Humanising Language Teaching
Year 4; Issue 5; September 02

The Sample Approach: Teaching Writing With Cambridge Examination Classes

by Nigel Harwood
Canterbury Christ Church University College, UK

This paper argues that the three principal methods of teaching writing, the product, process, and genre-based approaches, are not mutually exclusive. Neither is adhering to just one of any of the three approaches always appropriate given a teacher's circumstances. It is claimed that in Cambridge examination classes these approaches can be combined and implemented as The Sample Approach, which developed out of the traditional product concept of using models for the students to imitate. Rather than exemplary pieces of text, however, the approach uses deliberately flawed 'samples'.

1. INTRODUCTION
2. 'REAL' WRITING & A REAL CLASSROOM SITUATION
3. THE PRODUCT & PROCESS APPROACHES
4. THE GENRE APPROACH
5. A CRITIQUE OF THE USE OF MODELS IN THE WRITING CLASS
6. MODELS & SAMPLES: THE SAMPLE APPROACH IN ACTION
7. JUSTIFYING THE SAMPLE APPROACH
8. CONCLUSION

1. INTRODUCTION

The aim of this paper is to argue that neither of the two approaches to writing which have predominated in English Language Teaching (ELT) contexts-the product and process approaches-can fully meet the needs of learners taking writing examinations such as those which are a feature of the Cambridge First Certificate, Advanced, or Proficiency papers. After providing a brief outline of the various difficulties which the teacher will typically be forced to confront when preparing an examination writing class (section 2), the tenets of the product and process approaches are summarized (section 3), before a third approach to the teaching of writing is described, which emphasizes genre and academic convention (section 4). A brief review of the literature on the use of models in the ELT writing classroom follows (section 5), before the Sample Approach is exemplified via material written for a Cambridge Advanced examination class (section 6). Section 7 defends the material against criticisms commonly levelled at taking a more conventional model-based approach, and section 8 closes with a summary and some concluding remarks.

2. 'REAL' WRITING & A REAL CLASSROOM SITUATION

Horowitz (1986: 141) recounts an exchange which took place at the TESOL '85 conference during a round table discussion on writing:

[S]peaker after speaker stressed the need for students to produce multiple drafts of papers in order to allow the process of evaluation to go forward. At one point, someone in the audience asked the panel to relate this dictum to essay examination writing. One of the members of the panel, having no ready answer, simply dismissed examination writing by claiming that it was not 'real' writing.

However 'real' we judge such writing to be, the fact of the matter is that for many ELT teachers around the world, 'helping students pass exams' is probably a fairly accurate job description. Or at least this is how the directors of the teachers' institutions-and some of their students-see their teachers' role. Since I anticipate some readers of this paper in more privileged positions objecting to this assessment, I now sketch out a typical scenario in institutions where I have worked which will hopefully illustrate the necessity of using an exams-based approach to the teaching of writing at times.

It is early-to-mid-October. On the first day of the new term, I find myself with a class consisting of students who (it rapidly transpires) have scraped a pass at First Certificate level; or, indeed, have not even taken the FCE exam in the first place. What a number of them have in common is that

(i) they are poor writers
(ii) they write as infrequently as possible in academic contexts, even in their own language
(iii) they are only acquainted with (as opposed to being comfortable with) the conventions which lie behind only a limited number of genres and text types in English (formal and informal letters and discursive compositions, perhaps).

The CAE exam is in June, or, if I am particularly unfortunate, in December, seven weeks away. Despite the fact that I already know after one or two classes I would advise a good proportion of them to defer, it will make no difference: the students have already registered for the test. Yet, according to the CAE Handbook (1998: 19) they could be asked to write any of the following:

• newspaper and magazine articles
• contributions to leaflets and brochures
• notices
• announcements
• personal notes and messages
• formal and informal letters
• reports
• reviews
• instructions
• directions
• competition entries
• information sheets
• memos

My job is to try to equip these learners for the task. Something drastic is needed!

As if all of the above were not formidable enough, consider also the constraints of the actual examination situation. Learners are given little time to respond to and write about the topic in question; and the topic itself is normally more or less predetermined, since learners are given only a limited amount of choice regarding the subject of their text. Yet as we shall see when we review the insights which the process approach has given us into the way text is composed in section 3 below, there is an inherent contradiction between the way that, on the one hand, the writer needs time to work through the recursive, convoluted writing process, and personal investment in the text to ensure the writer's voice will emerge; and on the other hand, the harsh reality of exam conditions and simply 'getting through' the test:

…[A]s test topics are rarely ones with which students become personally engaged, the concept of inquiry, of discovering meaning, becomes moot in the context of the timed writing assessment. … Similarly minimized, too, is the role of revision as an integral part of the recursive writing process. The testing situation precludes substantive additions, deletions, and shifts in organizational structure, changes which…are dependent on the amount of time students have to explore their ideas… . Test situations…allow students to make at best a few surface changes after they have hurriedly completed their essays. Thus, in most timed writings, the recursive nature of the revising process…is reduced to an occasional notation in the margin… (Wolcott 1987: 40-41)

Before I present an example of the approach which I believe at least partially resolves this contradiction, however, I discuss the product, process, and genre approaches to writing in more detail.

