain kinds of knowledge can be privileged in that controlled environment, and only certain kinds of behavior can be accepted.

And though this dissertation is not about basal readers or the ways in which the culture of schools maintain a structure of control that is not conducive for all student-learning, we need to understand that when we talk about agency and engagement, we must attend first to the ways in which schools systematically keep students from feeling a high degree of agency and engagement in many of the classroom tasks they take part in.

Everything from bells to bathroom passes send messages to students that they are not in control of their learning spaces or their time.Homework assignments, project deadlines, reading lists, multiple choice tests.These all send messages to students about what should be learned and how.They signal how someone else will determine whether the material was learned and how well.And school is a place where a “good attitude” means a student has learned to comply with rules governing behavior and learning.

At Riverton Middle School students experience a number of different classroom styles that charge them with copying overhead lecture notes all hour to engaging in hands on experiences with historical artifacts.They experience a situation where they are allowed to use a bathroom pass twice within a marking period to being allowed to leave whenever they need to as long as no one else is out of the room.In some language arts classrooms they are told what to read, what projects they will do and how those projects will be designed. In other language arts classrooms students have more choice and voice regarding what they read, how they will be assessed, and what shape such things as projects will take.

In my classroom where this study was carried out, students were given choices within a pre-defined area as to what they would study.For example, they were asked to annotate a poem, but allowed to choose that poem from a collection.They were asked to create a hypertext web on a famous American, but allowed to choose which person they would study.Within those projects they were allowed to let their interests take them wherever they lead, so that if a student wanted to focus just on Harriet Tubman’s life as a slave, he or she could.If a student wanted to focus just on her experiences leading slaves to Canada, he or she could.And, students had the option to switch topics if they found the one they had chosen was not working out for them.

Computers in the Classroom

Most of the research on computers in the K-12 classroom deals with word processing.Most of that research indicates that word processing enables students to write longer pieces and to revise more easily.Because hypertext composing and word processing share some similarities, and because little research has been done on hypertext composing, I will review here the literature on word processing in the classroom in the hope that it will provide a foundation for thinking about hypertext, a later text creation technology.

Word processing is perhaps the most apparent use of computer technology in language arts classrooms, and there is a growing body of research that indicates that word processing enables the writing process. Costanzo asserts that word processing software supports the recursive nature of the writing processes, that word processing software programs encourage writers to fluidly shift between generating ideas, editing, conceptualizing and reconceptualizing text.Costanzo goes on to say that computers give inexperienced writers “access to alternatives that might otherwise remain invisible” (17). In other words, inexperienced writers have a tool that lets them create “professional” looking text, which they can change efficiently. In short, computers allow inexperienced writers to behave like their more experienced counterparts. Certainly this is true for students who create hypertext documents as well.Not only are they able to efficiently and fluidly move back and forth through the recursive stages of the writing process, they are able to employ color and graphics, and while it is true that the can do this with word processed documents, hypertext includes links to other documents in the web.This graphic and textual element makes hypertext different.

Ilana Snyder, in a 1994 study that examined the effect of word processing on students, concluded that word processing helped foster more student engagement in the composing process, and that, when compared to a group who used pen and paper for composing, word processing seemed to help students generate better, more complex text (157). Other studies, too, show that computers seem to impact student motivation and engagement (Baker 1987, Hawisher 1986, Lichtenstein 1992). In my general observations over a period of three years, I found that students experienced consistently higher degrees of engagement in hypertext composing than when they composed word processed documents.

Ronald Owston, in a 1992 study that compared the writing samples of 111 eighth graders who composed two pieces of writing, one “by hand” and the other using a word processing program, observed that when students used computers to write, they moved blocks of text, and continuously revised in other ways as they composed. He concluded that students created better pieces of writing when they could compose using a word processing program than when they composed “by hand” (270). This ability to move blocks of text is possible during hypertext composing as well.And we cannot forget that hypertext carried with it the ability to not only move text, but to hide it.This, perhaps, is the ultimate experience in “moving” text—the ability to hide it behind a link.

