How do we read? This question has been at the forefront of experimental psychology since the early days of the discipline.
Edmund Huey, an American psychologist, was among the first to use experimental methods to investigate the science and pedagogy of reading. But writing in 1908, he recognised the huge challenge facing reading researchers.
“Problem enough, this, for a life’s work, to learn how we read!” he wrote. “A wonderful process, by which our thoughts and thought-wanderings to the finest shades of detail, the play of our inmost feelings and desires and will, the subtle image of the very innermost that we are, are reflected from us to another soul who reads us through our book.”
Over the past 100 years, psychologists have made immense progress in understanding this “wonderful process”. We have a very good understanding of the cognitive operations involved in skilled reading – how we as fully literate people read – and we also know a great deal about how children learn to read. From this viewpoint, it is perplexing to witness the so-called “reading wars” – the controversies and debates about how reading should be taught – which have plagued educational policy and practice for decades.
Teaching children how to read is surely one of the most important jobs of the state. After all, being able to read well provides a young person with access to education and learning, to employment, to health and wellbeing, and to culture. It allows them to contribute to and benefit from society. Every child has a right to read well, but the Organisation for Economic Cooperation and Development (OECD) estimates that about 20 per cent of 15-year-olds in developed countries have unacceptably low levels of literacy, which is blocking them from participating effectively in society.
We have recently considered how the science of reading can be better harnessed to help teach reading and break the cycle of disadvantage. By bringing together more than 300 research studies, journal articles and chapters, we have produced an overview of what is known about how children learn to read. In this wide-ranging review, Ending the “reading wars”: reading acquisition from novice to expert (bit.ly/ReadingWars), we also explain some of the barriers and confusions that have unnecessarily fuelled the reading wars. What follows is a summary of our paper, exclusively in the UK for Tes readers.
1. Why phonics has to be the first step
The term “reading” has breadth. If a teacher said “Alice is the best reader in her class”, what would she mean? It might mean that Alice is able to recognise words and read them aloud accurately. It might mean that Alice reads a lot and can readily learn new information from text. Or perhaps Alice is adept at literary analysis. Reading means different things to different people and it also means different things depending on the age of the child.
Reading is the act of constructing meaning from print – transforming symbols on the page into the message intended by the writer. This transformation requires the orchestration of a large number of cognitive and linguistic operations, from recognising the identity of a letter – even when it is written in a different case or font – to inferring the mental state of a character described in a story.
The challenge facing beginning readers (and their teachers) is substantial. Three sets of skills need to be acquired: cracking the alphabetic code; becoming fluent at recognising words; and understanding written text. These are not completely independent capacities that develop separately, one after the other. But there is a strong consensus that learning to crack the alphabetic code – ie, understanding the relationship between printed words and spoken language – is a necessary foundation for learning to read.
When we as skilled adults encounter a word while reading, we do not normally linger. Our sense is that we see a word and immediately link to its meaning – and often we do: we quickly understand the difference between sail and sale, even though they sound the same. We do not consciously “sound out” words, although we can when we need to – and sometimes it is hard not to. Take the last word in this passage from Roald Dahl’s The BFG: “You must not be giving up so easy...the first titchy bobsticle you meet and you begin shouting you is biffsquiggled.”
If adults routinely draw on automatic connections between print and meaning, one might conclude that this is what children need to be taught to do from the outset. But this is not the case. The evidence base is clear in showing that the journey towards establishing strong connections between print and meaning starts with children establishing links between print and sound. This is non-negotiable – not because of ideology or pedagogical preferences – but because it is how our alphabetic writing system works.
One reason why the reading wars have persisted is because of the view that children can learn to read in the same way that they learn to speak and understand. But reading is not like spoken language.
Like learning to walk, children are born with the ability to acquire spoken language simply through interaction with their environment. We have no such predisposition for learning to read. Presented with a library of books, a child will not usually learn to derive meaning from the sets of curves, lines and dots that make up writing. Instead, reading is a learned skill that typically requires instruction.
Precisely how children learn to map those curves, lines and dots onto meaning depends on the nature of the writing system that they are learning. Imagine the task of a child learning the meanings of the printed words “cat”, “cut”, “can” and “rat”. These share some visual symbols, but they have totally different meanings.
