chapter 2 homechapter 2 introductionreaders forumwriting projectsrelated links
home instructor resourcescall for essays composition resourcesdocumentation adviceglossaryfeedback

Reading #5
The Computer as a New Writing Space
By Jay David Bolter
guidepost | discussion questions | thinking critically | writing topic

Consider this simple example of electronic writing. The text is a continuous prose paragraph, displayed on the computer screen for the reader to read in the traditional way. Some of the words are in boldface; the style indicates that there is a note on that word or phrase, something more to be said. To retrieve the note, the reader points with the cursor at the text in boldface and presses a button. A second window then opens on the screen and presents a new paragraph for the reader to consider. The reader examines the note and may then return to the original paragraph.


In one sense this is simply the electronic equivalent of the footnote used in printed books for hundreds of years. Instead of looking to the bottom of the page or the end of the book, the reader aims the cursor and the computer retrieves and displays the reference. The machine is merely handling the mechanics of reading footnotes. But there is this important difference: the second window can also contain boldface phrases that in turn lead the reader to other paragraphs. The process can continue indefinitely as the reader moves from one window to another through a space of paragraphs. The second paragraph is not necessarily subordinate to the first. A phrase in boldface may lead the reader to a longer, more elaborate paragraph. One paragraph may be linked to many and serve in turn as the destination for links from many others. In a printed book, it would be intolerably pedantic to write footnotes to footnotes. But in the computer, writing in layers is quite natural, and reading the layers is effortless. All the individual paragraphs may be of equal importance in the whole text, which then becomes a network of interconnected writings. The network is designed by the author to be explored by the reader in precisely this peripatetic fashion.


Such a network is called a hypertext,and, as we shall see, it is the ability to create and present hypertextual structures that makes the computer a revolution in writing. The computer as hypertext invites us to write with symbols that have both an intrinsic and extrinsic significance. That is, the symbols have a meaning that may be explained in words, but they also have meaning as elements in a larger structure of verbal gestures, Both words and structures are visible, writeable, and readable in the electronic space.

Writing Places


With or without the computer, whenever we write, we write topically. We conceive of our text as a set of verbal gestures, large and small. To write is to do things with topics--to add, delete, and arrange them. The computer changes the nature of writing simply by giving visual expression to our acts of conceiving and manipulating topics. A writer working with a word processor spends much of the time entering words letter by letter, just as he or she does at a typewriter. Revising is a different matter. With most word processors, writers can delete or replace an entire word; they can highlight phrases, sentences, or paragraphs. They can erase a sentence with a single keystroke; they can select a paragraph, cut it from its current location, and insert it elsewhere, even into another document. In using these facilities, the writer is thinking and writing in terms of verbal units or topics, whose meaning transcends their constituent words. The Greek word topos meant literally a place, and ancient rhetoric used the word to refer to commonplaces, conventional units or methods of thought. In the Renaissance, topics became headings that could be used to organize any field of knowledge, and these headings were often set out in elaborate diagrams. (See Ong, 1958, pp. 104-130.)* Our English word topic is appropriate for the computer because its etymology suggests the spatial character of electronic writing: topics exist in a writing space that is not only a visual surface but also a data structure in the computer. The programmers who designed word processors recognized the importance of topical writing, when they gave us operations for adding or deleting sentences and paragraphs as units. They did not, however, take the further step of allowing a writer to associate a name or a visual symbol with such topical units. This important step lends the unit a conceptual identity. The unit symbol becomes an abiding element in the writer's thinking and expression, because its constituent words or phrases can be put out of sight.

*Editor's note: Bibliography for parenthetical references not provided here.


On a printed or typed page, we indent and separate paragraphs to indicate the topical structure. Within each paragraph, however, we have only punctuation, occurring in the stream of words, to mark finer structure. A better representation of topical writing is the conventional outline, in which major topics are designated by Roman numerals, subtopics by capital letters, subsubtopics by Arabic numerals, and so on. Each point of an outline serves to organize and situate the topics subordinate to it, and the outline as a whole is a static representation, a snapshot, of the textual organization. The conventions of outlining turn the writing surface into a tiered space in which the numbering and indentation of lines represent the hierarchy of the author's ideas. A paragraphed text is the flattening or linearization of an outline.