In 1984, Gail Womble observed that sixth grade students spent more time with their writing when they were allowed to use computers and seemed more willing to make more changes in their emerging texts.She also observed that they seemed to develop a better sense of audience when using a computer for composing as opposed to using traditional pen and paper (35).Owston, too, in his work with eighth graders who were very familiar with computers and word processing found that their pieces of writing, when scored holistically, were consistently judged higher than the works of students who wrote by hand (252).We will see in this dissertation how growing familiarity with a web editing program and the nature of hypertext itself, allowed students to create we more visually and textually complex webs.

In a 1992 study conducted in a whole-language classroom, James Strickland found that, because of the already language-rich environment, the use of computers in writing fostered a more collaborative learning community, which, in turn, supported students in their writing (17).This collaboration was due partly to the collaboration that arises among students who are engaged in literate conversations about texts, and partly to the fact that computers screens allow for better viewing of text by more than one person. It is also due to the fact that computer text can be easily changed.When collaborating students discuss their emerging text and make decisions about that text, they can easily take action regarding those decisions.

Numerous studies also indicate that students generate more writing and better quality writing, measured in a number of different ways, when they are allowed to use word processing programs (Kurth 15,Diaute 143, Baker 117).That students generate more text indicates heightened senses of engagement and motivation.Students are more fully engaged in the act of composing and motivated to generate longer pieces. I found this to be true for hypertext composing as well. But not only did students seem to write more, they wrote using a surprisingly sophisticated sense of how to use other textual features such as images, text placement, and color.

It is important to point out that not all studies drew researchers to the conclusion that computers have such a positive effect on student writing.In a 1992 study involving thirty-two fifth graders in New Jersey, Nora Lichtenstein concluded that student use of computers for writing had no significant effect on the quality of the writing (93). Students who composed using a computer were allowed to do so once a week in a computer lab.Students were pre-tested and post-tested and their writing was evaluated holistically.

In a 1996 study involving thirty-eight sixth graders, nineteen of whom used paper and pencil, and nineteen of whom used computers for composing, researcher Lois Mayer Nichols found that the pieces composed by students using a computer did not significantly differ from those written by students using pencil and paper. However, Nichols noted that students who used a computer for composing wrote longer pieces than those who used paper and pencil. Other studies looked at the number of drafts students created using pencil and paper as opposed to a computer and concluded that there was very little difference between the number of drafts students created using a word processing program than the number created with paper and pencil. Zeni (Literacy 80) points out that many of the studies that look at the differences between paper and pencil composing and word processing look at the number of drafts generated.But Zeni explains that the whole concept of drafts really belongs to the paper and pencil mode of composition.It is no longer appropriate when computers are used for composing to think of first drafts and second drafts.A single draft evolves organically during the time writers begin and decide a piece is finished. Jane Zeni (Writinglands 93-94) warns that computers and the flexible programs that come with them cannot transform rigid teaching practices into process friendly results.

The differences between book culture text and electronic text may go beyond simple differences in how it is created.Or rather, those simple differences may have more profound effects on learning and composing than we initially think. Klemm and Moran stress that word processing alters the nature of text and, thus, alters the way in which we read digitized text (135). Bolter believes that electronic text is “non-print, undark, dry, unimprinted, prone to sailing off” (86) and for that reason is volatile.The reader, according to Bolter, loses track of where he or she leaves off and where the writer begins (86).Certainly this is true for hypertext where the reader charts a path through the text, a path that no one, including the writer, anticipated.But perhaps the biggest difference isn’t so much with the text itself, but with the expectations the reader brings to the text. Hypertext readers know they will have to make choices.They may go into the text with the expectation that they will be initially confused, that they will have to suspend a certain “satisfaction of knowing” until they navigate further across the docuverse.And oddly, though readers may have to suspend satisfaction, they may also feel a sense of agency, of being in control of the text.It could be that hypertext overtly invites readers to insert themselves by way of the cursor or the hand that indicates a hypertext link.Book technology seems to fix a certain notion of author as authority.Hypertext challenges us to rethink that role.But I must point out that other texts do this also.I only suggest that hypertext more overtly challenges us through the cursor, the hand icon, and the links.

