Reading and writing non-fiction - it's complex and not boring!

I have always had a funny relationship with how we deal with non-fiction texts in schools. For years, I have felt that we just haven’t got it right. We oversimplified the teaching of non-fiction in terms of triplets (writing to advise, inform, instruct) and reduced it down, largely, to simple techniques. If you are writing a blog, you must have X, Y and Z. We are spending a lesson looking at writing to advise today, people. No matter what level or approach of teaching non-fiction, it was largely reductive and simplistic. For this reason alone, teachers would rather teach stories rather than non-fiction. You can see the collective yawn when it comes to Paper 2 on the GCSE exam paper. Even the teachers go: do I have to? It is ‘so’ boring!

Why do we have this issue with non-fiction? Why is it always seen as the poor alternative to fiction? Narrative is the key thing. We love a narrative. We love structure, order and patterns. When we read a story, we enjoy the progression of one narrative and see how the writer structures that narrative and plays around with the order and patterns of things. For the most part, we are reading about one person and their relationship with others.

I’d be bold to say that, as humans, we are wired for story. We read, speak and write stories on a regular basis and it is the thing that both entertains and informs our lives. Through stories we understand the world. Non-fiction does have elements of story, but doesn’t have emotional connection or structures and patterns. If I am reading a novel, I know that Tom, the main character, is my compass for the story. Read a piece of non-fiction and the writing moves from Tom, Betty and rice  and then spins to Victorian greed. The subject and focus moves. Constantly. I am not following one person’s story, but hundreds of micro narratives that logically or thematically link but don’t have a clear cohesion with a clear personal narrative.  

That’s just one issue with non-fiction texts, so what else is there? 

# Denseness of ideas. One text can cover so many different ideas encapsulated in four paragraphs. It can cover religion, ethics, spiritualism, class issues and geographical poverty in one small text. Rather than tell a narrative of one of these ideas, instead the writer has threaded a narrative around these elements and there will not be a Tom to unite these things together. 

#Mind model and picturing. As we read a story, well most of us, we construct a model or image in our brain. We visualise the story as it is going on. Tom is sad. I picture in my head Tom crying. Non-fiction doesn’t easily allow for this mind modelling. For example: The French Revolution was an economic problem.  

# Reading speed. Students read and approach non-fiction like it is fiction. We read fiction quickly and at a pace so we can get to the next bit. Completion is key. Non-fiction cannot be read quickly. You can, but largely it is a slower paced read as we digest an idea or think about a point mentioned. 

# Inferences related to different knowledge domains. I like Charlie Brooker as a writer and I have used a number of pieces of his work in lessons, but the amount of knowledge students need to understand a paragraph. He will make inferences that relate to politics, TV shows from the 80s, sex and many other things. Often, I find I have to spend my time teaching the prior knowledge than actually read the text.  

#Non-fiction is largely written with a literary quality. Like narrative, writers cycle through nouns when describing one aspect. That tends to be fine when talking about characters in a car. When making inferences, you can make some easy assumptions. There’s only three people in the car it could possibly refer to. Imagine how hard it will be to make an inference if you are talking about a global war with hundreds of key players. 

For a lot of the reasons above, I can see that examiners tend to take non-fiction texts that lean toward fiction in their approach. This you can certainly see in the GCSE English Language exams. They tend to have two or three people and focus on one moment or event. So, how do we help our students get better at reading non-fiction? 

[1] Talk about synonyms and how writers cycle through nouns. 

Look at how Queen Elizabeth I could be referred to in a text. If a student doesn’t know that these all refer to the same person, then their understanding will lack precision. 

Queen Elizabeth

Elizabeth I 

Queen

Queen of England Queen of Ireland

monarch

head of state  

ruler

sister

daughter

Henry VIII’s daughter Mary’s sister

Anne Boleyn’s daughter

Virgin Queen

Protestant  

 

It isn’t just people that this relates to. Look at how the government can be ‘named’ in a text. 

Assembly 

Parliament 

Regime 

Powers that be

Governing body 

Powers 

Senate 

Congress  

State 

Authority 

Ministry 

Rule 

Council 

Cabinet 

Establishment 

How we do this depends on the text. There might be space in the curriculum to do this before reading the text, but you can do this when reading and make it part of the online inference process. The first sentence refers to the government. What word in the second sentence also means government? 

[2] This / It 

Students need support with ‘it’ and ‘this’ because they rely on students making cohesive inferences. Lots of students are experts on using this and it at the start of a sentence. Yet they aren’t experts when reading a text. That’s because the cohesion between ‘this/it’ and the subject it refers to isn’t so clear. There may be sentences between them or the subject isn’t so clear. 

