Redefining Literacy:
The Multilayered Grammars of Computers

by Cynthia Selfe

It is old news by now to note that computers have revolutionized the nature of written communication and, thus the nature of literacy. We take for granted the fact that increasing numbers of writers and readers are using computers to generate, manipulate, share and store written text and graphics.

But it may be news to contend that the real work of developing computers as aids to literacy has just begun. If the most reluctant among us now accept that computers have had an impact on literacy, not even the most enthusiastic claim to know just what that impact has been or how we should deal with it in our teaching. In fact, we are just beginning to get an idea of how radically our definition of literacy changes when communication activities are mediated by computers. Until we better envision this computer-mediated literacy--and our vision must accommodate reading, writing and technology--our profession can hardly expect to provide students with the skills they need to function as literate members of our technologically supported society.

Nowhere is there greater potential for achieving a degree of success in this exceedingly complex venture than there is in the numerous computer-supported writing centers that have been set up in schools and institutions across the country. These nontraditional communication environments are valuable to us because they are sites in which language production and learning occur consistently within a technologically dependent context. Thus they can serve as laboratories in which we inform and update our notion of literacy as it is practiced in computer-supported communication environments. In these centers, we begin to gather the data that will eventually allow us to construct a new vision of literacy as it has been, and continues to be, affected by computers.

To date, I should add, we have only begun this data gathering and have had time only to identify the barest outlines of this new vision. Nor is rapid progress probable. Teachers and researchers trying to define the changing nature of literacy engage in a complex version of old connect-the-dots game--one in which the players never see the entire picture they are trying to construct because it is still in the process of being formed. Often in this game our perspective is limited; each one of us sees only one or two of the dots and must experiment with different methods of connecting them. But even the sketchy and incomplete representations we have been able to put together are exciting enough to warrant our continued and increased attention.

We can progress more rapidly and effectively than we now do in our attempts to define literacy as it relates to computer-supported communication. To do so, we need to share information about the dots each one of us has already connected, to exchange educated guesses about the final form that the picture is taking. In this chapter, I work toward that end, sharing some preliminary observations about the changing nature of literacy and offering a rough sketch of what literacy may now involve within the context of computer-supported literacy communication environments.

TRADITIONAL LITERACY

To understand how computers have altered the nature of literacy in the past decade, we need to begin with a broad baseline definition of that concept. And, if we agree for the purpose of this discussion to focus on print literacy rather than, for example, visual literacy or cinematic literacy, this definition need not be overly complex. Most of us would agree that literacy involves both reading and writing, and concerns the way in which human beings make meaning from printed texts by interpreting content in light of their own purposes and needs.

We might also add to this definition that the responsibility for this process of constructing meaning cuts both ways in a literate culture, that it involves both readers and writers of text. Indeed, we know that literate individuals collaborate, in a sense, to make meaning, engaging in unique kind of asynchronous, two-sided interpretation of personal experience, even when one party never meets the other. Finally, we might add that literacy training involves teaching individuals the shared system of conventions associated with reading and writing.

I will focus on this last point about conventions because it pinpoints the crucial reason why computers affect literacy so immediately. By conventions I mean those standard rules, variously referred to as "grammars" by Gumpert and Cathcart (1985) or as "formats" by Altheide (1985), which govern such things as arrangement, structure, form, and appearance of text. Armed with these rules, individuals can predict some things about the basic nature of texts that they read and write.

Let's consider the grammar of the page, for instance. In western cultures, a page is read starting in the upper left-hand corner and proceeding to the lower right-hand corner in a line-by-line fashion. It is composed of a predictable set of structural components, consisting of letters that combine into words, words that combine into sentences, and sentences that add up to paragraphs, pages, sections, and chapters.

Pages have a number of formal conventions as well: running heads, numbers, standard margins on the sides and at the top and bottom, footnotes, titles, subtitles, or headings. Finally, pages have physical conventions: they are made primarily of paper and bound so that they can be quickly flipped through; their aspect ratio (ratio of width to height) is approximately 2 to 3. Within individual books, pages are a fixed size, arranged in a fixed order, and thus represent static structural units of text that do not change from reading to reading. No matter how many times we read a copy of a book, a given piece of material will be on the same page in the same location; the page will be in the same order in the book. Pages, in this sense, are immutable. In fact, we use this particular characteristic of pages to locate information within the spatial context of a text. If we want to locate information again in the future, we note its spatial location, referring to a particular page by number. In sum, these grammatical conventions that we learn during our literacy training allow us to "cope" (Strassman, 1983) with the world of print, to anticipate the characteristics that printed texts share, and thus to use these texts efficiently.

