This raises a number of questions for the teaching of writing in the South African context, particularly its effectiveness in preparing students for the world of work. As is the case in many countries, the teaching of writing has not had the same attention paid to it as reading instruction; many teachers are less skilled in the teaching of writing and time requirements and fair assessment remain contentious issues. Writing is a complicated activity. Successful writers need to master a range of skills. These skills include knowing the complex requirements of genres. Writers need to know the social practices in which genres are embedded and the meaning and function they have in communities. An implicit or explicit knowledge of the numerous cognitive skills and processes required to produce a piece of writing is also required. Writing relies on careful thought that draws on but is different from the processes required for reading. Thus, with the importance of writing increasing globally, and schools a key institution in which writing is taught, it is necessary to understand the effects of writing pedagogy.

The range of writing tasks and genres required across schooling as envisaged in the Curriculum and Assessment Policy Statement (CAPS) documents (Department of Basic Education [DBE] 2011) is beyond the scope of this article; therefore, the focus here is on the argumentative essay. Argument manifests itself in numerous ways in our daily lives. The rules of engagement, discourse structures and topics that result in differences of opinion vary across communities, within and between age groups and professions. The importance of being able to present and evaluate argument is captured globally in many curriculum documents; CAPS claims that students who can construct and evaluate arguments demonstrate an ability to be thoughtful and reasonable (DBE 2011:8–9, 11).

But, as Newell et al. (2011:297) argue, the view that argumentative reading and writing lead to ‘reasonableness and thoughtful consideration of a topic’ is a false assumption. In fact, little is known about how these skills develop over time. They argue that whilst argumentative reasoning is emphasised, there is little work that addresses the methods teachers use to develop students’ argumentative writing. This gap is the starting point for this article.

There is also little research on writing in schools in the South African context. Research that has been conducted draws attention to the paucity and poor quality of writing and writing pedagogy in English as a first additional language (FAL) (Hendricks 2008) and as a home language in primary school classrooms (Gains & Graham 2011). Using a Vygotskian perspective Mendelowitz (in press) argues that CAPS’s overemphasis of a ‘back to basics’ discourse undermines creative writing. Her examination of Grade 7 teachers reveals how their conceptualisations and enactment of creativity and imagination enable and constrain children’s ability to harness the higher order skill of creativity in their writing. In her consideration of current writing research in South Africa, Dornbrack (n.d.) has called for a more explicit focus on writing pedagogy.

Although much writing research focuses on the processes of writing or classroom practices, our focus is not on what the teacher has done in the class but rather on student take-up. An analysis of a Grade 10 student’s material products provides us with insights into her level of genre mastery and the constitution of writing practices (Ormerod & Ivanic 2000). This approach, we argue, enables us to identify specific pedagogic practices and oversights. We show how the conflation of the process and genre approaches in the CAPS document increases the challenge of teaching writing in FAL contexts. The analysis also highlights the complexity of the genre.

Before we present the sample text, we discuss and problematise the approaches to the teaching of writing presented in the CAPS document for English First Additional Language for Grades 10–12. This is followed by a discussion on argumentative writing. The article then outlines the methods used for this research.

Whilst process pedagogy has traditionally been used in home language writing, it is not without its critics, particularly for first and second additional language contexts. One of the primary criticisms is the ‘resolutely asocial’ and ‘wholly individualistic’ nature of the pedagogy (Atkinson 2003:4). Process pedagogy views writing as an ‘abstract internal process’ (Atkinson 2003:5) that underplays the complex social and ideological powers that shape (and are shaped by) dominant writing conventions. Hyland (2003) argues that:

because process approaches have little to say about the ways meanings are socially constructed, they fail to consider the forces outside the individual which help guide purposes, establish relationships, and ultimately shape writing. (p. 18)

It is critical to consider, in a country with an oppressive history such as South Africa, what these forces might entail, and how they play themselves out in the educational arena. We would do well to consider Delpit’s (1988) concern, writing from the context of an African American teacher’s experience:

that adherents to process approaches to writing create situations in which students ultimately find themselves held accountable for knowing a set of rules about which no one has ever directly informed them. Teachers do students no service to suggest, even implicitly, that ‘product’ is not important. Students will be judged on their product regardless of the process they utilised to achieve it. (p. 287)

