Concerns about construct validity, the level to which a test reflects the theory underlying the test (Richards & Schmidt, 1992), in written responses to test items have existed since at least the year 681CE some 75 years after the introduction of the Keju, or Chinese Imperial civil service exam system, when the format of the test was changed to one that was believed to test originality of thought and content knowledge rather than rote learning (Suen & Yu, 2006). However, attempts to consider the nature of, and debate about, written discourse can be traced back a further 1000 years to the work of Aristotle and his analysis of rhetoric (Lord, 1981).
Whilst continuing doubts about what the writing process entailed and concerns about what could be infered from samples of students' writing have persisted, numerous attempts to define the construct have been put forward. However, Freedman, et al. still felt able to state as late as 1983 that the study of writing had "the aura ... of a discipline in its early stages of growth" (p.1). 10 years later, Wood (1993) stated:
"There is no generally consistent theory of what constitutes language ability, including writing ability. Typically, the views on which aspects of writing skills should be measured, and the best ways of measuring them tend to differ from one study to another" (p.53)
This was a view echoed by Purves (1992) in his analysis of the methods used in first language writing assessment in a number of countries which showed that there was little agreement in what constituted good writing. Around the same time, Grabe & Kaplan (1996) claimed that writing pedagogy was not informed by an agreed theory of what learners needed to know about the writing process and what they needed to be able to do, or what social factors affected the selection of appropriate writing skills.
Whilst there is little agreement in the literature on what writing consists of, a broad historical overview of the influences that have shaped the nature of writing might be given as follows. As mentioned above, the first attempts to consider the nature of writing are usually attributed to Aristotle and his contemporary Isocrates. Their analysis of rhetoric, both spoken and written, saw it as the ability to communicate "the right thing at the right time to solve a public problem [which the composer is able to do] because they know how to put the shared beliefs and values of the community into practice"(Miller, 1991, p. 57).That is, the Aristotelian idea was that writing was rooted in the conventions of human relationships and as such required a consideration of the ideas of the writer and the reader (Freedman, et al., 1983). The ideas of Aristotle dominated the conception of writing till at least the 16th century and the ideas of Petrus Ramus regarding the separation of the study of argumentation from the study of rhetoric and its placing in the formal study of logic (Sharratt, 1991).
The idea that started to form from this point and held sway till the 1960's has been termed the current-traditional rhetoric (Matsuda, 2003).It can be seen as deriving ultimately from the Romantic notion that the process of writing was "mysterious, inscrutable and hence unteachable" (Freedman, et al., 1983, p. 2) and that the only thing that could be analysed was the mechanics of "style, organization and usage". The idea focused on the written product and the assumption that language forms perform certain functions and that writing involves the appropriate selection of functions to achieve the purpose of the text. The process that the writer has to engage in is one of identifying the functions necessary to fulfil the writing task and arranging the linguistic realisations of these functions to form the text. Pedagogy based on this influence typically involves a focus on 'paragraph models, grammar and usage rules' (Grabe & Kaplan, 1996, p. 31) and, whilst not just focusing on the sentence level, sees paragraphs and larger text structures as functional units that need to be constructed and arranged in similar ways to the construction of sentences. The current-traditional view of writing affects much that still occurs in writing classes particularly at the university level and most textbooks still make use of its view of writing to some extent (Hyland, 2004).
A realisation that the current-traditional beliefs did not take account of the processes that writers engage in and a desire for a more humanist evaluation of the act of writing (Freedman, et al., 1983) led to what has been called the 'tacit tradition'(Freedman, et al., 1983) and, certainly in its classroom application, the process approach. This can be seen to some extent as a return to the ancient tradition particularly in its borrowing of much from the core values of the Aristotelian tradition. These include its acceptance of the importance of the inventive process of the writer, the moral code of the writer, the need for a logical basis of the text and the need to fully respond to the audience and contexts in which the text is written and read (Freedman, et al., 1983; Miller, 1991).
It added to this earlier tradition, though, by its use of the analysis of language from other disciplines such as sociolinguistics and creativity theory (Freedman, et al., 1983). This synthesis resulted in it seeing the production of written text as the traversing of a plane, as represented in figure 1, bounded by the role and demands of the encoder (or the writer), the decoder (or the reader), and the reality being represented by the text itself. The act of writing is thus seen as a process of balancing the demands of each element to create a piece of communication that most closely satisfies the goals of the communicative act. This process "is a non-linear, exploratory, and generative process whereby writers discover and reformulate their ideas as they attempt to approximate meaning" (Zamel, 1983, p. 167).