The significance of these conventions has been explored by scholars like Gumpert and Cathcart (1985), McLuhan (1967) and Ong (1982) who claim that the conventions associated with specific media determine not only how we see text but how we view the world as a whole and how we construct our reality. As evidence, Ong (1982) cites, among other examples, Goody's (1977) research on indexes to support his argument that until humans used print to represent their thinking in an essentially linear format on a collection of pages, they had no conception of retrieving a piece of information by identifying its spatial location. Thus, Ong suggests, when we invented indexes to help us find written information efficiently within such a context, we also discovered the essential concept of "indexing," the notion of cataloguing and retrieving information by recording its location in space. Because indexes have no equivalent in the natural world, until we created the print medium and the convention of "index," we had no way of envisioning, or even thinking of, systematic information storage and retrieval based on spatial location.

For another example of this important concept, we can look to film and television. Recent work with these media suggests that humans observe media conventions, for example, slow-motion, split-screens, and objects revo lving in space; that they internalize them; and that they later apply these conventions as "tools of thought" to problems unconnected with the medium itself. For instance, in a study that concentrated on the "zoom shot" as it has been defined by the film and television industry, Salomon (1972) found that viewers who watched the technique of "zooming-in" on film could internalize the convention and then apply it as a problem-solving strategy in other situations. The medium (television/video) and a convention of the medium (the zoom shot), in other words, provided subjects with a new way of constructing reality and meaning that they could not have acquired through observation of the natural world.

LITERACY AND COMPUTERS

If we accept that the conventions of specific media determine how we construct reality, we can begin to see how the use of computers as communication aids might affect literacy in two important ways. First, computers add several new grammars to the lists of things that individuals must learn before they become successfully literate in a computer-supported environment. We can posit grammars associated with computer keyboards and with computer screens, and grammars related to the use of computer networks or printers. These new kinds of literacy are layered over and have a substantial impact on the tasks of reading and writing. Second, computers change the way we "see" text and construct meaning from written texts. Like the concepts of "indexing" and "zooming-in," some of the conventions associated with computers do not exist in the natural world, and these conventions change the way in which we think about communication problems.

To explore both of these claims, we can look for a moment at the grammar of the computer screen. Although some conventions of the printed page carry over into, and indeed, are mimicked by the conventions of a computer screen, the principle conventions of the two media are quite different. We have noted that pages are static structural units of a longer, spatially represented text; the text on a page does not change with time. Screens do not represent structural units of text; rather, they are temporal windows on a virtual text. Virtual texts, unless they are translated into the print medium, exist only in the memories of the computer, the reader, or the writer.

Because text on a screen is a temporal rather than a structural unit, a reader of screen text lacks some spatial-contextual cues to which a reader of page text has access. A reader of a book can gauge its length in one glance and get an idea of format, organization, and arrangement by flipping through pages. In contrast, the reader of a virtual text on a collection of screens can see the whole text only in his or her mind, and must keep track of individual screens in the same way. Given this lack of spatial-contextual cues, movement through a virtual text is often considered more difficult than movement through a printed text, either slower (when scrolling is done line by line), or more erratic (when one screen dissolves instantly into another at the touch of a key).

Screens also have different formal conventions than do pages. They are not numbered; their margins are fluid and easily modified by writers and readers; the aspect ratio of screens, at 4 to 3, mimics that of a television rather than of most books. And screens have cursors, windows, and menu lines--characteristics not shared by pages.

Physical conventions also vary between the two media: unlike pages, screens are made of glass or plastic; unlike pages, screens must be activated by electric power sources. The text on a screen consists of phosphorescent pixels that are dynamic in their formation and in their display; text on a page, printed in ink, is static. On a screen, writers and readers can add and delete text, change margins and spacing, construct and reconstruct text physically; the text exists only in a fluid form. On a page, text is immutable.

THE EFFECTS OF LAYERED LITERACY

At this point, we have identified two very different sets of conventions that individuals must learn if they hope to function literately within computer-supported communication environments: conventions of the page and conventions of the screen. This information suggests two concerns: How much of an impact do these different grammatical conventions have on individuals? Can people who have acquired literacy in the print medium acquire comparable computer-based literacy skills? This process of layering literacy, stacking in effect one grammar on top of another, has a profound impact in that individuals making the transition from page text to screen text must change the ways they read, write, and make meaning from text.