The inclusion of the notion of ‘text types’ in the CAPS curriculum provides one way to teach ‘the rules’. Learners are required to ‘learn and apply knowledge of the structure and features of different text types’ (DBE 2011:35). Specifying text types with their accompanying linguistic features and structures goes some way to offer explicit instruction of the ways language works in social contexts (Hyland 2003). This is in keeping with genre pedagogy, which argues that teachers should explore ‘ways of scaffolding students’ learning and using knowledge of language to guide them towards a conscious understanding of target genres and the ways language creates meanings in context’ (Hyland 2003:21). In the sample of writing discussed here, it is clear that little explicit knowledge of the argumentative genre or of how grammatical choices shape meaning have been provided to this student writer; thus, the process approach operates at a superficial level.

The genre and process approaches traditionally set specific stages or processes to follow. By conflating the two in the CAPS documents the specificity of each approach has been lost. This conflation requires teachers who can read between the lines by drawing on prior knowledge of both approaches, as well as understand the significance of each (missing) step in both approaches. This is a challenge for any teacher but more so for teachers who have had little access to these approaches in their own education and training.

The steps for Planning/Prewriting are as follows:

1. Analyse the structure, language features and register of the text type that has been selected.

2. Decide on its purpose, audience and context.

3. Brainstorm ideas for the topic, using, for example, mind maps.

4. Discuss the criteria that will be used to evaluate the piece of writing.

5. Research the topic, for example, in a library and select relevant information.

6. Identify main ideas and supporting detail.

The steps for Drafting are as follows:

7. Choose appropriate words, for example, in a narrative use evocative words and phrases to make the writing vivid.

8. Organise ideas in a logical sequence so that the argument flows smoothly in an essay.

9. Organise ideas and/or images so that a story makes sense

10. Establish an individual voice and style.

11. Read drafts critically and get feedback from teacher and classmates (DBE 2011:35).

Step 1, Step 2 and Step 4 for Planning/Prewriting contain a genre element. But the first two steps alone require intensive intervention to make the genre rules explicit. This is crucial in a FAL context. The process of planning is embedded in the steps for brainstorming, research and identification of ideas (where organisation of said ideas is implied). The Drafting steps require a rough draft in which genre and language are foregrounded. Oddly, the step 8 refers specifically to argumentative writing: ‘organise ideas in a logical sequence so that the argument flows smoothly’ despite the fact that not all texts have arguments. This is followed by ‘organise ideas and/or images so that a story makes sense’ (DBE 2011:35). It is not clear why logical organisation is linked to argument and narrative and not to other text types since all texts require logical coherence. Step 7 flags individual voice, which is inappropriate for some text types such as agendas and minutes. The final step, step 11, that requires peer feedback is problematic in a FAL context where social and linguistic capital may not always be in place to offer ‘critical’ (DBE 2011:35) feedback. The final process step, Revision/Editing/Proofreading, gives a list of language and structural features that should be revised but does not mention revising content or ideas¹.

The consequences of conflating each approach can be further illustrated if one examines Flower and Hayes’s (1981) notion of planning. Planning requires goal setting, content generation and the organisation of content in terms of the developing text (McCutchen 2008). When one examines this meaning against the writing steps in the CAPS document, these elements of planning are implicit rather than explicit. This raises a potential problem for teachers. If the purposes of planning (a process step) are not well understood, then discussions about purpose or audience (a genre step) will remain at a superficial or decontextualised level. This impacts on goal setting and content generation. Also of concern is the linear way in which the sub-steps are laid out in CAPS. Planning is a recursive process (Hyland 2003). Merely following the steps in CAPS raises the possibility that what is most important, the process of thinking through ideas and revisiting them, is left out. Whilst it is not the job of a curriculum document to outline the underpinnings of theoretical positions or approaches, the disjuncture between policy and reality, where many teachers lack the pedagogical knowledge to implement these requirements, ultimately sets students up for failure.