3. THE PRODUCT & PROCESS APPROACHES

Mindless, repetitive, anti-intellectual… The product approach to writing has been accused of all of these. In EFL contexts it was rooted in Behaviourist Theory:

The learner is not allowed to 'create' in the target language at all…[T]he use of language is the manipulation of fixed patterns;…these patterns are learned by imitation; and…not until they have been learned can originality occur… (Pincas 1962: 185-6)

The approach seemed willing to sacrifice learner motivation at the altar of this 'correctness'. When would the 'patterns' Pincas talks of ever be sufficiently mastered to allow for meaningful student involvement? The approach merely resulted in 'mindless copies of a particular organizational plan or style' (Eschholz 1980: 24); and the entire activity of writing was seen as 'an exercise in habit formation' (Silva 1990: 13).

Hairston's (1982) seminal paper pointed out further flaws in the product paradigm. Proponents of the product approach viewed the composing process as linear, proceeding 'systematically from prewriting to writing to rewriting' (Hairston 1982: 78). However, their assertions were based on intuition rather than on solid research evidence (Hairston 1982: 78), and when the actual composing process of writers was analyzed it was found to be in no way linear:

[Writing] is messy, recursive, convoluted, and uneven. Writers write, plan, revise, anticipate, and review throughout the writing process, moving back and forth among the different operations involved in writing without any apparent plan. (Hairston 1982: 85)

The product approach led students and teachers to believe that the planning stage began and ended in the initial period of composition. Yet in reality, not only did proficient writers 'rehearse' what they wanted to say before any plan was produced, but also planned throughout the writing process rather than exclusively at the start (Zamel 1983). Using a process approach, this finding was easily transferable to a classroom context where the rehearsal stage could be simulated in the form of discussions between learners, or between learners and the teacher, at any stage of composing (Johnson 1996). In addition, it was felt that the product approach, while allowing for a certain amount of revision, seriously underestimated the importance of rewriting generally. Effective revision would only result from a proper appreciation of the audience the writer was addressing (Flower and Hayes 1980) and a preoccupation with ensuring the text was reader-friendly and easy to follow. Teachers needed to cultivate a sense of responsibility in their learners for being one's own critic (White and Arndt 1991). Not only did this mean that multiple rewrites were preferable to the single rewrite which the product lesson normally limited writers to, it had far-reaching implications for the teacher's role. Whereas the teacher was concerned with grammatical accuracy in the product classroom, the preoccupation with clarity, organization, and true self-expression in the process lesson meant that the onus was now on the teacher to facilitate, rather than merely judge student writing (Johnson 1996; Killingsworth 1993).

4. THE GENRE APPROACH

An approach to writing which stresses the importance of the particular genre the student writer is attempting to approximate has much in common with what Silva (1990: 16-17) calls the 'English for academic purposes approach', where

learning to write is part of becoming socialized to the academic community-finding out what is expected and trying to approximate it. …The reader is a seasoned member of the hosting academic community who has well-developed schemata for academic discourse and clear and stable views of what is appropriate. The text is a more or less conventional response to a particular task type that falls into a recognizable genre.

Recently there have been a number of corpus-based studies of academic writing (e.g. Hyland 2000, 2001; Salager-Meyer 1994; Swales et al 1998; Tang & John 1999) which have deepened our understanding of the salient features of some of these genres, and which have also strengthened the case for taking a genre approach to writing. This is because the diversity of linguistic features and rhetorical structures from genre to genre revealed by corpus data was perhaps formerly underestimated. In short, the genre approach teaches us that readers have certain expectations about what writing in a certain genre will look like, both in terms of organization and linguistic features.

To summarize, I maintain that while an emphasis on the writing process can offer our learners many valuable lessons, exam class situations such as the one described in section 2 above require something rather distinctive. Owing to its reliance on samples and models, traditionally seen as part of the old product paradigm, together with its incorporation of a review stage associated with the process methodology, I claim this Sample Approach is an example of a 'genre/product/process' slant on writing, or what Dyer (1996: 316) has called a 'process-product hybrid'. However, there have been a number of criticisms levelled at the use of written models in the classroom which are now reviewed before I exemplify the Sample Approach.