There are, of course, highly charged political problems that accompany our culture's transition from traditional print literacy to a multilayered, computer-based literacy, and at the heart of these problems lies the issue of equal opportunity and education. Not everyone is equally capable of acquiring facility with computer-based grammars. given the relatively recent advent of computers, and our continuing reliance on hard-copy texts, individuals over the age of seven learn the literacy conventions of the page much earlier than they do those connected with the screen; they practice these grammars more frequently, and they are more comfortable with them. As one of my students put it the other day, most people speak computers as a second language. If we carry this analogy even further, we can predict that some level of interference will be present for these people as they supplement their print literacy conventions with the literacy conventions associated with computers.

This situation becomes more disturbing when we realize how closely individuals' economic security, career success, and academic progress are tied to an ability to master computer-based grammars. Increasingly, employees, students, and teachers are required to learn computer systems quickly and under highly stressful situations. Students in many colleges, for example, are expected to acquire computer-based literacy skills on their own in order to compete in their English composition classes. The teachers of these classes, who seldom feel responsible for matters of computer-based training, often neglect to provide their students with adequate training; frequently the instructors do not understand the complex nature of the literacy task that they are assigning. The same situation is true of executives in business or industry who see computers as quick fixes to office or production problems and require employees to read and write on sophisticated computer systems without adequate training.

In some ways, our culture has already begun to cope with this multilayered literacy problem, with the fact that people must be, or become, literate consumers and producers of both page and screen text. For example, many commercial word-processing packages come equipped with translation aids that help readers and writers move back and forth easily between the grammar of the screen and that of the page. At the bottom of every screen in these packages is a ruler line or status line that gives information both about the screen text and translation of that text to hard copy. Typically, this line identifies the name of the virtual text being displayed on the screen and indicates how this text would appear if displayed on a traditional page, noting the spacing, page breaks, margins, and page numbers as they would appear in a print medium.

Similar translation devices exist in full-page, what-you-see-is-what-you-get (WYSIWIG) editors such as Radius Full Page Display for the Macintosh computer. These devices, because they mimic the aspect ratio of a printed page, can display a portion of virtual text just as it would appear in the print medium. The purpose of the Radius Display, and similar machines, is to make work with virtual texteasier by electronically simulating some of the structural and formal conventions associated with a printed page.

However much translation aids moderate the effects of multilayered literacy, though, they do not eliminate them. Individuals still must learn to cope with the conventions of the screen and the page, modifying reading and writing processes accordingly. I speculated earlier that some people may never achieve equal fluency for reading and writing in both grammars. This premise, based on compelling preliminary evidence, deserves further investigation.

We examine the case for reading first. We know from recent investigations, for example, that individuals read screen text more slowly (Hansen, Doring, & Whitlock, 1978), less accurately (Wright & Lickorish, 1983), and less effectively (Haas & Hayes, 1986) than they do text printed on a page. In these investigations, researchers noted that readers had more difficulty getting an overview of a text and found it harder to locate specific information (Haas & Hayes, 1986) on most computer screens than they did on a page. In addition, individuals proved less accurate in proofreading text on computer screens than they did on paper (Wright & Lickorish, 1983).

Although researchers have yet to determine why text on a computer screen is more difficult to read than text on a page, they suspect differences in variables such as resolution, font size, reading angle, and color may be responsible (see Haas, Chapter 2 of this volume). Another explanation, however, may be equally as credible. Given the fact that the computer screen is a relatively new medium, we might posit that these studies show the effects of a "media generation gap" in Gumpert and Cathcart's (1985) terms: individuals trained originally to deal with the grammatical conventions of a printed page have not achieved a similar fluency in coping with the grammar of a computer screen. I point out, in addition, how this generation gap might become especially political in nature, when individuals trained for much of their lives to rely on hard copy must learn, in a matter of days or weeks, coping with screen copy and must do so under highly trying and demanding circumstances.

Writing, in a similar way, provides evidence of the various layers of literacy required in computer-supported writing environments and problems involved in acquiring literacy skills during a period of transition. We can, for instance, examine the comments of this writer who continued to compose her drafts with paper and pencil, even after she had become basically competent in writing on a computer screen:

I feel I can express myself better {with paper and pencil}...like I'm in control of the situation. Maybe I'm too far away with the computer. I mean the screen is there, and I am here {gestures}. With paper and a pencil, I'm touching the words. Also they {the words} look like you wrote them, not like the machine wrote them. (Selfe, 1985, p. 57)

My words are in there {points to the computer}, but not in the same way or in the same form as they are here {points to hard copy}... To me, writing changes magically somewhere between the screen and the printer. There is a transformation that occurs. I can write on the computer, and I can write on paper--but I need both to survive. (Selfe, 1984)

THE IMPLICATIONS OF MULTILAYERED LITERACY

In order to assess accurately the implications of these observations for teachers of English, we consider again the speculative model of grammars that may be functioning within a computer-supported communication environment. Each one of these grammars represents new layers of complex conventions that individuals must learn if they hope to function successfully and literately within such an environment. As our use of computers and networks becomes more sophisticated, we will undoubtedly add additional layers of literacy in English composition classes. Some teachers of technical writing, for example, are already requiring that students acquire expertise in databases, graphics programs, and spreadsheets for use within these classes.