In considering academic writing in tertiary institutions Janks (2012) argues that the teaching of writing focuses mostly on left-brain skills but that writing requires both the right-brain operations for generating ideas and left-brain operations for organisation, structure and crafting. Creativity and logic both have a role to play in writing (Janks 2012:8; Mendelowitz in press). Considering the ways in which planning is a creative act for which the brain needs time to think (and rethink) reinforces the importance of thorough planning and brainstorming. It is also the launching pad for crafting the essay where logic and order come to play.

Despite a rich data set, our reason for focusing on one exemplar text is underpinned by Delpit’s (1988) argument that product is important, because it is from this that students are judged. Although cognisant of other writing samples in the data set, choosing a ‘high-performing’ student’s text is useful because such students are most likely to demonstrate a take-up of teachers’ input. In addition, Ormerod & Ivanic’s (2000) work with young children’s literacy, located in a sociocultural paradigm, reveals how literacy practices can be inferred from the materiality of completed texts. Drawing on their work, we analyse how this student’s use of semiotic resources provides insight to the writing practices that (dis)enable the production of this specific text. Critical discourse analysis further enables us to disassemble the ideational, interpersonal and textual (Halliday 2006) meanings of this essay as we work to understand what the text tells us about the construction of argument as a writing practice.

The topic for the argumentative essay is taken directly from a third term unit in the textbook. In line with the CAPS document, the outcome of the unit is not an argumentative essay per se. Rather it provides a range of tasks to scaffold the eventual writing of an argumentative essay in Grade 12. Evidence from all the project schools’ observations suggests that this would have been the first time students engaged with argumentative writing. The textbook tasks include a listening comprehension on ritual slaughter in which students are required to take notes listing the arguments for and against the topic. Students are then required to construct an argument list, taking one side of an argument and listing two points for six different topics (the ritual slaughter essay was one topic). In the following unit students are required to structure a paragraph for a formal debate.

In contrast, it appears the teacher chose the ritual slaughter topic to focus on essay writing. There is no evidence that learners had completed the listening comprehension. Rather than constructing an argument list, as was suggested in the textbook, Busi’s prewriting task is an ‘argumentative’ paragraph:

Ritual slaughter is not a form of animal cruelty

It is not a form of animal cruelty because it is the culture of Black South Africans. The people who say it is animal cruelty, they are simply undermining our culture. For the fact that they say ritual slaughter means that they don’t see anything wrong when animals are slaughtered in a butchery.

This was followed by a mind map the next day (see Figure 1).

On the surface, the pre-writing paragraph appears to meet the requirements of the first step of the process approach, Planning/Pre-writing. Pre-writing tasks for an argumentative text can have several functions. They can unpack the topic, generate ideas, provide evidence for claims, examine both sides of an argument and consider how the argument will be structured. All of these functions are an important part of planning and do not necessarily require separate writing tasks. But this paragraph does none of these things sufficiently. Rather it presents two claims why ritual slaughter is not cruel: it is the ‘culture of Black South Africans’ and ‘they’ (white South Africans perhaps?) undermine black culture because, if ‘they’ were really concerned about animal cruelty, all forms of animal slaughter would be considered cruel. The number of claims is insufficient for an essay. This pre-writing paragraph is a puzzling starting point for helping students to plan an argumentative essay. This possibly indicates the teacher’s limited engagement with the curriculum documents, the textbook, or her understanding of the role of planning for this genre.

A more obvious starting point would be to unpack the topic and the parameters of the debate. An understanding of the terms ‘ritual’ and ‘cruelty’, which is taken for granted in Busi’s paragraph, is essential. For teachers who resist conflict situations in classrooms (Newell et al. 2011), working with semantics may be a way for students to engage critically without heightened emotions. If the genre approach had been used, the class would possibly have begun by first analysing the ‘structure, language features and register’ of arguments (DBE 2011:35). Explicit attention to grammatical structures such as modality and tense can deepen understandings of structure and register.