There are of course, similarities between word-processing and hypertext composing.Both spaces allow for text insertion and other manipulations.Both are electronic.And both allow for the use of color and images.But Bolter points out the differences between these environments.One, of course, is the hypertext link.Most word processing programs allow for the insertion of a hypertext link, but this speaks more to the growing use of hypertext and the world wide web as a publishing space than it does the similarities between these two electronic environments.Both allow for the insertion of images, but there is an important difference between the use of images in a word processed document and the use of images in hypertext. Bolter points out that images in word-processed text are not really part of the text.Images are “merely allowed to coexist with the verbal text” (26). Hypertext allows images, even video, “to sit among the other textual elements for the reader to examine” (26).The image becomes part of the structure of the whole text, the larger text of any given lexia.In many hypertexts these images are links to other other pieces of text, and so they are in a sense portals that lead to other spaces within a textual environment.

Evolution of Hypertext

This dissertation will assume that hypertext represents a new discursive form born partly out of the need to make associations between and among the growing amounts of information present in the textual world. To consider the ways in which hypertext alters the nature of text, we need to look at how hypertext developed and the cultural changes that facilitated the acceptance of hypertext as a textual environment.

Many attribute the concept of machine-linked text to Vannevar Bush who, as Director of the Office of Scientific Research and Development in 1945 envisioned a Memex, a microfilm reader that would access information quickly and allow the reader to search for related information by typing a key word into the system.Bush envisioned a mechanical device built into a glass-topped desk equipped with a microfilm reader. Bush’s Memex, had it been built, would have served as an interactive library or encyclopedia, much as the world wide web serves as an interactive encyclopedia today.The hypertext we see on the world wide web has its technical origins in the beginning of the computer age when people like Douglas Englebart began dreaming of the possibilities.Englebart, the inventor of the mouse, initiated the Augment Project at Stanford Research Institute.His goal was to link and cross-reference research materials, much as Bush envisioned.One aspect of Augment, oN-Line System (NLS) included a number of hypertext characteristics.

The term hypertext was coined by Theodore Nelson in 1965 to describe non-sequential electronically linked text.According to Nelson, hypertext was a means of linking documents to create a web of inter-related concepts that would allow readers to follow associative paths in much the same way that Vannevar Bush envisioned.Nelson, over the years, developed a system he called Xanadu named after the Coleridge poem.Nelson envisioned a system of documents “measureless to man” where a reader could follow his or her own interest over a network.In addition, anyone who wanted to add a document to the network could do so.Although Nelson released Xanadu to mixed reviews in August, 1999, most people recognize a version of Xanadu in the world wide web. And, actually, some of the early world wide web developers were influenced by Nelson’s ideas.

Over the years Nelson developed his definition and concept of hypertext.As of 1990’s he was defining it as non-sequential electronic text that contained links that are controlled by the reader (Snyder, Hypertext 24).The important elements in this definition are “electronic”, “reader controlled,” and “non-sequential.”These are very important elements. But at first look they are not necessarily peculiar to hypertext.

The dynamic of reader control adds the other important, indeed dramatic element to hypertext. Derrida, Bahktin, and Barthes all allude to the shifting boundaries of text, the capacity of text to do more than sit on a page and denote only a single meaning.All three challenge the assumed linearity of writing, or text.Bakhtin, cited in Landow (Convergence 28) writes “Each passage is a node, a point of intersection or focus, on which converge lines leading from many other passages in the novel and ultimately including them all” (in Landow, Convergence, 28).Hypertext brings that challenge forward.Bolter reminds us that there is a gulf between the reader and the writer, a gap that is mediated by the technology that helped create the text. But he adds that with hypertext, the reader is “made aware of the author’s simultaneous presence in and absence from the text, because the reader is constantly confronting structural choices established by the author” (30).These choices are in the form links, visual reminders that the reader can choose where to travel next.

