ABSTRACT
Phase models of the text-production process basically assume that different writing activities predominate during different time periods in the process. These time periods of activities can be recognized by more or less homogeneous time-series dynamics in the data. Most of the newer models describe the writing process as incremental, increasing at every level. Farreaching decisions such as those concerning topic planning and local actions such as correcting a typographical error are possible at any time but not equally functional and therefore not equally probable at all times. Furthermore, phases can overlap each other and recur cyclically. Three types of phases of text production are strongly supported in the academic discourse on writing research and education: planning, formulation, and revision phases. They can be traced to earlier models, such as that of Hayes and Flower (1980). At that time, writing processes were investigated with simple tasks in experimental settings. However, text production in situ, for example as an activity in professional life, also includes understanding and determining the task as well as the interfaces for implementing the text product. Relevant research has been carried out by Keseling (1992) on planning in the pauses while writing and speaking; Hayes and Nash (1996) on planning while writing; Van der Geest (1996) on text planning from the perspective of professional writers; Wrobel (1997) on modeling formulation processes; Wrobel (2002) on “mental pretext,” a cognitively produced but not yet transcribed idea of a formulation; Baurmann and Ludwig (1985) and Rau (1994) for an overview of revision in the writing process; Sommers (1980) and Flower, Hayes, Carey, Schriver, and Stratman (1986) on revision strategies used by students; Van den Bergh, Rijlaarsdam, and Breetvelt (1994) on the relationship between the revision process and text quality; Severinson-Eklundh and Kollberg (2001) on revision patterns used by various authors; and Allal, Chanquoy, and Largy (2004) on revision in writing education. Writing undoubtedly is the central activity in the production of written texts. However, studies on text-production structure show that reading
phases more or less systematically accompany the entire process of text development. The type and purpose of reading vary, contingent on the stage of text development and on the related text. Depending on the related text, two types of reading processes can be differentiated: productionoriented reading, to monitor the developing text product, and sourceoriented reading, to understand texts being referred to that are written by other authors. Important research on reading processes in text production has been done by Bracewell and Frederiksen (1982); Ludwig (1983); Nelson Spivey (1990); Flower et al. (1990); Rau (1994); and Jakobs (1999). Writing and reading activities occur at all levels of text production, including (a) graphomotoric functioning, (b) formulation, (c) intermediate products, and (d) text versions. But the mode, frequency, and interplay of these activities depend on the progress in solving the emerging task. Therefore, the activities can be taken as indicators for writing phases.