For English teachers, this notion of multilayered literacy suggests a wide range of implications for our definition of literacy; for the ways in which we deal with reading and writing, or readers and writers; for the manner in which we set up and operate computer-supported communication environments in support of English composition programs.

It seems reasonable to assume that we will have to deal with layers of literacy instruction, just as our students deal with layers of literacy demands. For some of these layers, I suspect, the task will be relatively easy. We will probably, for instance, continue to teach reading and writing skills much as we do now--emphasizing multiple drafts, successive refinement of ideas, peer critiques and editing, and conferences--regardless of whether writing is completed and shared on screen or in hard copy.

However, to address the succeeding layers of literacy described in this paper, we will also have to begin identifying strategies for coping with the grammars that computers add to communication tasks. On the level of screen grammar, for example, we may have to teach students new or different methods for reading texts on screen to supplement the instruction they have received in reading hard-copy texts. We may have to show them, for instance, how to move about in a virtual text: how to use search-and-replace commands to jump from section to section, how to reformat text for easier screen reading, or how to incorporate different kinds of intratextual organizational markers for readers who view texts only on a screen.

Similarly, we shall have to identify reading and writing strategies for helping students deal effectively with the conventions of word-processing packages, computer systems, and networks. The problem involves identifying and mastering these strategies ourselves, and this can be accomplished only by a collective effort. Until teachers and writers begin cataloguing and sharing the literacy strategies that they have used in computer-supported writing environments and testing them on different populations, our efforts will remain scattered and unsystematic.

We are not limited to our on resources in this effort to identify and teach effective strategies. In fact, the more I work with students in computer-supported communication environments, the less adequate I feel to undertake this task alone. I suspect that our generation, raised as we have been on the print medium, may not be capable of generating some of the more creative strategies for coping with the conventions of computer screens, computer systems, or computer networks. Our conceptual vision in these areas may be limited by the media generation to which we belong and its pencil-and-paper strategies (Selfe & Wahlstrom, 1988). We may have to turn to our own students for help, observing the literacy strategies they develop on their own for coping within computer-supported communication environments.

In support of this premise, I can describe the experience of teaching computer-supported writing courses at Michigan Tech. As part of these classes, which are frequently taught in the Center for Computer-Assisted language Instruction (CCLI), students are frequently required to write their journals or papers on our local area computer network, exchange drafts with their peers by posting them on a computerized bulletin board, confer with a teacher through an electronic mailbox, or participate in an electronic conference.

In this computer-supported writing facility, some faculty members in our Humanities Department prefer to run "paperless" classrooms--storing handouts and lecture notes on the computer and asking students to hand in their writing assignments on-line. In these classes, most reading and writing is done via the medium of the computer screen rather than the printed page. Soon after we began this practice, faculty within our program began to notice that the writing assignments students did for viewing on the computer screen were different in both form and content from the writing assignments they produced for viewing in the print medium. One of the first characteristics that became evident, for instance, was that students had learned on their own initiative to write text to be read on screen rather than on the page, thus handicapping those students who printed such text out in a print medium. In screen-based papers, for example, students incorporated flashing segments of text (light background color and dark characters rather than the usual dark background and light characters)--all conventions that do not print legibly in hard copy. In addition, students used formatting conventions that were screen-based rather than page-based, often using shorter paragraphs that could be viewed on a single screen, breaking the text into screen-sized chunks so that readers could more efficiently use the "page-up" and "page-down" keys as they read, and centering text within the window of the page.

A second, and perhaps more unusual, characteristic of screen-based texts involved the use of color as a visual cue to underlying logical content and structure. Students, for example, used three different colors to signify primary, secondary and tertiary headings; used two different colors to "paint" contrasting arguments contained within a single paragraph; used color-coding to identify thesis or topic statements and the evidence that supported these central ideas. These "painting" strategies are important because they go beyond mere decoration of a text to represent a visual revelation of logical structures. As Bernhardt (1986) notes, such visual cues heighten readers' "awareness of categories and divisions, changing the ways people conceive classificatory divisions" (p. 66). Bernhardt (1986) also maintains that teachers of writing should encourage students to "experiment with visual features of written texts" so they can "increase their ability to understand and use hierarchical and classificatory arguments.

