When I recently asked for ideas for future posts, many people suggested one on metadiscourse. This topic certainly needs addressing, but I have been dragging my feet about doing so. And the reason for this reluctance is simple: I find metadiscourse difficult to teach. The great thing about writing this blog is that I have now had to think about why I find it a hard topic to teach. What I have realized is that I struggle because I am using both a simple and a more complex understanding of metadiscourse. Let me begin by explaining these two ways of looking at metadiscourse.
First, the simple version: metadiscourse is often presented as the writing that we do about our writing, rather than about our topic. I find this definition to be simultaneously useful and limited. It is useful because it makes intuitive sense and because it highlights the crucial—and often neglected—task of explicitly guiding the reader through a piece of writing. It is limited because it overlooks all the other things that we do as writers to help our readers understand and accept our ideas. This brings us to the more complex understanding of metadiscourse: the linguistic strategies that we use to manage the evolving relationship between writer, reader, and text. In the words of Hyland and Tse (2004), metadiscourse is “the range of devices writers use to explicitly organize their texts, engage readers, and signal their attitudes to both their material and their audience”. I love this definition because it offers a valuable description of what we can accomplish through our writing choices; I think all writers benefit from thinking of writing in this multifaceted way. But as a teacher, I also value the conceptual clarity of dividing writing about writing, on the one hand, from writing about the topic, on the other. Essentially, an unsophisticated understanding of metadiscourse ends up being, for me, a valuable teaching strategy.
For the purposes of this post, I am going to follow my classroom practice and call this writing about writing, ‘signposting’, and I am going to call the more subtle devices that we use to structure the three-way relationship between text, reader, and writing, ‘metadiscourse’. According to this breakdown, signposting is, of course, a form of metadiscourse. (Other key forms of metadiscourse: making transitions between ideas; providing evidence for claims; offering elaborations of key points; managing different degrees of certainty; signalling authorial attitudes; seeking to engage the audience; and introducing an explicit authorial presence. These forms, taken from Hyland and Tse, overlap with one another and with what I am calling signposting. I will devote a later post to discussing all of this in more detail.) The reason I like to pull signposting out and treat it separately is the tendency of novice academic writers to neglect it. In the student writing I encounter, metadiscourse is generally already there—and just needs to be better understood so as to be used more effectively—while signposting generally needs to be increased.
It is very common for me to say something like this to a student in a one-on-one session: ‘You have introduced your topic, but you have not introduced your paper.’ In other words, the reader knows what the paper is about, but not how the paper itself will proceed. The interesting thing for me is that this is an area in which I find some students are actually quite reluctant to follow my advice, either out of their own inclination or because they believe that their advisor is against that sort of writing. My approach in these cases is to acknowledge the way that signposting can make writing appear laboured and then suggest that they try it, at least provisionally. If they really hate it, they can always take it out later. There is a real benefit to using signposts, even as a kind of scaffold: the very act of writing such phrases—for example, ‘In this paper, I will’ and ‘After a discussion of x, I will turn to y’ and ‘This paper will be divided into four sections’—gives clarity to the writer as well as to the reader. If you are one of those people who really dislikes the way that sort of writing sounds—I confess that I don’t see the problem, but my own writing tends to be somewhat pedestrian—you can remove it later or, even better, turn it into something that sounds more sophisticated. As your understanding of your own writing deepens, it is often quite easy to move from ‘In the first section, I will discuss X’ to ‘Given the centrality of X to any treatment of Y, this paper will begin by demonstrating the internal complexity of X within the context of Z’. Even this highly generic example suggests how we can use signposting (‘this paper will begin’) within sentences that also serve to deepen the reader’s understanding of the topic.
One final point for today: the central idea behind signposting is that the reader needs to know how your text will be structured. In those cases in which the reader doesn’t need to know, signposting will obviously be unnecessary. I can think of two such cases: one, a really short text in which the reader will know what’s what soon enough; and, two, a text with a completely predictable structure. Sometimes a student, in a laudable attempt at clarity, will end the introduction of an IMRAD-style paper with something like this: ‘The method section will discuss the method. Then the results will be discussed. Finally, concluding remarks will be given in the discussion section.’ Since the reader already understands the form that the author is using, they don’t need that information. You can usually tell that you don’t need to be providing structural hints of this sort when you find that you are writing redundant-sounding sentences like these.
This post has just been a general introduction to the idea of signposting and metadiscourse. In a future post, I will look in detail at the various forms of metadiscourse. I also hope to write a separate post on the related question of using the first person in academic writing. In the meantime, please raise any questions or confusions below.
Source: Hyland, K., & Tse, P. (2004). Metadiscourse in academic writing: A reappraisal. Applied Linguistics, 25 (2), 156–177.
fantastic post. and super helpful.
–yes, the idea of foreshadowing how your paper will proceed is something students need to work on–at least my students do (i teach comp 101 classes). i often (VERY often) encounter students who flit from idea to idea to idea (and there is a certain kind of logic at work–“Reality TV is superficial. Americans just want to gossip. We shouldn’t be so nosy. We just want to watch famous people because they are famous.”–but there isn’t the kind of development of a central message which i want to see in an essay…the kind of unfurling of an idea with specific premise points, etc.
–perhaps if i really demand the students include a bit how-i-will-proceed signposting, they will grasp how to treat their thesis in a more analytical way. for me, perhaps signposting can help the students bring more critical-thinking skills to their work…and less “info-dumping” as they search for ways to fufill the assigned word count.
again, thanks for the post.
stray question: why do we teach the definition essay? why is important students learn how to produce one?
Thanks for the post. It’s both informative on a theoretical level, and practical for someone teaching writing (like me). Thanks!