As puzzling is the brainstorm, completed the next day, that follows the paragraph, which appears to have a similar function of generating ideas (Figure 1). Possibly the bullet point in CAPS, ‘Brainstorm ideas for the topic, for example, mind maps’ (DBE 2011:35) has been reduced to a formula in this classroom: brainstorm = mind map. The visual nature of the mind map, which should be a generative tool, appears to be reduced to a meaningless task as is evident from the generic cloud bubble with four or five words attached. The same reduced visual appears in students’ English books across the project schools. The paragraph and the brainstorm do not show evidence of right-brain thinking. A range of brainstorming techniques could have been utilised when planning for argumentative writing (see Buehl 2009), including the example presented in the Teacher Guide (listing arguments for and against in a table).

FIGURE 1: Busi’s brainstorm for the argumentative essay.

CAPS recommends that ‘research’ is required to ‘select relevant information’ (DBE 2011:35). Positioning ‘research’ in the Planning/Prewriting stage implies that reading and discussion are necessary. Researching topics enables content generation for both sides of an argument and provides evidence for claims. There is no evidence that Busi has had the opportunity to do this – even in a school with a paucity of resources, texts on this topic were provided in the Teacher Guide. Research would enable students to see knowledge gaps, partial understandings or flawed thinking and potentially adjust their own position in the argument. But this is a sophisticated epistemological move where ‘a side-by-side analysis of [students’] prior knowledge’ with new information enables a view of ‘knowledge as mutable, as incomplete, and open to revision’ (Buehl 2011:135). Had Busi had access to other texts, or had the brainstorm been collaborative, she may have had the opportunity to see the limitations of her thinking in the first sentence of her paragraph (‘It is not a form of animal cruelty because it is the culture of Black South Africans’). Discussion questions such as ‘Can cultures be cruel?’, ‘If ritual slaughter is not a cruel practice, what prevents it from being cruel?’, ‘How does one know this?’, ‘What evidence is there to support this claim?’, ‘Where does this evidence come from (experiential or text-based evidence)?’ and ‘How reliable is this evidence?’ would have extended her thinking.

This highlights another shortcoming in the CAPS documents. The generic requirements of an argumentative essay hinge on the validity of claims made; otherwise, the strength of an argument is diminished. The outline of the process approach steps does not provide any space for critical engagement or evaluation of content to consider claims. In addition, the definition of the argumentative genre in CAPS does not specify evidence as a requirement for this essay, only that ‘elaboration’ be present (DBE 2011:37). ‘Elaboration’ is substantively different from ‘evidence’. One may argue that questions of validity and evidence would be covered in a class discussion of assessment criteria, but this presupposes a discussion, a reliable rubric and that teachers understand the rules of evidence themselves (Newell et al. 2011).

Additionally, access to relevant research texts provide not only content knowledge but also register and discourse markers that identify arguments. For FAL speakers, access to texts that model both content and form (e.g. structure, grammar) is essential to their development as writers. When students come from communities that do not necessarily engage in this form of argumentation, and may not have access to people with academic knowledge (Buehl 2011), reading for writing is essential. When one examines Busi’s ideas in the essay below they are not without merit. This is a capable student. However, the ideas require development which reading would provide. We argue that writing pedagogy needs to include careful, sustained planning that integrates reading and discussion in order to develop thinking and reasoning.

Successful planning should provide a smooth transition into the essay. The final Planning step in CAPS asks that main and supporting ideas be identified (DBE 2011). There are no textual traces (e.g. numbering, highlighting, etc.) in Busi’s brainstorm or paragraph to show these connections. Busi adds the words ‘differences’ and ‘respect’ to the brainstorm, which it could be argued develop the idea of culture but the visual does not indicate how. A new idea, ‘the law’ is also inserted. These are not substantively developed in the essay:

Busi’s essay

Ritual slaughter is whereby animals are slaughtered during a ritual.

It is not cruelty to animals because animals are slaughtered in a butary [butchery]. People are fighting towards the fact that people slaughter the animal on their own. This is done because we respect our culture. We slaughter animals because it is our culture and we have to understand that we are not the same and we do different things.