Faced with trying to learn these meanings, the child might develop a strategy that whenever the first symbol is different, the word refers to a rodent-like creature. This strategy might work for a while, but will quickly unravel when faced with new words like “ran” and “rot”. This example shows why encouraging children to guess words from context, or to rely on salient visual cues, results in gains that are short-lived at best.
Fortunately, learning to read in English does not require children to memorise individual words. Its alphabetic writing system means that the visual symbols represent the sounds of the language. Turning back to the previous example, once a child learns the letter-sound relationships in “cat”, “cut”, “can” and “rat”, they can use their knowledge of the sound of each word to access its meaning. They also have the capacity to generalise – to read new words that they’ve never seen before – like “ran”, “rut”, “nut” and “ant”.
Alphabetic systems are extraordinarily efficient – they allow a vast amount of information to be communicated using a small amount of code. It’s not surprising that they are so common across the world.
To read independently, children must be able to understand the alphabetic code, and the purpose of phonics is to teach this directly. Myths about phonics abound: “it teaches children to read nonsense”; “it interferes with reading comprehension”; “English is too irregular for phonics to be of any value”; and “phonics is boring and puts children off reading”. Our research explains why all of these claims are misguided. Phonic knowledge provides the necessary foundation for what comes later, but it is only part of the story. It is necessary, but not sufficient.
2. The importance of morphemes and morphology
Once children have mastered the alphabetic principle, they have the starter kit to read words for themselves and to begin to develop the rapid links between print and meaning that underpin skilled reading. This is a slow process of building expertise in which children harness their powers of perception, memory and language to learn and generalise from their experience with print. But how does this expertise develop and how can it be fostered in the classroom?
One implication for teaching takes us back to the nature of the English writing system. When first learning to read, children encounter short words in print where the letters are code for sounds, as in “lock”, “pick” and “pack”. But as children increasingly encounter longer words, letters and groups of letters also become code for meaning, like in “unlock”, “unpick” and “unpack”. Here, the letters “un” are contributing something important about meaning in a way that can apply to many other words, too. This type of coding is known as “morphology”, and a “morpheme” is a linguistic term used to describe the minimum meaning-bearing unit of language.
Most English words are built by combining morphemes – for example, “unlock”, “lockable”, “relocked”, “locksmith”, “headlock” – and we know that skilled readers use knowledge of morphology to compute the meanings of words rapidly. So it makes sense that there should be explicit instruction on this aspect of the writing system, once alphabetic knowledge is established.
One barrier to this is that teachers are not always taught about morphology. Many are increasingly skilled in phonics – quite appropriately – but are less aware of the ways in which morphemes communicate meaning and govern spelling. This suggests a possible gap in teacher training.
3. Reading experience counts
Children need more than explicit instruction on how the writing system works. They need many years of reading experience. It is well known that children who are good readers tend to read more, and in so doing, become better at reading. Establishing this virtuous circle must be a central goal of primary schooling. Motivating poor readers to read is not easy, but the question of how to do it should not be divorced from the question of how best to teach them. On the contrary, one clear and achievable means of maximising motivation is to ensure that a child has solid basic skills and considers being “a reader” a key part of their identity.
This, in combination with sufficient text experience, produces a powerful word-reading system that allows children to recognise the vast majority of the thousands of words they see each day, efficiently and easily.
4. Teaching the means to create meaning
So far, we have been talking about words. Being able to read words well is, of course, the critical “front end” of reading, but constructing meaning from text requires much more, and “construction” is a key term.
As we read, we dynamically build a mental model, sometimes called a “situation model”, culminating in a rich interpretation of the text that goes far beyond what is explicitly stated. In most cases, comprehension is not simply a verbatim record of what has been read.
To illustrate, consider this short text: “Denise was stuck in a jam. She was worried what her boss would say.”
To make sense of this, we need to recognise and understand the words and their contextually appropriate meanings – that “jam” probably refers to a traffic jam, not the fruit preserve. We need to go beyond individual words and make connections between them – that “she” and “her” in the second sentence refer back to Denise. On reading the text, we are likely to conclude that Denise was on her way to work, but was running late due to heavy traffic. But the text doesn’t say this explicitly – it needs to be inferred using our knowledge of the meanings of words, the rules of syntax, our background knowledge and appreciation of how the world works.