Of course readers of book culture text have always had the option of browsing through a text, but some texts seem to forbid that more than others, and some contexts, such as classrooms, seem to discourage that.Hypertext, however, seems to invite readers to choose consciously, deliberately, which path they may follow.Again, this has always been possible, but hypertext provides an electronic means of doing this, and even textual permission to break the sequential rules of textual organization.Hypertext instantiates that breaking.It provides a deliberate multi-linear space that constantly re-centers itself as each new text block is called into being by a reader. And this overt, kinesthetic (meaning the physical moving and clicking of a mouse) means of choosing a textual path carries some interesting implications.

Michael Joyce defines hypertext as a visual form that “embodies information and communications, artistic and affective constructs, and conceptual abstractions alike into symbolic structures made visible on a computer-controlled display” (19). What is important here is that Joyce, like Bolter, identifies hypertext as a visual form. And though all written text is visual, hypertext seems to lend itself to topological rhetoric. This rhetoric shines through in such concepts as readers “navigating” through a web, and in the fact that hypertext writers often create a “map view” of their webs.Internet readers “surf.”In addition to these common terms, Joyce refers to a city of text. Nelson calls it a docuverse.Both terms refer to the collection of linked text that makes up a hypertext web, and both use the topology metaphor.These are all somewhat “navigational” terms that suggest that hypertext is a world over which a reader travels.This graphic concept of the texts within a web, whether deliberately linked by a single writer or “casually” linked by many writers who may be known to each other or unknown, is an important concept when thinking about how students negotiate writing within a webbed environment.

Hypertext Research Literature

Only a few studies have focused on hypertext composing. All of them indicate that hypertext composing fostered a great deal of intrinsic motivation on the part of students.In one study involving twenty-ninth graders enrolled in an American History class, students in the class were asked to create a hypermedia presentation that would be used to teach their peers about various topics dealing with American history.Students had a great deal of freedom in designing their hypermedia documents.Researchers concluded that students researched and wrote in more depth when they could use hypermedia composing software as opposed to more traditional composing tools (Connell 252).

Dene Grigar notes that her Texas Woman’s University students wrote more text during their hypertext project on Gertrude Stein’s “Tender Buttons”, where they contributed to a web that Grigar had already begun.Their work, according to Grigar, deepened their knowledge and understanding of a difficult piece of literature (39).

Gretchen Lee found the same to be true for her sixth graders when they created a hypertext web about a book they read as a class.Their knowledge and understanding of a class novel increased as they worked on their web site.Lee attributes this to increased motivation perhaps prompted by the novelty of working on documents that would be published on the world wide web (NCTE convention).

Bradford Barry concluded in his dissertation Writer Motivation, Rhetorical Purpose and Classroom Web Publication Projectsthat hypermedia composing nurtured a high degree of rhetorically-based, intrinsic motivation among college freshman writers.He called this “rhetrinsically-based motivation” and defines it as “self-sponsored rhetorically based writing” (2).

None of these studies, however, investigated the composing processes students used.Ellen Barton and Ruth Ray point out that most of the research conducted in computers and writing has attempted to explain how or defend the ways in which technology can improve writing instruction (279).However, this dissertation is an investigation of the composing processes, an attempt to trace the ways in which three students interacted with their own emerging hypertexts.The purpose is also to argue that a hypertext composing environment enabled students to fully engage in and reflect upon their emerging texts. Other forms of writing may do that, as well; however, in the three years that I have included hypertext composing in my classroom, I did not see other forms, both electronic and non-electronic, engage students so consistently as did hypertext.