5. A CRITIQUE OF THE USE OF MODELS IN THE WRITING CLASS

Eschholz (1980) details various criticisms of using a model approach to writing instruction:

• Models have the potential to intimidate students, since they are quite clearly of a higher quality than the class themselves can produce
• The models chosen may be inappropriate to the learners' needs in terms of length and style
• Students should be given the chance to engage in the writing process themselves before they are presented with an ideal
• The study of models can result in the learner sacrificing content to style:
  By studying forms and organizational patterns first students come to see form as a mold into which content is somehow poured…[S]tudents have no commitment to what they are writing, and care only for how they write it. (Eschholz 1980: 24)
• A model approach can mean time which could have been spent investigating the writing process is spent on reading the models
• The writing process is ignored as the finished product gets priority:
  In essence students are encouraged to know what their essays should look like before they have written them…all before writing the first sentence of what should be an exploratory rough draft. (Eschholz 1980: 25)

Similarly, Watson (1982) raises an additional concern:

• The classic 'product approach' to writing involved students more or less copying or manipulating the model in various ways: turning declaratives into interrogatives, for example. As a result, not only is the language produced patently inauthentic, but 'the risk of boredom is great' (Watson 1982: 9).

However, while both Eschholz (1980) and Watson (1982) underline the pitfalls of an approach to writing which involves the slavish aping of models, they nevertheless come down in favour of using them to teach composition, Watson (1982: 13) describing them as 'an indispensable resource'. Indeed, many well-known advocates of a process approach also accept that models can enhance writing instruction:

[I]t is important for us to preserve the best parts of earlier methods for teaching writing: the concern for style and the preservation of high standards for the written product. I believe we also need to continue giving students models of excellence to imitate. (Hairston 1982: 88) [A] place exists for a model, but of an abstract kind. The model is not to be mimicked, but is to offer a means of organizing ideas in a culturally appropriate manner. (White 1988: 12)

So it is not a question of being in one camp or the other: the product/process approaches are not irreconcilable. It is rather a question of how, not whether, to use models: how to use them effectively while avoiding the dangers Eschholz (1980) and Watson (1982) identify. I will return to these warnings later in this paper and attempt to show how I believe my materials have avoided these dangers.

6. MODELS & SAMPLES: THE SAMPLE APPROACH IN ACTION

Before providing material which typifies my approach, I should first differentiate between models and what I call 'samples'. By models, I mean 'ideal' compositions which the students would do well to utilize and exploit for the purposes of their own work. These contain no known generic or structural 'errors'. However, I prefer a more critical, discovery-led approach initially focusing on various samples: that is, specially-prepared texts composed by the teacher of, or supposedly of, a certain genre for the class to evaluate, which intentionally fall foul of what I have found from marking students' scripts to be some of the more common exam 'pitfalls'.

Hence Figures 1-4 below consist of a typical Cambridge Advanced Examination (CAE) question; an intentionally flawed sample text I have written in response to this question; worksheets which ask the learners to evaluate and identify the strengths and weaknesses of my sample text; and teacher's notes which summarize briefly the sample text's main strengths and weaknesses.

The sample text (Figure 2) is of the brief note or message genre the CAE candidate is expected to have mastered.

Figure 1: Typical CAE Notes/Messages Question

You are studying English at your local college. As part of the course you have to do a project on the different types of English spoken around the world. You have received three notes connected with your project. Read these notes carefully, and then, using the information in the notes, write the letters listed below.

Now write
(a) a note to the English library (write about 50 words)
(b) a note to Jane (write about 50 words)
(c) a letter to Chris Smith (write about 150 words)

COMMENT:

As the gap between the instructions and the exam questions indicates, the 3 letters the student is required to read are missing. If the reader is to make sense of what is happening, these 3 texts need to be read. The texts can be found on PAGE 9 of the correct version of the paper I've attached with this email.

Figure 2 below comprises the intentionally flawed sample text I composed in response to question (b) in Figure 1 above:

Figure 2: Sample note, written in response to part (b) of Figure 1 above

Flat 132,
Student Halls,
St John's College,
Oxford.

3rd March 2002.

PS
Oh-I've only got a few weeks left-so I'd appreciate it if he'd get a move on!

105 words

Figure 3 asks learners to grade the sample text in Figure 2:

Figure 3: Students' tasks centred on the sample note

i) Discuss the strengths and weaknesses of the note in groups, then mark it out of 5 according to the categories in the table below.

/5
APPROPRIACY OF STYLE
ORGANIZATION
LENGTH
RELEVANCE: DOES IT ANSWER THE QUESTION?

ii) Now state the main strengths and weaknesses of the note:

MAIN STRENGTHS

MAIN WEAKNESSES

Finally, Figure 4 provides details of what are, in my view, the main strengths and weaknesses of the sample text. Once the learners have attempted to evaluate the sample and identified the text's strengths and weaknesses individually and in groups, the teacher could use the notes below to ensure learners have at least identified the most salient points:

Figure 4: Brief teacher's notes on the tasks

MAIN STRENGTHS