When people report animal slaughter to the law, they are simply undermining our culture. I don’t see anything wrong when people slaughter on their own because it is not cruel because animals are slaughtered in a butery [butchery]. I don’t think that people that report animal slaughter are vegetarians they also eat meat.

Some people may slaughter on their own because they feel the meat from the buttery is not tasty enough. For example chickens are grown from the buttery [butchery] un-naturally, they use machines and clone the chicken. They chlone [clone] it inorder [sic] to make the chicken fat and look tasty when it is not even close to being tasty. So people decide to take the matter to their own hands.

What I’m trying to say is that animals are not only slaughtered for rituals but also for people’s well being, because the chicken from the shop is not healthy enough and many people get heart diseases dew [sic] to the meat from the buttery [butchery]. So it is better to slaughter the animal on your own.

The essay comprises five paragraphs. The first paragraph consists of one sentence: ‘Ritual slaughter is whereby animals are slaughtered during a ritual’. Busi has recognised the genre convention of providing a definition to start the essay. But she has not used the definitions to set up the parameters of the debate. She has only taken one aspect of the topic, ritual slaughter, and not considered it in relation to animal cruelty in the essay. In addition to this, her definition has a circular logic. Had the planning stage entailed a careful unpacking of the topic it would have better supported her needs as a FAL learner. She would have had the time to work out the meanings of the two abstract concepts and their interrelatedness. The advantage of setting out the parameters of the argument at the outset is that it can work as a writing frame. For a novice FAL writer such framing can lessen the cognitive load.

Busi’s planning texts do not develop the question of culture. This alerts us to assumptions that underpin topic choice. From a socio-cultural perspective, the topic of ritual slaughter would seem to be appropriate for Xhosa students. The knowledge of how ritual slaughter is practised would be shared cultural knowledge within the classroom. But shared knowledge can be both an advantage and a disadvantage. We posit that topics that refer to shared practices may be especially challenging because of the implicit (invisible) nature of these practices. It is unlikely that Busi would ever have had to explain and justify the practice of ritual slaughter to people in her predominantly Xhosa community. Additionally, thinking through actual practices and translating them meaningfully in English may have little relevance in her real life.

Ideologically, this topic sets up an African/Western binary with animal slaughter being positioned against a Western belief of animal rights. We are not arguing that topics with connections to students’ lives should be avoided. Rather, teachers may need to help students make the familiar unfamiliar, and be aware of how discourses might interpellate students, thus narrowing the possibilities of what they can write. Pedagogically, addressing the notion of audience draws attention to possible pitfalls in topics. But this means that students have to be clear that the audience is not the teacher (Newell et al. 2011), otherwise there is no need to explain shared or tacit knowledge. We contend that this is what Busi might have done because in a community space there is no need to explain shared knowledge.

On a first read of Busi’s essay, her argument seems illogical, repetitive and one-sided. In order to unpack these entangled ideas we analysed the text using a theme and rheme analysis, which reveals some interesting trends in the way she constructs this essay (see Table 1 and Table 2).

Thematic prominence is achieved with the first position in a clause (Halliday 2002:206); the remaining words constitute the rheme. In an argumentative essay one would expect to find prominence given to the key claims and concepts (animal cruelty, the law, culture) as well as discourse markers of argument (‘but’, ‘on the other hand’, ‘in addition’). What is striking in this theme and rheme analysis is the lack of prominence Busi gives to the key elements of her topic (see Table 1). Rather, she gives prominence to people (8), animals (5) and pronouns (14). When one examines who the pronouns signify (see Table 2), a binary emerges between the people who slaughter the animals versus the people who report the slaughter. Busi’s use of ‘we’ and ‘I’ aligns her to those who slaughter animals and hints at a racialised discourse of ‘us and them’ with Busi being on the ‘side’ of ritual slaughter. Busi uses ‘we’ three times to refer to a more universalised notion of people where she strongly states, ‘we have to understand that we are not the same and that we do different things’. Here she draws on a human rights discourse to argue her point.