What has become obvious to the teachers within our program is that the writers in our classes who compose their texts for reading on computer screens have invented and exploited a new set of literacy skills that their teachers never imagined. Using different fonts, font sizes, symbols, highlighting, and graphic elements, they have not only adjusted their writing to the conventions of the screen and the computer, but have also reconceptualized the content of their assignments in terms of these conventions. It seems possible that the grammar of the computer, the word-processing package, or the computer itself changed the way in which writers think and express thought.

These insights raise the issue that there may be other computer conventions that similarly affect our world view. Whether, for example, the convention of the virtual text as it exists in a computer and on screen will encourage us to develop new ways of handling large texts in our mind's eye, leading us to a new conceptual ground that exists somewhere between the print literacy we know now and a literacy of the virtual text that we have yet to define; whether students who learn the concept of windowing in a word-processing package will then apply this literacy convention in new and creative ways to other communication situations; and whether writers who learn to reformat screen texts will somehow see printed text in a different way.

In other words, our work is cut out for us. Our profession will have to work diligently in the next few years to identify and explore the changing nature of literacy within a computer-supported writing environment and to consider the implication of these changes for our teaching. We are going to have to do this job even as the computer industry continues its explosive growth. I maintain that computer-supported writing labs and classrooms may be the best site for this work because they offer us the opportunities for both research and teaching. In these facilities, we can observe the changing literacy demands of our students, gather evidence of the impact of these demands, develop teaching strategies that help our students cope with these demands, and then test these teaching strategies on a variety of populations. That job will be exciting and creative for us all.

Acknowledgments

I want to acknowledge here those intellectual debts I owe to colleagues at Michigan Technological University. First, I have to thank Billie Wahlstrom for so graciously sharing with me the electronic journal entries from her Mass Communication class and for helping me make some sense of these "new" texts. I used these journal entries to corroborate the observations I made of journal and electronic conference entries in my own computer-supported writing classes. Second, I thank Marilyn Cooper. Over a memorable dinner, she helped me to understand many important differences between grammars of the page and of the screen. (I was so excited that I couldn't eat.) The three of us are navigating by triangulation as we attempt to connect the dots of computer literacy.

REFERENCES

Altheide, D. L. (9185) Media Power Beverly Hills, CA: Sage Publications.

Bernhardt, S. A. (1986). Seeing the text. College Composition and Communication, 37 (1), 66-78.

Goody, J. (1977). The domestication of the savage mind. Cambridge, England: Cambridge University Press.

Gumpert, G., & Cathcart, R. (1985) Media grammars, generations, and media gaps.Critical Studies in Mass Communication, 2 23-35.

Haas, C., & Hayes, J. (1986) What did I say? Reading problems in writing with the machine. Research in the Teaching of English, 20 (1), 22-35.

Hansen, W. J., Doring, R., & Whitlock, L. R. (1978). Doing the same work with hard copy and with CRT terminals. International Journal of Verbal Learning and Verbal Behavior, 10, 608-613.

McLuhan, M. (1967). The Medium is the message: An inventory of effects, New York: Random House.

Ong, W. (1982). Orality and literacy: The technologizing of the word.London: Methuen.

Salomon, G. (1972) Can we affect cognitive skills through visual media? An hypothesis and initial findings AV Communication Review, 20 (4), 401-422.

Selfe, C. (1985). Computers and the composing processes of students. Unpublished raw data.

Selfe, C. (1985). The electronic pen: Computers and the composing process. In J. Collins & E. Sommers (Eds.), Writing on-line: Using Computers in the Teaching of Writing (pp. 55-66). Upper Montclair, NJ: Boynton-Cook.

Selfe, C., & Wahlstrom, B. (1988). Computers and Writing: Casting a Broader Net with Theory and Research. Computers and the Humanities, 22, 57-66.

Strassman, P. (1983). Information systems and literacy. In R.W. Bailey & R.M. Fosheim (Eds.), Literacy for life: The demand for reading and writing (pp. 115-121). New York: Modern Language Association.

Wright P., & Lickorish, A. (1983). Proof-reading texts on screen and paper. Behavior and Information Technology, 2 (3), 227-235.

This Chapter first appeared in: (Eds.) Gail E. Hawisher and Cynthia Selfe, (1989). Critical Perspectives on Computers and Composition Instruction (pp. 3-15). New York: Teachers College Press.
Converted to HTML by Ted Phillips, (Sep 1996)