The focus on critical access by the Writing Centre is especially important in the light of ongoing student protest action across Higher Education in South Africa over the last three years which began with the Rhodes must fall campaign, and the ongoing calls for decolonization of the curriculum. Ngugi wa Thiong’o describes how colonialism on the African continent not only included the taking of land and resources, but importantly included the «mental universe of the colonized, the control, through culture, of how people perceived themselves and their relationship to the world» (1986, 16).
Voice is a key concept in Higher Education which can enable student awareness of their own agency within the constraints of academic practices. It is not a new concept, but the investigation here is new in that it focuses on voice across modes, media and genres. Voice is an elusive concept with multiple definitions. It can be used as a proxy for an essentializing humanist view, as in finding your own voice (Elbow 2007). Or, in progressivist pedagogy, voice can be seen as taking a certain standpoint and also as a critical term for formulating an alternative pedagogy (Cope & Kalantzis 1993). Usefully, Clark and Ivanič (1997) speak about writers creating a discoursal self through the act of writing and the discourses they enter into as they write.
Why then use a mode-specific term like voice which is so closely linked to the mode of speech (and writing)? The concept of voice comes with a particular history, and with particular theoretical resonances which allow us to look at text in a way that draws on theories of writing and an academic literacies perspective. An academic literacies approach to writing takes into account issues of identity, institutional relationships of discourse and power, and the contested nature of writing practices (Lea & Street 1998, 159). The field of academic literacies has a long history of theorizing agency that a term like stance or style may not have. Stance refers to authorial point of view (Hyland 1999) whereas voice is more concerned with the relation between the writer and reader. I use the term voice to refer to the way a sign-maker establishes presence in a text through the choice and use of semiotic resources (signifiers of authorial engagement), as well as positioning in relation to other sources, and positioning in relation to the audience and site of display.
Instead of thinking about voice as aligned to coherence, which is perhaps a more liberal humanist notion of voice, I think about voice as negotiating and containing contesting voices. This is Bakhtin’s (1981) notion of dialogism which emphasizes voice as a site of struggle. However, these choices are not infinite, and voice is subject to contextual conditions which are located within larger patterns of inequality and power relations (Blommaert 2005). This notion of voice operating within constraints provides a useful way of thinking about voice and agency. There are contextual constraints which include institutional culture, as well as discursive practices, such as disciplinary genres. Besides semiotic alternatives for meaning-making, there are also semiotic constraints within different modes and media, as we saw with the manga comic on screen versus on paper.
To stay with this idea of constraints for a while, Simon Bell (2016) reports on student writing projects in the context of the Higher Education subject of design. The projects were characterized by artificially imposed constraints such as: writing within image and shape restrictions; writing within font restrictions; and writing within word restrictions. Bell argues that paradoxically more weight is given to each word by using so few. This sharpens the focus on the words, «making them work harder to produce meaning» (Bell 2016, 141). This is a lot like a PechaKucha style presentation. In this type of presentation, there are only 20 slides in the powerpoint. Each slide is to last 20 seconds and they are to be comprised of images and no writing. Here the constrictions of form can also enable a new kind of thinking and a tightening of argument, for example, finding an image to illustrate a concept could tighten or broaden your notion of that concept. Constricting time can also provide a certain energy to a presentation in the way that slides containing a list of bullet points cannot. However, as with any form, there are certain gains and losses. Definitions of concepts are difficult to do in images and are often less precise than when realized through writing. For instance, I could try to capture my definition of voice in four images – as being about presence in text, choice of semiotic resources, positioning in relation to other sources (citation), and in relation to audience (see Abbildung 1).
Abbildung 1: Definition of voice portrayed visually
Choice is represented by a fork in the road, citation as a head amongst smaller heads located in a speech bubble, and audience is held in the palm of your hand. It is clear that the images function more metaphorically here and less clearly and explicitly than a written definition would. How well this definition works is thus debatable. In addition, these images are all taken from open access websites. These kinds of generic images are problematic in some ways, as the sign-maker has to force a connection between the image and the concept. My point here is that constraints can enable creativity and different ways of thinking, but also that certain modes and genres are better suited for different functions (and definitions tend to work better in language than in image).
A framework for academic voice
In order to engage with the concept of voice, we need appropriate semiotic categories for analysis that work across modes. I would like to propose two categories for looking at academic voice. The first is authorial engagement which concerns the extent to which authors choose to engage with their audiences and subject matter (Hyland 1999) and is realized through different mode-specific conventions. The second category is citation. Citation is central to the construction of academic voice and is marked by positioning in relation to a particular discourse community. I will begin by focusing on authorial engagement.
Authorial engagement
Authorial engagement is expressed through choice, namely the type and the composition of the representation. These representational choices depend on context – aptness for purpose and audience. In writing, authorial engagement can be revealed through attitude markers, the use of pronouns, active and passive voice, and rhetorical questions. Authorial engagement can even be established through the visual design of the writing such as typographical choices, the use of punctuation (for instance, the use of commas to pause and breathe), layout, the use of white space, spelling (is it grey or is it gray?) and ways of asserting identity through cool visual lexicons (such as homophones is short text messages).
In images, authorial engagement can be demonstrated through aspects such as the choice or type of image, how the image is sourced, the use of colour and even colour saturation, composition (what is foregrounded and backgrounded), and social relations between represented participants in an image (see Archer 2013). In terms of composition, Kress and Van Leeuwen talked about the concepts of given and new in their 2006 book, Reading Images. A grammar of visual design. Perhaps they have moved away from these more prescriptive notions now, but the fact remains that in composition, position matters. For example, in the past, the figure of Johnny Walker on the whiskey bottle was represented as walking from right to left. This directionality was seen as signifying movement into the past and it was then changed to the current logo where he walks from left to right, signifying walking into the future. Positionality and directionality as such are thus important semiotic resources for meaning-making.
Now, I will look at a how authorial engagement is realized in infographics and data visualizations. It is important to be aware of the construction of these texts as the assumptions underlying the numbers are generally hidden and numerical representations are often regarded as more factual and objective than other kinds of evidence. Here we need to look at: what is measured and summarized; the type and choice of information graphic; the composition of the graphic (Prince & Archer 2014). There are design choices to be made here in terms of size, shape, colour and composition in order to represent an argument to a particular audience in the most apt way. One could, for example, choose the default design option when creating a bar chart to represent data. This would comprise blue bars on a grey background. There are a host of other design choices that could be made, however, such as using