The excessive use of pronouns draws attention to the challenges of writing argumentative essays. The written genre requires a level of rationality and objectivity that distances the writer from her subject. Applebee and Langer (2006) argue that students write more narrative than analytic texts. The consequence of this is that students have stronger narrative repertoires. In order to meet the requirements of this essay, we posit that Busi draws on her narrative repertoire that allows for greater pronoun usage in order to make her arguments. Busi is also likely drawing on a social practice of oral arguments where people who strongly identify with a cause use pronouns to reference their group solidarity. Busi’s school knowledge and cultural practices are insufficient to meet the requirements of this genre. Had a genre approach been used, where a nuanced discussion of purpose, audience and register was a core feature of pedagogy, it would have given her footholds into academic writing.

The final process stage in the CAPS documents requires students to do substantive linguistic editing. Busi did not complete this step, so her final product is in fact a rough draft. This stage implies that students are linguistically attuned to the requirements of the genre. It is unlikely, judging from the absence of the genre approach in her other texts, that even if Busi had been given the opportunity to redraft, she had the linguistic resources required for editing. This is evident in her overuse of ‘because’ (5) in theme position. It does indicate her realisation of the need to provide reasons or causes. But, only using ‘because’ does not allow for the incremental development of an argument, nor does it allow for counterarguments. The result is an underdeveloped text which raises questions about how the rules of evidence were explained (Newell et al. 2011).

We would argue though that there is evidence of reasoning in Busi’s essay. She has keyed into a range of current discourses. These include a racialised discourse (‘when people report animal slaughter to the law they are simply undermining our culture’), human rights discourse (‘we slaughter animals because it is our culture and we have to understand we are not the same and we do different things’), dietary and health discourse (‘I don’t think that people that report animal slaughter are vegetarians, they also eat meat’; ‘the chicken from the shop is not healthy enough and many people get heart diseases’) and discourses of science and genetic modification (‘chickens are grown un-naturally, they use machines and clone the chicken … to make the chicken look fat and tasty’). The problem is that she makes implicit claims within an overly ‘pronouned’ text. For example, she conflates butcheries with abattoirs and cloning laboratories. These ideas need to be interrogated and the rules of evidence applied. Access to reading texts would have assisted Busi to refine her thinking, increase her content knowledge, see how other discourse markers can be used and thus extend her language of argument. This again illustrates the importance of a planning process.

In order to analyse these texts, it was also necessary to examine the curriculum requirements. What became evident was the conflation of the process and genre approaches which raises questions about curriculum assumptions of teachers’ knowledge. This points to a need for teachers to be encouraged to engage critically with the curriculum. Given that anecdotal evidence shows that teachers avoid reading documents, the responsibility falls on teacher training institutions and the Department of Basic Education to find alternative ways to get teachers to critically engage with policies and resources.

Often in-service workshops are generic in nature, glossing over specific genre requirements. The data shows areas where Busi needed more support. We contend that this is a result of her teacher not fully understanding the importance of both process and genre steps for argumentative writing. One area is planning and its role in supporting and developing thinking and reasoning. This connection emphasises what a cognitively complex act writing is and the need to provide explicit support at each stage. This makes a case for professional writing workshops that focus on specific genres and, more importantly, develop the ‘pedagogical confidence of the teacher not just in command of linguistic understanding, but also how that linguistic understanding might be applicable … to the development of writing ability’ (Myhill 2005:80).

Teaching complex genres also requires detailed planning of more than two lessons in order to meet the needs of FAL students. This means addressing the perception that teaching writing involves standalone lessons. To meet student needs an integration of all the literacy skills is necessary. Students need time to think, discuss ideas and read. This requires challenging teacher expectations of the forms of capital students have and their ability to think and reason. It also points to the need for teacher trainers to re-examine their assumptions and practices.