When it uses the word ‘it’ was it referring to? 

[3] Pronouns 

This is slightly linked to the previous one, but I felt it needed addressing in its own way. In the same way writer’s don’t want to mention the name Queen Elizabeth, they will rely on using pronouns in addition to synonyms. We need to remind students of this when reading. Who does the ‘her’ refer to? This again is one of those things that students need to understand the use of pronouns and the connection between that pronoun in a sentence and the reference to the noun in the previous sentence.  

[4] Topic sentences are key for inferences later in the paragraph. 

Like sentences, the rest of the sentence depends on you recalling the subject of the sentence. Inferences later in the text are reliant on a reader recalling the first thing described in the topic sentence. That’s why it is important for students to read, know and commit to memory the topic sentence. Plus, it pays for us to keep going back to the topic sentence as we read. All those pronouns and synonyms probably relate to the initial subject mentioned in that first paragraph.  

[5] Reading speed is the key. 

We need to model the slowness of reading non-fiction. The stop and start nature of it. The reading purpose should be clear. It isn’t to get to the end without dropping off to sleep. It is to get a detailed understanding of an idea or topic. That involves lots of checking and rechecking what has been read.

[6] Questioning. 

Students view non-fiction in simplistic terms and so their default reading relies on skimming and scanning. Therefore, questions asked by teachers need to work against skimming and scanning. Questions should be testing understanding and not rely on the search of knowledge. How did Queen Elizabeth feel about the rebellion? 

Let’s stop treating the reading of non-fiction in fiction reading terms. We do a disservice if we do this. My reading of non-fiction is built from decades of reading. I love reading non-fiction texts, but we need to stop the dangerous oversimplification at all levels. Lean in to the complexity of non-fiction and teach real texts with the understanding of its complexity. Reading non-fiction needs changing. 

 

Thanks for reading, 

Xris 

Sources: 

Effective Teaching of Inference Skills for reading -  Anne Kispal (2008)

https://www.nfer.ac.uk/publications/EDR01/EDR01.pdf

Children’s inference generation: The role of vocabulary and working memory - Nicola Kate Currie, Kate Cain (2015)  

(PDF) Children’s inference generation: The role of vocabulary and working memory (researchgate.net)

Online inference making and comprehension monitoring in children during 

reading: Evidence from eye movements – Holly Joseph , Elizabeth Wonnacott and Kate

Nation (2021) 

https://journals.sagepub.com/doi/full/10.1177/1747021821999007

Infer, infer, they’ve all got to infer for me

 'The ability to make inferences is, in simple terms, the ability to use two or more pieces of information from a text in order to arrive at a third piece of information that is implicit.'

'one of the underpinning bases of inference is vocabulary’

Anne Kispal 2008 

During the Birmingham ResearchEd talk I gave on reading, there was a small interruption. One of the stewards arrived to find out if there was an owner of a Jag in the room. Nobody responded and casually made a joke about what we could infer from that piece of information. 

• The person is blocking another person in the car park. 

• The person has left their lights on. 

• The person has quite a bit of money. 

• The person likes status. 

In fact, there are quite a number of inferences I can make and some might be bordering on supposition and not really grounded inferences. What that one point highlighted was the amount of previous knowledge needed to build that one little inference. You need to know that a Jag meant a Jaguar car. You need to know that if people ask about a car then there’s usually several reasons why. You need to know that Jaguars are quite a flashy and expensive car. There’s a lot of information needed to build up those inferences. That’s why I have built cars and car manufacturing into a unit of work. I am joking.

I have taught for several years now and at the heart of reading and English teaching is inferences. I can teach students everything under the sun, yet when students are on their own, they need to be able to infer something from the text, independently. On their own. Unstructured. Unaided. In an exam. Or, in real life. When we read a poem, we make inferences. When we read a novel, we make inferences. When we read non-fiction, we make inferences. We are constantly making inferences about elements in English. 

Let’s boil the types of inference we make in English to a more precise level. 

[1] We make inferences about a character / setting or thing in a story. 

What do they think, feel or really mean?  This is usually based on tone, language, body language or actions. 

[2] We make inferences about a reader’s thoughts or feelings towards a text. 

What is the reader thinking at this point?

[3]  We make inferences about the writer’s intent. 

What do you think the writing is trying to say about X here? 

Those three pillars of inference are what we focus on a lot in English. We could add more, but, for the time being, three is enough for now.  Across our Trust, we are tracking how these elements are taught over KS1 to KS3. We see that inference and the ability to make inferences is a huge milestone in education. We are working on the decoding element and reading fluency, but inference, for us, is a massive leveller for students. It is what stops students from making big leaps in their thinking. If a student struggles to, independently, make inferences from a text, then they often can’t get to the next rung on the ladder. Skimming and scanning does not get you far. There needs to be something more. 