The word processor, which imitates the layout of the typed page, also flattens the text. It offers the writer little help in conceiving the evolving of the text. Although the word processor allows the writer to define a verbal unit in order to move or delete it, the definition lasts only until the operation is complete. But if the word processor offers the writer only temporary access to his or her structure, another class of programs called outline processors makes structure a permanent feature of the text. An outline processor sets the traditional written outline in motion. A writer can add points to an electronic outline in any order, while the computer continually renumbers to reflect additions or deletions. The writer can promote minor points to major ones, and the computer will again renumber. The writer can collapse the outline in order to see only those points above a certain level, an action that gives an overview of the evolving text. In short the writer can think globally about the text: one can treat topics as unitary symbols and write with those symbols, just as in a word processor one writes with words.


Writing in topics is not a replacement for writing with words; the writer must eventually attend to the details of his of her prose. The outline processor contains within it a conventional word processor, so that the writer can attach text to each of the points in the outline. But in using an outline processor, writers are not aware of a rigid distinction between outlining and prose writing: they move easily back and forth between structure and prose. What is new is that the points of the outline become functional elements in the text, because when the points move, the words move with them. In this way the computer makes visible and almost palpable what writers have always known: that the identifying and arranging of topics is itself an act of writing. Outline processing is writing at a different grain, a replication on a higher level of the conventional act of writing by choosing and arranging words. The symbols of this higher writing are simply longer and more complicated "words," verbal gestures that may be whole sentences or paragraphs.


In an outline processor, then, the prose remains, but it is encased in a formally operative structure. With a pen or typewriter, writing meant literally to form letters on a page, figuratively to create verbal structures. In an electronic writing system, the figurative process becomes a literal act. By defining topical symbols, the writer can, like the programmer or the mathematician, abstract himself or herself temporarily from the details of the prose, and the value of this abstraction lies in seeing more clearly the structural skeleton of the text. It is not possible or desirable that the prose writer should become a mathematician or that human language should be reduced to a system of logical symbols. The result of giving language wholeheartedly over to formalism would simply be the impoverishment of language. On the other hand, the electronic medium can permit us to play creatively with formal structures in our writing without abandoning the richness of natural language....



The goal of conventional writing is to create a perfect hierarchy, but it is not always easy to maintain the discipline of such a structure. All writers have had the experience of being overwhelmed with ideas as they write. The act of writing itself releases a flood of thoughts--one idea suggesting another and another, as the writer struggles to get them down in some form before they slip from his or her conscious grasp. "I only wish I could write with both hands," noted Saint Teresa, "so as not to forget one thing while I am saying another" (see Peers, 1972, vol. 2, p. 88). Romantics like Carlyle founded their psychology of literature upon this experience. The experience is not limited to saints and poets: many, perhaps most, writers begin their work with a jumble of verbal ideas and only a vague sense of how these ideas will fit together. The writer may start by laying out topics in an arrangement less formal than an outline: he or she may organize by association rather than strict subordination. Teachers of writing often encourage their students to begin by sketching out topics and connecting them through lines of association, and they call this activity "prewriting." What students create in prewriting is a network of elements--exactly what computer programmers mean by the data structure they call a network. The computer can maintain such a network of topics, and it can reflect the writer's progress as he or she trims the network by removing connections and establishing subordination until there is a strict hierarchy. In the world of print, at least in nonfiction, associative writing is considered only a preliminary.


Association is not really prior to writing, as the term "prewriting" suggests. Association is always present in any text: one word echoes another; one sentence or paragraph recalls others earlier in the text and looks forward to still others. A writer cannot help but write associatively: even if he or she begins with and remains faithful to an outline, the result is always a network of verbal elements. The hierarchy (in the form of paragraphs, sections, and chapters) is an attempt to impose order on verbal ideas that are always prone to subvert that order. The associative relationships define alternative organizations that lie beneath the order of pages and chapters that a printed text presents to the world. These alternatives constitute subversive texts-behind-the-text.


