Many teachers have been there before.
Years ago, as a newly-graduated primary school teacher, I was teaching a Year 5/6 class, when a student named Sally (pseudonym) handed me a draft report written for our science class. As I read through one of her paragraphs, my internal teacher radar went off. I could sense that this paragraph was poorly written. But when I opened my mouth to offer feedback, no useful guidance came out. I realised at that moment that I lacked the personal knowledge about language (aka the metalinguistic understanding) to know exactly which aspects of language were problematic in the text.
Faced with this problem, I offered her a classic, vague cop-out:
“Sorry, Sally. This paragraph doesn’t read very well. You’ve got time. Can you go back to your seat and have a go at fixing the flow?”
I’ll never forget the expression on Sally’s face as she sat back down. She stared at her page in confusion. And reflecting on this both then and now, she had every right to feel confused. I’d told her that her paragraph was dodgy, but I hadn’t given her the tools of language to understand why this was the case, or the strategies that would help her to fix it.
The interaction with Sally really unsettled me as a new teacher. I imagined how terrible it would be for a similarly positioned new medical doctor to offer a vague response after hearing that their patient was suffering from abdominal pain: “Yeah, your stomach does seem to be in a spot of bother. You’ve got time. Can you go back to your house and rest for a bit until it’s better?” Not good.
Developing my own metalinguistic understanding
As someone who’s always wanted to be a halfway decent teacher of English, this experience fuelled a desire to properly learn my stuff when it came to student writing development. In hindsight, the subsequent years I spent completing a whole PhD about this possibly took my mission a bit far, but I really never wanted to find myself in that situation again.
There are several theories of language that could help bridge my considerable metalinguistic knowledge gap, but being in Australia, it made sense to turn to Michael Halliday’s Systemic functional linguistics (SFL). This theory underpins much of the Australian Curriculum: English (as explained in Exley, 2016), so learning about it helped me understand several previously perplexing content descriptions. While I’m no longer working with Year 5/6 students directly, I now have the responsibility and privilege of training preservice teachers so they (hopefully) never find themselves in the same position with Sallys of their own!
However, providing teachers with a linguistic toolkit like SFL is only half the battle. We also need to understand what happens in a student’s mind as they actually use those tools. To understand the need to develop preservice teachers’ metalinguistic understandings before they finish their degrees, and why writing is so cognitively draining for novices, I’ve found it useful to learn about a framework from cognitive scientist John Vervaeke. Vervaeke is the Director of the Cognitive Science Program at the University of Toronto. With his colleagues, he introduced two helpful concepts for the argument I’m forming here, known as relevance realisation and the salience landscape. These concepts may sound a bit lofty and academic, but they do a great job at explaining the value of explicit instruction and also what went wrong for me as a new teacher.
The cognitive process influencing our writing choices
To make sense of these terms, consider how the human brain processes the world around us. At any given moment, like right now, your brain is flooded with millions of data points: the stray dust particles in the air, the cellular biology of your hand, the history of the room you are sitting in, the pixels on the screen you are reading right now, and many more. In cognitive science, trying to navigate this infinite sea of possibilities is called a combinatorial explosion. If your brain tried to process all of it, the system would crash.
So, why don’t our minds constantly crash?
According to Vervaeke and colleagues (2012), our minds survive the information onslaught by running a background filter called relevance realisation. This is the unconscious, dynamic process our brains use to sift through the surrounding chaos and shrink reality down into a manageable size, selecting only what matters to meeting our immediate goals. We unconsciously choose to notice what’s relevant to our goals, while filtering out the rest as irrelevant noise.
The immediate, conscious result of this filtering is your salience landscape (i.e., your actual, lived experience of reality). When your background filter works properly, the things relevant to your goal visually and mentally glow with importance (they pop out at you), while the background noise fades away. In short: relevance realisation is the unconscious information filter; the salience landscape is the more manageable picture you’re left with.
Importantly, relevance realisation is a blank process until it is calibrated by constraints, tools, and background knowledge. Without explicit knowledge, a person’s filter has no criteria to judge what is important. This is where metalinguistic understanding comes in!
Relevance realisation and school writing
Still with me? Great. Let’s relate this to writing in a school context.
When a student sits down in front of a blank page, they experience their own combinatorial explosion of infinite possible choices. Inside their minds they experience a chaotic flood of thoughts, episodic memories, background knowledge, and vocabulary choices. To make language choices that express meaning clearly, their internal filter has to tune out the noise so that the exact words they need glow in their salience landscape, ready to be written down.