Our knowledge of Denise, her boss and the situation might prompt us to elaborate further, and what we conclude from the text will be coloured by this. Perhaps Denise is always late? Maybe it’s an important meeting that she cannot miss? Perhaps her boss is unreasonable and Denise fears getting the sack? Or maybe there’s no traffic at all and Denise is stuck in a metaphorical jam. There are other possibilities, too, all licensed by this simple, two-sentence text.
As we read, the text prompts us to pull in relevant knowledge, and we use our language and memory resources to process this information as we read – to make connections between elements of the text and to draw inferences about the intended meaning. The enormous literature on the psychology of reading comprehension makes clear that it is not a single entity that can be taught directly. Nevertheless, there are clear pointers as to what should guide effective teaching.
One important conclusion is that, beyond reading words, many aspects of reading comprehension are not specific to reading. Instead, they are features of language comprehension more broadly. What follows is that those who start school with low levels of language are at a huge disadvantage from the outset when it comes to reading comprehension. As recent discussions of “the language gap” and “the word gap” have made clear, explicit instruction in oral narrative and vocabulary needs to start early – and to continue throughout the school years. Not only does this promote language and oracy, it benefits reading comprehension, too.
Once children are able to read fluently, there is clear evidence that explicit instruction in strategies to help comprehension has an impact. This includes instruction in clarification, summarisation, prediction, and in helping children to make inferences and to actively monitor their comprehension. These strategies can be learned quite quickly and applied to new reading material.
Importantly, however, strategies can only take us so far. Without having access to the relevant knowledge to understand the content, comprehension will fail. And note, too, that these reading strategies do not help beginner readers learn to read words. Instead, they depend on children being able to read words accurately and fluently – the critical “front end” of reading discussed earlier.
While strategy instruction might be quite quick, the acquisition of knowledge, discipline-specific language and cultural literacy is gradual and continuous. It relies on rich input, much of which will come from reading experience itself. The implication of this is clear: teach children to read and then provide opportunities for varied, extensive and successful reading experiences.
5. Time for a truce
Our research has led us to draw two conclusions about why the reading wars have continued. First, phonics has been unfairly criticised, partly because people have not always understood why phonics works for alphabetic systems. And while practitioners know there is more to reading than alphabetic skills, a full discussion of other aspects of reading acquisition has been lacking. As a result, calls for a greater focus on phonics instruction can seem unbalanced.
The evidence is there to get the balance right. Though the term “balanced literacy” is in widespread use, it is too often used to describe approaches with “a bit of everything”. In our view, it is time to reclaim a term such as balanced instruction and recast it in a more nuanced way that is informed by a deep understanding of how reading develops.
This requires training teachers in how the writing system works, and how the broader language system it represents underpins the process of becoming a reader. Reading is multifaceted – children must learn alphabetic decoding, fluent word reading and text comprehension – but this does not mean that instructional time should be devoted to all of these skills at all times.
Rather, instructional regimes to support these various abilities are likely to be most effective at different and particular points in development, and precious teaching time should be structured to reflect this.
Returning to Huey’s words from more than 100 years ago, we are reminded not just of the magnitude of what we do when we read, but also of the status of reading as an extraordinary cultural invention. He argued that to “completely analyse what we do when we read” would be at the pinnacle of achievements “for it would be to describe very many of the most intricate workings of the human mind, as well as to unravel the tangled story of the most remarkable specific performance that civilisation has learned in all its history”.
Those tasked with the difficult job of teaching reading are opening children to this wonderful process – and all that follows from being able to read well. The evidence is now there to support teachers with this endeavour, and to finally end the reading wars.
Kate Nation is professor of experimental psychology at the University of Oxford and directs its language and cognitive development research group. Kathleen Rastle is professor of cognitive psychology at Royal Holloway, University of London, and directs its language, learning and cognition lab. Anne Castles is professor of psychology at Macquarie University in Sydney and deputy director of the ARC Centre of Excellence in Cognition and its Disorders