Smith and Curtin argue that because students are immersed in a time of ever-increasing technological development, it is possible that their methods of thinking and processing information differ from those of past generations.In their view, the pluralism of identities and perspectives emerging today and the increasing domination of computer technology directly affect the structures of educational systems.They warn that the very institution of education as we know it is challenged by these cultural shifts and that the total reassessment of curriculum and schooling is necessary (xxix). This dissertation may shed some light on the ways in which that reassessment can be directed.

Theoretical Perspectives

The purpose of this section is to explore some of the theoretical foundations of hypertext and the effects it may have on our notions of text and textuality.

Hypertext as challenge

One of those notions regards the theory that hypertext challenges us to abandon conceptual systems based on the idea of center, margin, and linearity, much the way Derrida challenges us in his playful approach to text in The Post Card.Derrida, writing about a post card that seems to reverse the positions of Socrates and Plato, writes:

What I prefer, about post cards, is that one doesnot know what is in front or what is in back, or there, near or far, the Plato or the Socrates, recto or verso. Nor what is the most important, the picture or the text, and in the text, the message or the caption, or the address.Here, in my post card apocalypse, there are proper names, S. and p.,above the picture, and reversibility unleashes itself, goes mad (13).

Derrida considers text and language as the free play of signifiers and his playful speculations about a post card depicting Socrates and Plato could very easily address the non-sequentiality of hypertext and the “hidden” texts that lurk within a larger “docuverse” or collection of webbed texts.It could also address theconstant shifting of visibility and invisibility of text.The whole notion is that behind a single link lurks the possibility of a docuverse that holds behind it more hidden docuverses.Or it could hold nothing, a continuance of invisibility, all there, or not there at the click of a mouse button. In hypertext, the concept of near and far fade.A block of text is as near and as visible as a click of the mouse, or as far and as hidden as the textual thread it lurks in.And that block of text may be first for one reader, and last for another. Or it may be first for a reader during one reading, and last for the same reader during another reading.Reversibility goes mad. Reversibility is the fact that one reader can reverse the order of the text, and another can invent a whole new text.Reversibility also pushes the question as to who the author is, and who the reader is.Barthes of course pushes this issue for print text, but hypertext, because the reader can literally rewrite the text by shifting the order in which lexias are accessed and read, pushes that issue one step further.

Landow believes that electronic texts like hypertext shift textual boundaries (again we see the topological metaphor) between the reader and author and between the student and teacher, so much so that our whole concept of text, authorship, literature, and reader will be redefined (23).In her book The Electronic Labyrinth, Ilana Snyder, like Landow, believes that hypertext is changing our notions of authorship.She and Landow note that the absence of textual autonomy and centeredness disperses the author.They, like Tuman, point out that computers shape the way we think, encouraging some kinds of thinking and discouraging others.Snyder uses the example of a blackboard where text is created with the assumption that it will be erased.Paper and pen writing encourages writers to attend to grammar and spelling and to use a more controlled type of thinking.Computers, according to Snyder, invite writers to think non-linearly and cooperatively (17). But again, this is not a new concept.

Snyder points out that “we organize our writing space in the same way we organize our thoughts, and in the ways in which we think the world itself must be organized” (69). She, like Landow, sees nothing short of significant changes in the world because of the kinds of shifts hypertext is calling us to make, shifts as profound as those that took place when a culture adopted print and slowly abandoned manuscript as a means of framing written language.

Hypertext and the Rhizome

Much of that shift comes from the notion of centeredness.An explanation of this can be taken from Delueze and Guattari’s metaphor of the rhizome, that ever recentering web of nutritious fleshy roots that spread along the surface of the soil and sprout green stalks.Each rhizome grows offshoots of itself, expanding the network of roots irregularly.The center constantly shifts and multiplies into numerous centers as the rhizomes grow.Stuart Moulthrop writes:

Seen from the viewpoint of textual theory, hypertext systems appear as the practical implementation of a conceptual movement that…rejects authoritarian, “logocentric [i.e., truth-affirming] hierarchies of language, whose modes of operation are linear and deductive, and seeks instead systems of discourse that admit a plurality of meanings where the operative modes are hypothesis and interpretive play (1).