The issue for novice writers is that their internal filters haven’t been trained yet. Because they lack an explicit, metalinguistic understanding of how language works, their brains can’t determine which structural or grammatical tools are actually relevant for meeting their immediate writing goal.
As a result, nothing glows as a suitable choice in their minds (let alone on the page!). The vast choices they could make in verb groups, noun groups, adverbials, conjunctions, text stages, and so on all blend together as undifferentiated options for the novice writer. Their salience landscape is like a featureless map where no single path stands out, leaving them unsure what to write.
The importance of building teachers’ metalinguistic understanding
This reality facing every novice writer is why teachers’ need to have their own developed metalinguistic understanding. When teachers are equipped with a robust linguistic framework from the word level to the whole text level, they can explicitly show students how to make choices that suit various writing purposes.
For example, if a student constructs weak noun groups in a science report (like Sally had), the teacher can draw on their metalinguistic understanding (which has calibrated their cognitive filter) to instantly diagnose the issue:
“Ah, I see that Sally has built noun groups using everyday, conversational language here, like the stuff. For this science report, her writing will be more effective if she uses precise, technical noun groups that pack relevant information into the text, like the hot chemical solution.”
But having the metalinguistic knowledge to identify a writing problem like this is only one part of the solution. Armed with this understanding, teachers must then use explicit instruction to build that same metalinguistic knowledge in their students, effectively training their internal relevance realisation filters so they know exactly which language choices matter in this context.
Building students’ metalinguistic understanding with explicit teaching
It is unreasonable to expect primary school students to intuitively discover how different written genres work on their own. Because their internal cognitive filters haven’t been trained to sift through infinite linguistic choices that might be relevant in the writing context, leaving them to experiment or fix the flow inevitably results in frustration for them and you (though it might admittedly drive you to complete a PhD 😅).
Navigating the specialised written registers of schooling requires calibrated relevance realisation. Without explicit knowledge of the language features that matter in a specific genre, students will struggle to see the path to successful writing.
A better approach is for teachers to deliver targeted, explicit instruction that highlights how texts are built at the word, sentence, paragraph, and whole text levels in the context of a specific writing task. Following the well-established Teaching learning cycle (Derewianka & Jones, 2022; Humphrey & Feez, 2016; Rothery, 1996), here is one example of how a teacher can gradually release responsibility for making these writing choices to students:
The Teaching learning cycle in action
Step 1: Deconstruction and explicit modelling:
The teacher works with the students to explicitly name and isolate a language feature based on their immediate needs. For example, using an effective mentor paragraph from a science report on the board, the teacher demonstrates how a scientist constructs noun groups.
They isolate the head noun (the technical scientific object), the pre-modifiers that come before the head noun in the noun group (such as articles, factual describers, and classifiers defining the object), and the post-modifiers that come after the head noun (such as prepositional phrases or embedded clauses providing extra detail). The deconstruction helps the students to visualise what successful writing looks like in this specific genre.
Step 2: Joint construction of a new text:
Next, the teacher displays a partially written science report paragraph featuring intentional gaps where the noun groups should be. This is actually one of several strategies from the work of Debra Myhill that I outlined in a previous post. Together, the class practices building noun groups that match the features highlighted in Step 1.
As the teacher scribes their suggestions on the board, they provide real-time feedback. Instead of a vague “let’s fix the flow,” the teacher can ask:
“Look at this noun group we’ve started building. It lacks a pre-modifier that would tell us what the liquid is like or classify what type of liquid it is. What descriptive words could be add here to pack more information into the sentence?”
The joint construction lets the students practise making the valued language choices with their peers and the support of their teacher. Importantly, this happens before they are asked to do so independently.
Step 3: Independent (or guided) writing:
The students are now ready to build their own noun groups. Those who are able can complete a matching independent gap-filling task, while students needing additional support engage in the same task as part of a guided writing session with the teacher or an aide.
This isn’t about rigid, formulaic writing where every child copies the same correct response. Because language offers infinite choices, explicitly teaching the linguistic patterns that characterise different genres is what actually equips students to run their own relevance realisation filters. Once they understand the distinct expectations of a specific genre, their internal filter knows what to look for, allowing the right choices to glow on their salience landscape.
