How do we know what a text really means?

In my research, I am interested in how teachers construct particular stories, or meanings, about digital technologies, the relationships between these stories and other stories of education, and the relationships between these stories and teachers' technological practices in the classroom. A core question for me will be how to understand, interpret, analyse and make meaning from the texts that are produced as part of this research. This seemingly basic question - how do we understand what a text means - is one which is not so easily answered. It is a methodological question: How do I, as a researcher, make claims that my analysis of a text's meaning is a valid one? It also gives rise to many conceptual questions: Where does the meaning of a text lie anyway - is it located in the text itself, in the author's intentions for the text, or in the reader's interpretation? When people disagree about what a text means, how do we decide who is right? What are the processes by which we derive meaning from a text - are some more legitimate than others and if so, who decides? I'm certainly not about to come up with final answers to all these questions here, but I am starting to think about some of the ways that have been taken to some of these questions. (And by 'text', I am not only referring to artefacts created using written language, but a set of meaningful signs that hang together to create a relatively coherent whole - a conversation, a film, a dance, etc.)


Critical discourse analysis seems to promise a theoretical and methodological approach that begins to answer these questions and provide a way for researchers to go beyond a naive and inevitably partial 'face reading' of a text. But reading some examples of CDA, I've been very uncomfortable about what some of them seem to be claiming. Often these approaches claim to deconstruct a text to reveal the hidden unpalatable political ideologies and power relationships upon which they are based.

Perhaps its my background in literary theory but analyses that claim to be able to reveal the 'hidden' meaning of a text set off great big alarm bells for me.

If such a technical and theoretical analysis is required to reveal the 'true' or 'hidden' political ideologies of the text, then how does the text have any effect on its reader? If it is only through a process of rigorous CDA that this underlying meaning can be grasped, then the readers or audience of the text are surely safe from any dangerous political connotations. What this seems to tell us is that what a text means is not just a theoretical question, but a practical, empirical question - we need to ask what a text actually means to particular people in particular situations. And it may not mean the same thing for all of them, because context is essential to the meaning of a text, and everyone brings different experiences to contextualise their understandings.

The 'true' meaning of a text as revealed through some CDA approaches is often taken to reveal what the author really meant, or thought. Authors (speakers, designers) could always make different choices about the words, images and other signs they use, but it does not necessarily follow that they have complete freedom and agency to choose any combination of words at all. There are social, cultural and organisational limits on what someone feels able to say and how they feel able to say it in any particular situation. We are not social 'dupes', mindlessly repeating whatever society tells us to, but neither do we always have the freedom, skills, time or other resources to say exactly what we want to. And authors often approach these compromises very explicitly and consciously in a sometimes complex process of negotiation that seems to be overlooked by an approach that equates the results of CDA as revealing the author's 'hidden' meaning.

As an example, I once sat in a seminar where researchers described how they had analysed the number of words, pictures and pages in books from a schoolchildren’s graded reading scheme and discovered that there was a consistent word-picture-page ratio, and then analysed this to reveal how authors’ conceptualised children’s reading development. At the time I was a publisher of children's literacy resources including reading schemes and the word-picture-page ratio was the result of a process of explicit decision-making and negotiation that resulted from a process of conducting market research with teachers, consulting with academic reading experts, taking into account organisational requirements to make a certain percentage profit and the costs of producing the books and making marketing decisions about the price we should sell the books for and the number of books we could sell. There were tensions in balancing all these considerations and compromises were made. The researchers' 'textual' analysis approach that placed all analysis on the text itself missed these conditions of production, and so simply equated the way that children's reading development was constructed in the text with authorial intention. This is all to say that the creators of a text may consciously be aware that their 'text' does not perfectly represent their own views, but is what they are able to do given the available circumstances and resources.

These problems are caused by separating the 'text' as an artefact from the context in which it was created, and the context in which it is interpreted by an audience. In doing so, it fails to distinguish between what texts tell us about its authors' meanings, the effects of texts on readers, and the meaning of a text 'in itself' (a question I'll leave for another day). Critical discourse analysis is usually described as a process of going back and fore between the text and its context, and when this is done well, I think these problems could be avoided.