Look at the GCSE Paper 2. It is full of students making inferences. Paper 1 adds in character inferences and inferences about the reader. 

Question 1	Question 2	Question 3	Question 4
Use inferences when deciding if something is true or false	Explain two inferences about a given topic	Make an inference about the writer’s thoughts and feelings Make an inference about the writer’s reasons for the choice.	Make an inference about two writers’s thoughts and feelings Make an inference about each writer’s reasons for the choice.

If I am honest, there has never been any explicit CPD about inferences in English. There’s been some programmes on inference training yet during all my time in teaching there’s been very little talk about it. Infer from that what you wish. I’ve had training on how to make Shakespeare engaging or how to incorporate ICT in my classroom, yet I haven’t really had any training on reading. CPD is often related to lots of ‘nice ideas, but where’s the pedagogy?’. 

If I am going to help students get better at making inferences, and more credible and convincing inferences, then I need to understand a bit more about how inferences are formed and how we can support them. Me asking a question alone doesn’t make students better inferers (eww - don’t like that phrase). Informing a student that a statement is supposition and not an inference helps them to reevaluate that original statement and build a better inference. Let’s mock up an example to prove a point. 

Student: I think that the Jag belongs to a headteacher. 

Teacher: Where’s your evidence for it to be a headteacher and not a consultant? 

The question leads students to explore the evidence they have and refine their inference. But, we don’t address supposition in lessons enough. If we think of how we discuss texts, a large part of the time is building inferences and not interrogating inferences. If there’s any supposition, we tend to seek out another idea from the same or a different student. We don’t interrogate how the student led themselves to that idea. In fact, we don’t interrogate inferences full stop. I’d argue we rely on students making so many inferences in lessons, yet we focus on the answer rather than the process. We listen out for good inferences and praise them with the idea students will repeat the process naturally or through some kind of osmosis. 

To explore the process of inference, all teachers need to know something about the process. Skimming and scanning is problematic and dominates schools, but all exams and teachers want students to make ideas about a text and not just find bits of information. The following is a simplification of key terms in Anne Kispal’s 2008 report on inference and it makes a good starting point for this aspect of reading.  

When does the inference take place? 

Online inference - during the process reading 

Offline inference - after the process of reading 

For lots of teachers, we make inferring a large offline experience. We have read the text and now let’s have some ideas. That’s why questioning in the reading process helps students. It models what we naturally do: build up a range of inferences during the reading process. If we are going to model effectively then we need to model the build of inferences during the process. 

Where in the text is the inference?  

Local inferences - a specific line or phrase 

Global inferences - whole text 

Good readers will use a combination of local and global inferences to build an idea. And some even more effective readers use precise local inferences to address inconsistencies and contradictions. Some of our weaker readers rely on just local inferences and we, as teachers, need to model how an idea can be constructed globally and locally. 

What in the text is being connected?  

Coherence inferences - a inference that links knowledge across a text 

Elaborative inferences - gap filling inference - using prior knowledge to form an inference 

Elaborative inferences are, usually, what we think of when we discuss inferences. Coherence inferences are probably the most underused and considered aspects of reading. We do it so automatically when reading a story, yet for non-fiction is a massive aspect students need to master. What does the ‘it’ and ‘this’ refer to? When we’ve read a story about a character called Tom, we get it is Tom the story refers to when the pronoun ‘he’ is used. Yet, in non-fiction you don’t have that narrative simplicity or support and you’ll have a cast of twenty people and twenty things to attach ‘this’ or ‘it to. ’ 

So once you have that understanding of inferences we can help support students better in lessons. We can put emphasis on students making online and not just offlines inferences. We can ensure students make local and global inferences. We can support students with coherence inferences across a text. 

Then, we can interrogate precise inferences and discuss them as practitioners. What would you infer from a line? What would be a good inference? What do students need to do to get that good inference? What skills and knowledge are they using? 

Here’s one to start off with. It is from Neil Gaiman’s ‘The Graveyard Book’ and it relates to a murderer wearing a type of glove.

What can we infer from the fact that he wore gloves made of the ‘thinnest lambskin’? 

Answers on a postcard or Twitter. 

I will be carrying on this reading thread in the next few blogs. 

Thanks for reading, 

Xris 

P.S. The document I refer to can be found here. Effective Teaching of Inference Skills for Reading - Anne Kispal (2008).

https://www.nfer.ac.uk/publications/EDR01/EDR01.pdf