Previous technologies of writing, which could not easily accommodate such alternatives, tended to ignore them. The ancient papyrus roll was strongly linear in its presentation of text. The codex, especially in the later Middle Ages, and then the printed book have made better efforts to accommodate association as well as hierarchy. In a modern book the table of contents (listing chapters and sometimes sections) defines the hierarchy, while the indices record associative lines of thought that permeate the text. An index permits the reader to locate passages that share the same word, phrase, or subject and so associates passages that may be widely separated in the pagination of the book. In one sense the index defines other books that could be constructed from the materials at hand, other themes that the author could have formed into an analytical narrative, and so invites the reader to read the book in alternative ways. An index transforms a book from a tree into a network, offering multiplicity in place of a single order of paragraphs and pages. There need not be any privileged element in a network, as there always is in a tree, no single topic that dominates all others. Instead of strict subordination, we have paths that weave their way through the textual space.... If all texts are ultimately networks of verbal elements, the computer is the first medium that can record and present these networks to writers and readers. Just as the outline processor treats text as a hierarchy, other computer programs can fashion the text into a general network or hypertext....

Writers and Readers of Hypertext


When we receive a written or typed letter, we hold in our own hands the paper that the sender also has handled. We see and touch the inkmarks that he or she has made. With electronic mail we receive bits of information that correspond to the tapping of keys on the writer's keyboard. We read this information as patches of light on our computer screen, and we touch nothing that the writer has touched. Like all other kinds of writing, electronic writing is an act of postponement or deferral. As writers, we defer our words by setting them down on a writing surface for later reading by ourselves or by others. The reader's task is to reactivate the words on the page and to devise for them a new context, which may be close to or far removed from the author's original context. There is always a gulf between author and reader, a gap that the technique of writing first creates and then mediates. In one sense the computer opens a particularly wide gap because of the abstract nature of electronic technology. On the other hand, the author has a unique opportunity to control the procedure of reading, because he or she can program restrictions into the text itself.


Computer-assisted instruction, for example, is nothing other than a hypertext in which the author has restricted the ways in which the student/reader can proceed. In typical computer-assisted instruction the program poses a question and awaits an answer from the student. If the student gives the correct answer, the program may present another question. If the student's answer is wrong, the program may explain the student's error. If the student makes the same error repeatedly, the program may present a review of the point that the student has failed to grasp. In most cases, these questions and explanations are texts that the teacher/programmer has composed and stored in advance. However, good programming can make these simple programs seem uncannily clever in replying to the student. In fact such a program takes on a persona created for it by the teacher/programmer, as it transfers the teacher's words into the new context of the student's learning session. In general, the reader of an electronic text 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. If the program allows the reader to make changes in the text or to add new connections (as some hypertext systems do), then the game becomes still more complex. As readers we become our own authors, determining the structure of the text for the next reader, or perhaps for ourselves in our next reading.


Electronic text is the first text in which the elements of meaning, of structure, and of visual display are fundamentally unstable. Unlike the printing press or the medieval codex, the computer does not require that any aspect of writing be determined in advance for the whole life of a text. This restlessness is inherent in a technology that records information by collecting for fractions of a second evanescent electrons at tiny junctions of silicon and metal. All information, all data, in the computer world is a kind of controlled movement, and so the natural inclination of computer writing is to change, to grow, and finally to disappear. Nor is it surprising that these constant motions place electronic writing in a kaleidoscope of relationships with the earlier technologies of typewriting, printing, and handwriting.

Bolter, Jay David, Writing Space: The Computer, Hypertext, and the History of Writing. Copyright 1991 Lawrence Erlbaum Associates. Reprinted with permission.


chapter 2 home | introduction | readers forum | writing projects | related links
home | instructor resources | call for student essays | composition resources
documentation advice | glossary | feedback

Site Map I Partners I Press Releases I Company Home I Contact Us
Copyright Houghton Mifflin Company. All Rights Reserved.
Terms and Conditions of Use, Privacy Statement, and Trademark Information