Like the constantly shifting center of the garden rhizome, a hypertext web shifts and grows irregularly, both for the reader who discovers new textual threads, and for the writer who creates them.These textual threads, this rhizomatic growth is akin to human expression that becomes fanned by new interests and new technologies and new voices made possible by that technology.Vannevar Bush envisioned this with his Memex.But Bush was hardly the first.

Bakhtin believes that human expression is the result of a multitude of voices linked in socially constructed webs of meaning, a rhizome of voices perhaps.Bakhtin would appreciate the overtly polyphonic, multivocal aspect of hypertext, not just in the “hidden” ways that Bakhtin refers to as a “borrowed light” but because of the hypertext link that connects bodies of text.And perhaps Bakhtin would be most charmed by the fact that these bodies of text are not linked along a central or primary axis. Ideally, there is no center to a hypertext web, only an opening to the web.The center shifts with each reader, and with each reader’s reading. The axis shifts with each new screen that is activated by a reader whose linking constantly resituates text. And in some cases a reader may add to a web, a docuverse, either by inserting an annotation, a response, a new piece of information into an existing web, or by linking to that web.This ability to add to a hypertext not only shifts the axis but it “blurs the boundaries of the metatext, and conventional notions of completion and a finished product do not apply to hypertext”(Landow, Convergence 58). The text becomes an instantiation of the rhizome.

Landow writes that hypertext blurs the boundaries between reader and writer and claims that, because of the nature of hypertext, the fact that the reader has to make choices and act upon those choices by clicking on a word or image, the reader becomes “active.”Perhaps it is important to point out here that all reading is active, but Landow makes an important distinction between book culture text and hypertext.Perhaps a better word to explain the role of the reader in this recenterable system is the word “deliberate.”Landow sometimes refers to a hypertext reader as intrusive.Hypertext requires the reader to make deliberate decisions about which path to take within a hypertext web, to intrude upon the author’s traditional domain, using the same powerful tools as the author. This blurring of reader/author roles holds important implications for writers and will be discussed later.

Hypertext as a Multi-Linear Dance

Much of this section will deal with hypertext reading, but the site of read hypertext presents important issues when we discuss the site of written hypertext. Murray regards electronic texts like hypertext as spaces where the writer becomes a choreographer, and the reader becomes a dancer.The reader and the writer collaborate to invoke the dance of text, “the thrill of exerting power over enticing and plastic materials” (153). And it is this concept of plasticity that is important.Hypertext is plastic in that it constantly re-centers itself according to the deliberate actions of the reader.No segment of the text takes precedence over another because the hierarchy of beginning, middle, and end do not exist.Again, hypertexts open.They do not begin.Landow reminds us that hypertext linking can center any given lexia, and that each time a reader selects a new screen or lexia, the reader has selected a new center.Landow believes this creates a:
 

…new kind of hierarchy, in which the power of the center dominates that of the infinite periphery.But because in hypertext that center is always a transient, de-centerable virtual center—one created, in other words, only by one’s act of reading that particular text—it never tyrannizes other aspects of the network in the way a print text does (Convergence 66).
Hypertext and Agency
One of the arguments this study will make is that hypertext provides an environment in which students can sense greater agency as writers. Murray believes that readers and writers of hypertext share agency.To be more precise, she believes that readers of hypertext share greater agency with the author than in book culture text.The difference may be due to the fact that by entering a computer environment, the reader physically alters the environment of the text through his or her participation because the reader inserts himself or herself simply through the act of deliberately selecting which lexias or writing spaces he or she will read. Oddly, writers who know they will share their agency with an invoked audience seem to sense greater agency in themselves.It may be that hypertext helps writers, especially student writers, gain a sense of audience.Creating book culture text, particularly book culture text produced for an audience of one, a teacher, is less likely to encourage this sense of audience. Student hypertext writers seemed to sense a great deal of control over the choices they made as writers and they were able to see the results of those choices, and base further decisions on those results.