The tacit traditional rhetoric was also seen as differing from previous analyses by its attempts to base its assumptions on research (Freedman, et al., 1983). One of the most influential pieces of research was that done by Bereiter and Scardamia (1987). They argued that writing needs to be viewed not as a unitary process followed by both novice and expert writers but as a set of "different processing models [followed by writers] at different development stages of writing" (Grabe & Kaplan, 1996, p. 117).
Their 'high road' and 'low road analogy to the writing process, or journey from blank page to communicative act, saw writing as being the object of one of two sets of cognitive processes for the construction of a text. In what they termed knowledge-telling, writing is the getting down on paper of the necessary facts to satisfy the initial goals identified by the writer and may result simply in a listing of ideas with little attempt at linking them to the neighbouring ideas or the text as a whole. However, in what they termed knowledge-transferring, writing is the active manipulation of ideas and the initial goals that may result in either or both of them changing as the text develops (Hyland, 2009).This active manipulation can be envisioned as a dialogue between the writer and the elements of the plane in figure 1. As Wertsch (1998) states:
In struggling with each paragraph, one is immersed in endless dialogic encounters with the voices of others, voices of one's earlier self, speech genres, previous and subsequent paragraphs ... (Wertsch, 1998, p. vii)
In the mid-1980's concerns started to appear about how the tacit tradition neglected the fact that the writer is operating in a wider social setting (Matsuda, 2003). It was felt that insufficient consideration had been given to the cultural conventions that limit the way the writer could construct a text to create meaning. The notion of an individual's "textual ownership", it was also felt, had been overplayed and that writing was not the application of a "decontextualised set of skills" done in isolation. Instead writing is seen as occurring in a social system in which the degrees of freedom open to the elements in fig. 1 are more limited than the tacit tradition allowed for. This suggests the need for writers to be socialised into the accepted conventions and for writing teachers to guide students towards these conventions.
This post-process approach to writing is still on the periphery as far as classroom practice is concerned. It is possible that it may gain ground, though, as it has been welcomed by some teachers who have concerns over the sometimes directionless nature of the process approach's desire for student-centred self-expression (Matsuda, 2003). Others, however, see a danger in the post-process approach of a step backwards to a situation where the manner of creating a text is determined by the uncritical presentation of models by the teacher to the students: a fear illustrated by the quote from a teacher given by Brooks (2009, p.90) "I suspect we're stuck in the realm of 'recurrent-traditional rhetoric'".
Bereiter, C., & Scardamalia, M. (1987). The Psychology of the Written Composition. Hillsdale: Lawrence Erlbaum Associates.
Brooks, R. C. (2009). Historicizing Critiques of Procedural Knowledge: Richard Weaver, Maxine Hairston, and Post-Process Theory. College Composition and Communication , 61 (1), 90-106.
Freedman, A., Pringle, I., & Yalden, J. (1983). Learning to Write: First Language/Second Language. Harlow: Longman.
Grabe, W., & Kaplan, R. B. (1996). Theory and Practice of Writing. Harlow: Longman.
Hyland, K. (2004). Second Language Writing. Cambridge: Cambrdige University Press.
Hyland, K. (2009). Teaching and Researching Writing (2nd ed.). Harlow: Pearson Education Limited.
Kinneavy, J. L. (1983). A Pluralistic Synthesis of Four Contemporary Models for Teaching Composition. In A. Freedman, I. Pringle, & J. Yalden (Eds.), Learning to Write: First Language/Second Language (pp. 121-138). Harlow: Longman.
Lord, C. (1981). The Intention of Aristotle's 'Rhetoric'. Hermes , 3, 326-339.
Matsuda, P. K. (2003). Process and post-process: A discursive history. Journal of Second Language Writing , 12, 65-83.
Miller, T. P. (1991). Treating Professional Writing as Social Praxis. Journal of Advanced Composition , 11 (1), 57-72.
Purves, A. C. (1992). Reflections on Research and Assessment in Written Composition. Research in the Teaching of English , 26 (1), 108-122.
Richards, J. C., & Schmidt, R. (1992). Longman Dictionary of Language Teaching and Applied Linguistics. London: Longman.
Sharratt, P. (1991). Introduction: Ramus, Perelman and Argumentation, a way through the wood. Argumentation , 5 (4), 335-345.
Suen, H. K., & Yu, L. (2006). Chronic Consequences of High-Stakes Testing? Lessons from the Chinese Civil Service Exam. Comparative Education Review , 50 (1), 45-65.
Wertsch, J. V. (1998). Mind as Action. New York: Oxford University Press.
Wood, R. (1993). Assessment and Testing: A Survey of Research. Cambridge: Cambridge University Press.
Zamel, V. (1983). The Composing Processes of Advanced ESL Students: Six Case Studies. TESOL Quarterly , 17 (2), 165-187.
First published: 2/4/15 updated: 3/4/15