When students learn how specific language choices create distinct effects on the reader, they can choose to follow the expected conventions or deliberately subvert reader expectations when this makes sense. Such decision making is a requirement of the Australian Curriculum: English in late primary school, preparing students for the increased writing demands of secondary school (for example, see AC9E6LE03, AC9E9LA03, AC9E9LE06).
Following this, students share and justify their unique noun group choices. Peers and the teacher can then offer targeted, specific feedback, relevant to the new learning. For a student like Sally, such feedback would offer much more useful guidance than a vague suggestion like fix the flow.
Writing education in schools today
Zooming out to look at writing instruction across Australia today, I can see many teachers following one of two opposing approaches. Unfortunately, neither approach is likely to set students up to handle the writing demands of secondary school. This is especially problematic for most boys, who we know struggle far more than girls in the transition between primary and secondary school.
Pitfall 1: Unstructured, choose-your-own-adventure style writing instruction
On one side, we still find unstructured writing experiences in many classrooms where students are simply asked to write reflectively or creatively with very little guidance at all. This approach, which was widespread in the days of whole language teaching, might be slightly more defensible when students come from highly literate backgrounds (i.e., those who already know what sounds right when writing because they have been saturated in books and rich conversations at home). Their relevance realisation filters have already been (informally) trained by their environment. But this unstructured approach is problematic for equity. It does nothing for students who haven’t grown up with the same cultural advantages and need explicit guidance to write in contextually appropriate ways.
Pitfall 2: Rigid, nuts-and-bolts style grammar instruction
On the other side, we can see the increasing uptake of bottom-up, nuts and bolts style writing programs. These approaches are far more structured, but they tend to heavily emphasise decontextualised grammar/syntax tasks, isolated sentence-combining exercises, and rigid paragraph formulas. We build up to the paragraph level, but don’t really get beyond this at all or until secondary school when it’s too late.
While it’s obviously important that students can construct coherent sentences and paragraphs, the problem with this kind of approach is the failure to recognise that sentences and paragraphs are never neutral; they are always deeply influenced by the writing purpose and the genre. A grammatically correct sentence in a science report can be entirely wrong for a critical response or a poem.
When we separate language choices from rhetorical purpose (even when we do so in highly structured ways), students don’t develop their capacity for relevance realisation. By learning to build small parts of texts in a vacuum, they ultimately struggle to transfer these isolated skills when asked to dynamically choose language features that suit different genres and audiences.
To be truly effective, writing instruction needs to explicitly teach students how grammatical and structural choices shift depending on the context. There needs to be a clear link between language choices and writing purpose. This is what will develop in them a profound tool for learning in every discipline and communicating what they’ve learnt to others.
How to really fix the flow
Reflecting back on that day in my Year 5/6 classroom, I now understand that Sally was simply a novice writer caught in a combinatorial explosion of infinite language choices. She didn’t need to go back to her seat and guess at how to make her text flow better. What she really needed was explicit guidance about how a science report is written with particular language choices. This teaching would allow appropriate choices to glow on her salience landscape, resulting in less confusion and a more successful text.
As educators, we owe it to the thousands of Sallys sitting in classrooms across the country to build our own metalinguistic toolkits. There are many ways to do this – one simple option is to browse through the online short courses for writing instruction on the PETAA website. Only when we master these tools can we confidently teach the important genres of schooling and diagnose precisely how students can improve their writing. At that point, we can stop asking them to create their own solution for fixing the flow, and instead give them the explicit tools to command written language for their own purposes.
References
Derewianka, B., & Jones, P. (2022). Teaching language in context (3rd ed.). Oxford University Press.
Exley, B. (2016). Secret squirrel stuff in the Australian Curriculum English: The genesis of the ‘new’ grammar. Australian Journal of Language and Literacy 39, 74–85. https://doi.org/10.1007/BF03651908
Humphrey, S., & Feez, S. (2016). Direct instruction fit for purpose: Applying a metalinguistic toolkit to enhance creative writing in the early secondary years. Australian Journal of Language and Literacy, 39(3), 207-219.
Myhill, D., Watson, A., & Newman, R. (2020). Thinking differently about grammar and metalinguistic understanding in writing. Bellaterra Journal of Teaching & Learning Language & Literature, 13(2). https://doi.org/10.5565/rev/jtl3.870
Rothery, J. (1996). Making changes: Developing an educational linguistics. In R. Hasan & G. Williams (Eds.), Literacy in Society (pp. 86–123). Longman.