It is important to point out again that hypertext is not the first or only textual innovation to do this. Roland Barthes, in “Death of the Author,” writes:
 

We know now that a text is not a line of words releasing a single ‘theological’ meaning but a multi-dimensional space in which a variety of writings, none of them original, blend and clash. The text is a tissue of quotations drawn from the innumerable centers of culture (116).
Though we do not generally think of hypertext as multi-dimensional, we can think of it as a tissue of text shreds and then re-integrates as each new screen is called up by a reader.And because the reader re-orders what the author created, the reader literally rewrites the text, reforms the hierarchy of the text. Again, this is not the exclusive domain of hypertext, nor is it a new phenomena.Ilana Snyder reminds us that in manuscript days scribes often altered the work they were copying.This blurred, even then, the boundaries between author and reader.Snyder adds that the tradition of print literacy privileges the author (45),although I would point out that it does not seem to privilege the student author.And certainly we must consider the situation of students as authors in a school setting when we discuss the act of composing in that environment.

Perhaps it is the “fixed” nature of book culture text that works against student-authored text in the school environment, an environment that often does not allow students to write in textbooks, that keeps students from inserting themselves “officially” in even this limited way. “Real” text is difficult to change in a valued way. It can be altered through graffiti.Or it can be added to a body of text after the fact, as happens with literary criticism assignments. But rarely does student text stand beside “real” text in a valued way. Once text is fixed, typeset, published in a book or journal, it is difficult to change.And certainly school culture bolsters the notion that text is fixed, both in meaning and in location.This is an illusion, of course. But it is a strong one. Hypertext composing in the classroom may change that.

Snyder points out that oral texts had many of the features that theorists claim are inherent in hypertexts.Oral texts could be revised at will by the speaker who altered stories depending on the prompts from an audience. But book culture text provided a new framing device for narrative and other forms (5-6).

Remember that Murray believes that in electronic text the “author” is procedural, like a choreographer “who supplies the rhythms, the context, and the set of steps that will be performed” (13). Murray says this about the reader, or as Murray calls him or her, the “interactor:”

The interactor, whether as navigator, protagonist, explorer, or builder, makes use of [a] repertoire of possible steps and rhythms to improvise a particular dance among the many, many possible dances the author enabled. We could perhaps say that the interactor is the author of a particular performance within an electronic story system, or the architect of a particular part of the virtual world, but we must distinguish this derivative authorship from the original authorship of the system itself (153). Murray is reminding us that each time a reader enters a hypertext web, the reader creates a “new” text, written by the choices he or she makes as she travels through the docuverse.This is not just the transactions that readers engage in.This is a heightened transaction, perhaps, a transaction heightened by reader choices and the simple clicking of a mouse.

Remember that collaboration is a key element in hypertext reading and writing.In a sense both the reader and the writer collaborate with the computer. But because the reader is physically required to execute some sort of command and to make a choice as to which command, the reader collaborates with the author in a deliberate and physical way. Landow believes the reader is always physically inserted in a hypertext because of the cursor that electronically marks his or her presence. We can view this as Murray does and say that the reader or dancer is collaborating with the choreographer or writer so that some version, perhaps a new one, of the author’s plan flashes onto the screen. And we can view this as Bakhtin’s constant centering self among the other(s) the text invokes.

The question again in this study is how middle school students negotiate the task of writing text that does not maintain an axis, a text that can be read in many different ways, text where the writer becomes the choreographer.

Research Methods

I began preparing for this study in the fall of 1999 when I announced to my eighth grade English students that they would be creating hypertext webs as part of their language arts curriculum that year, and that later in the year I would be asking for volunteers to participate in my dissertation study.For the next few months as I taught the students how to build web sites, I watched with more than a teacher’s eye for students who seemed to be able to articulate at length their reflections on the processes they used as they composed.I was particularly interested, for the purposes of this study, in students who had a high degree of metacognitive awareness.I was also initially looking for students who could build intricate webs.I will go into more detail about how I selected the subjects for this study in Chapter Two.

