Writing Technology:
Studies on the
Materiality of Literacy

By Christina Haas

Chapter 4
Materiality and Thinking:
Effects of Computer Technology on Writer's Planning

Having reported results of her studies into computer users' problems with on-screen reading and revising, Haas next tackles problems with planning that writers experienced when composing on the computer. In a think-aloud protocol study conducted at a technological university in the 1980s, Haas randomly selected 10 experienced and 10 student writers and had them compose essays in each of three conditions -- pen-and-paper, word processing alone, and a combination of the two.

In designing this study, Haas was extremely thorough, taking pains, for example, to exclude confounding variables by conducting pretests of the users' expertise with the text processing program and by training subjects in protocol methods. This study yielded four results:

Haas believes that results of this study on planning should provide little surprise to writers who have used both computers and pen-and-paper. What is surprising to her, however, is that current composition theory fails to predict such results because it generally sidesteps the relationship between material tools and mental processes. In ignoring this relationship, composition theory has implicitly bought into the myth of technological transparency: "an assumption that the technologies of writing have no impact on the mental processes or cultural functions of writing and that writing is writing is writing, regardless of the writing medium" (p. 114).

Haas believes that the computer system used in her study specifically supported certain kinds of writing processes, such as the production of text, but failed to support others, such as the "creation of diagrammatic, conceptual notes" that writers often made with pen and paper. Such criticisms are valid for most word processing programs on the market, but Haas takes little notice of how other writing technologies support "mapping" and other traditional heuristic techniques of planning. Nowhere in her discussion does she mention how hypertext-based systems might support such planning. One such system is the UNC Collaboratory's Writing Environment , which allows authors to map their evolving text in hypertextual nodes, but still supports linear representation of the final text (see Smith, 1994). Still, most word processing programs today fail to support such planning activities, and until they do, pen and paper may be the best medium for many writers to discover their topic and plan out their writing task.

Return to Front Node / Book Table of Contents