Atkinson, D., 2003, ‘Writing in the post-process era: Introduction’, Journal of Second Language Writing 12, 3–15. http://dx.doi.org/10.1016/S1060-3743(02)00123-6

Brandt, D., 2009, ‘Writing over reading: New directions in mass literacy’, in M. Baynam & M. Prinsloo (eds.), The future of literacy studies, pp. 54–74, Palgrave MacMillan, Houndmills.

Buehl, D., 2009, Classroom strategies for interactive learning, 3rd edn., International Reading Association, Newark.

Buehl, D., 2011, Developing readers in the academic disciplines, International Reading Association, Newark.

Cain, S., 2012, ‘The rise of the new groupthink’, The New York Times, viewed 22 June 2013, from http://www.nytimes.com/2012/01/15/opinion/Sunday/theriseofthenewgroupthink.html?pagewanted=all

Delpit, L., 1998, ‘The silenced dialogue: Power and pedagogy in teaching other people’s children’, Harvard Educational Review 58, 280–298.

Department of Basic Education (DBE), 2011, Curriculum and Assessment Policy Statement Further Education and Training Phase Grades 10–12, Pretoria, South Africa.

Dornbrack, J. n.d., ‘Writing pedagogy and CAPS: A South African case study’. (Unpublished article.)

Flower, L. & Hayes, J., 1981 ‘A cognitive process theory of writing’, College Composition and Communication 32(4), 365–387. http://dx.doi.org/10.2307/356600

Gains, P. & Graham, B., 2011, ‘Making space for expressive and creative writing in African primary schools: A two site action research study in Kenya and South Africa’, Reading and Writing 2(1), 77–94.

Gallupe, R., Dennis, A., Cooper, W., Valacich, J., Bastirnutti, L. & Nunamaker, J., 1992, ‘Electronic brainstorming and group size’, Academy of Management Journal 35(2), 350–369. http://dx.doi.org/10.2307/256377

Hadaway, N. & Young, T., 2006, ‘Negotiating meaning through writing’, in T. Young & N. Hadaway (eds.), Supporting the literacy development of English learners, pp. 15–167, International Reading Association, Newark.

Halliday, M., 2002, On grammar, Continuum, London.

Hendricks. M., 2008, ‘Capitalising on the dullness of the data: A linguistic analysis of a Grade 7 learner’s writing’, Southern African Linguistics and Applied Language Studies 26(1), 27–42. http://dx.doi.org/10.2989/SALALS.2008.26.1.3.418

Hyland, K., 2003, ‘Genre-based pedagogies: A social response to process’, Journal of Second Language Writing 12, 17–29. http://dx.doi.org/10.1016/S1060-3743(02)00124-8

Ivanic, R., 2004, ‘Discourses of writing and learning to write’, Language and Education 18(3), 220–245. http://dx.doi.org/10.1080/09500780408666877

Janks, H., 2012, ‘The discipline and craft of academic writing: Building writing capacity in institutions of Higher Education’, Reading & Writing 3(1), Art. #25, 9 pages. http://dx.doi.org/10.4102/rw.v3i1.25

McCutchen, D., 2008, ‘Cognitive factors in the development of children’s writing’, in C. MacArthur, S. Graham & J. Fitzgerald (eds.), Handbook of writing research, pp. 115–130, Guilford Press, New York.

Mendelowitz, B., (in press), ‘Permission to fly: Creating classroom environments of imaginative (im)possibilities’, English in Education.

Myhill, D., 2005. ‘Ways of knowing: Writing with grammar in mind’, English Teaching: Practice and Critique 4(3), 77–96

Newell, G., Beach, R., Smith, J. & Van Der Heide, J., 2011, ‘Teaching and learning argumentative reading writing: A review of research’, Reading Research Quarterly 46(3), 273–304.

Ormerod, F. & Ivanic, R., 2000, ‘Texts in practices: interpreting the physical characteristics of children’s project work’, in D. Barton, M. Hamilton & R. Ivanic (eds.), Situated literacies: Theorising reading and writing in context, pp. 91–107, Routledge, New York.

Rao, Z., 2007, ‘Training in brainstorming and developing writing skills’, ELT Journal 61(2), 100–106. http://dx.doi.org/10.1093/elt/ccm002