However, I will point out here that as part of my regular classroom practice, I ask students to reflect, in writing, on their composing processes.I generally prompt those reflections with a question.Often that question was merely “What have you accomplished today and why did you choose to work on that aspect of your project?”Sometimes I would ask “What is causing you difficulties and what do you think you need to do to ease those difficulties?”Or, I sometimes ask ”What frustrations and/or successes are you experiencing and why do you think you are feeling frustrated and/or successful?” Often toward the end of a project I would ask students what they would do differently if they had to start the same project over again.

The idea behind these questions was to help students develop a meta-discourse about their composing that would support the kind of reflections I think are important in literacy development.I also thought that such questions would help me decide which students would ultimately become part of my study, assuming they volunteered.And, as it turned out, getting students to volunteer for the study was not difficult. Again, I will go into more detail about this in Chapter Two.

But it is perhaps important to point out here that I had been observing students compose hypertext webs for the past three years, and what I had observed throughout the year in all the students in the class, including the initial volunteers, was also what I observed in the three subjects who are the topic of this dissertation.

This study, then, looks at the hypertext composing processes of three eighth graders at Riverton Middle School. It uses post work-session reflections written by the study subjects, post-project interviews and observations conducted as the students composed their webs to inform me as to the composing processes they used. I also use the students’ hypertext webs, as well as any pre-writing artifacts the students created.In addition to that I make note of each student’s previous hypertext composing experiences and discuss those experiences in order to provide a context for the specific project detailed in this study.All three students in the study composed two hypertext projects prior to the study project.

Approach to Research Design

I designed this study along a “subjective-objective continuum” as discussed in LeCompte and Preissle’s Ethnography and Qualitative Design in Educational Research (1993).It was necessary, of course, for me to integrate my knowledge of the classroom, the students, and the topic of the study and balance that as much as possible in order to as objectively as possible make note of the processes the students were using.LeCompte and Priessle point out that “Ethnography builds the subjective experiences of both participants and investigator into the research frame, thus providing a depth of understanding often lacking in other approaches to research” (44). I, of course, entered the research study with some initial assumptions about how students composed hypertext documents, based on my experience teaching middle school students about hypertext composing.Those assumptions were that students would, indeed, use a process in order to compose, and that they would experience heightened engagement levels.

My research was complicated by the fact that I knew my subjects very well, and I was in charge of the curriculum in the classroom.I collected my data with LeCompte and Preissle’s advice in mind, to “approach data collection as if it were the construction of a legal argument, every alternative of which had to be considered and every step and state of which had to be justified to a jury” (55).In order to do this, I observed students in the act of composing hypertext webs, I interviewed them, I initially conducted tape-recorded think-alouds that were not included in the research data on the three study subjects, and I collected student-made artifacts in the form of hypertext webs and the in-process materials that made up those webs.I also conducted informal “chats” with the study subjects as they worked.I took field notes during these brief conversations. I would ask myself “What am I seeing?”And I would also ask myself “What am I NOT seeing and what is significant about that?”

During the interviews that I set up after each study subject finished his or her study project, I made sure I asked the following questions.
 

Who was your audience for the web?
How did you decide on your topic?
Why did you set up your lexias to look the way they did?
What were you thinking about when you created your links?
How did you keep track of the textual threads in your web?
What kinds of successes and frustrations did you feel as you were composing your web?
In order to analyze the data I collected on the three study subjects, I did a simple comparison and contrast exercise where I could see both the commonalities among the three study subjects, and the areas where they different in their approaches to hypertext composing.For this, a simple three-circled venn diagram was useful in the beginning. Later, I used a table to sort out differences and similarities between the three study subjects.
In Chapter Two, I will provide further background information on the study, including why I decided to make hypertext composing part of my classroom curriculum, and I will supply more contextual information about the study and the environment in which the study was conducted.