David Harvey
Retired district court judge
Introduction
I have been a reader all my life. For some reason decoding letters and making sense of them in the context of reading came easily. I could read before I went to school and no books were forbidden in our household when I was young – a rule I maintain to this day. I work in a room surrounded by books – my happy place.
The decline in literacy is a problem and has been with us for some time. When the “experts” decided that phonics was no good and a “whole language” approach was more attuned to comprehending meaning, literacy was on a slippery slope. Couple that with the use of keyboards rather than pencils and kids today cannot read cursive script because they have been brought up on typescript on a screen.
This article is about reading and the current malaise which is at last being addressed in our schools. It advances a case for continuing the current programme and sticking to the course. We deprive our young of their future if they cannot read properly and write elegantly.
Whether or not the results of the election and the cobbling together of coalitions will change the focus has yet to be seen. I imagine the left will want to reassume the dumbing down position led by the lanyard and ‘taonga’ wearers of the PPTA. And that would be a tragedy.
The Malaise – A Collapse in Reading
New Zealand is not the only country to suffer from a decline in reading. It has been a problem across the developed world. The causes are complex. It is simplistic to blame smartphones and teenagers’ use of them. The evidence suggests the problem goes further back – to the first years of school.
A child’s inability to decode the symbols that make words fluently at age five or six does not develop the stamina, vocabulary and background knowledge that deeper reading demands. The problem then compounds year by year. It results in weak comprehension at 15 and an inability to handle complex texts at university and in later life.
Rose Horowitch’s essay in the Atlantic (August 2026) sets out the scale of the decline in the United States. The fall in reading cuts across every age group, gender and level of education. Even the groups who have traditionally read the most – retirees, women and university graduates – have fallen away.
Horowitch’s central theme is that as far as America is concerned it is not just illiterate but postliterate. She argues that people can still decode individual words, but they are losing the higher-order abilities of comprehension and synthesis that make reading worthwhile.
The statistics tell the story. Horowitch sets them out.
In 2004, 28 per cent of American adults read for pleasure on any given day; by 2023 that had fallen to 16 per cent. Fewer than half of adults read a single book of any kind in 2022.
Among 13-year-olds, the number who say they rarely or never read for fun has risen from eight per cent in 1984 to 29 per cent in 2025, and children like reading less with every year they grow older.
Basic skill has slipped too: nearly 30 per cent of adults now cannot paraphrase or draw inferences from a multipage text, up from under 20 per cent in 2017.
Meanwhile, the average attention span on a screen has shrunk from around two-and-a-half minutes in 2004 to about 47 seconds.
That decline in standards appears at every stage in education, which suggests that the problem starts early and is perpetuated throughout the system. And it doesn’t stop at school but continues into university.
In a 2024 study at two Kansas universities, English majors were asked to read the opening paragraphs of Dickens’s Bleak House. A quarter of them took its figures of speech literally, concluding that dinosaurs were walking the streets of Victorian London, and only five per cent came away with an accurate, detailed understanding. Even at Harvard, staff report students who regard reading whole texts as an unnecessarily difficult way of acquiring information.
Horowitch suggests that some responsibility lies with smartphones and short-form video that have displaced the quiet, sustained attention that reading requires.
But in addition brain science explains why all of this matters. Reading is not natural. We don’t have any built-in machinery for it. The brain must adapt areas that are designed for speech and object recognition and neuroscience tells us that the brain needs exercise – it masters what it practises and drops that which it does not. If you don’t keep reading over time it becomes harder and harder to make sense and meaning from the written word.
The Importance of Reading
Reading is the foundation of sustained analytical thought. Literacy made possible philosophy, modern science, history and the informed citizenry democratic institutions assume. Walter Ong and Neil Postman have argued that writing enabled extended focus and logical deduction in ways oral culture did not.
Horowitch argues that a postliterate society reverts to habits of oral culture: slogans over argument, emotion over evidence, and personality over record. Media theorist Joshua Meyrowitz and commentators across the political spectrum note how social platforms reward simple, inflammatory content over nuance. A population that cannot follow a long argument is easier to mislead.
The New Zealand Picture
Derek Cheng, writing in the Herald has demonstrated that reading and maths have plummeted since 2018, and the hoped-for return to pre-Covid levels did not come.
Over the life of the study, New Zealand’s average reading score has fallen from 529 in 2000 to 497 in 2025, while the OECD average has fallen from 494 to 461, with most of the OECD’s decline coming since 2018.
An OECD report matched some of the findings in the Horowitch analysis. There is a shrinking capacity for careful reading and sustained attention. Students are more likely to give hasty, superficial answers. OECD education director Andreas Schleicher points to falling reading for enjoyment and the shift to skimming social-media feeds.
New Zealand’s scores held steady between 2022 and 2025 while the OECD average kept falling. But holding steady is not the same as recovering. The Herald article statistics show that in 2025, 23 per cent of New Zealand 15-year-olds were below that floor, up from 21 per cent in 2022. Reading was the only subject in which this group grew. These are young people who will leave school struggling to understand the everyday texts that adult life, work and citizenship require.
The longer view is even more sobering. The average New Zealand reader in 2025 was more than a year of schooling behind their counterpart in 2009, and once sampling bias is taken into account the gap is closer to two years.
Dr Nina Hood of The Teachers’ Institute told the Herald that achievement fell sharply from around 2010–12 and that there has been no real shift towards improvement since.
In other words, New Zealand has stopped falling for now, but it has not yet begun to climb back. The OECD also advises caution with New Zealand’s 2025 figures, because the sample did not meet its participation standards.
Changes in Approach
Since term one 2025, every New Zealand primary school has been required to teach reading through structured literacy, alongside a new English curriculum.
Before this, the way children were taught to read depended heavily on the school they attended and even the teacher they happened to get.
Some schools had already adopted structured approaches; others relied on approaches, such as using pictures and context to guess unfamiliar words, that structured-literacy advocates argue leave many children behind.
The central change is that teaching is now nationally consistent and grounded in the research on how children actually learn to read.
Structured literacy itself is straightforward to describe. Children are taught explicitly and in a planned sequence how written letters correspond to spoken sounds, how to blend those sounds into words, and how words are built.
Once decoding becomes automatic, the method builds towards fluency, vocabulary and comprehension. The government committed $67 million to the rollout, including training for teachers.
The reforms then add a system for catching problems early.
New entrants take a short phonics check after 20 weeks and again after 40 weeks at school, reading aloud a list of 40 words, some real and some invented, to show whether they can genuinely sound out unfamiliar words rather than recognise memorised ones.
The checks were available in 2025 and became compulsory in 2026. Their purpose is diagnostic: to tell teachers, in a child’s first year, who needs extra help.
That help is now more readily available, with 349 full-time-equivalent specialist structured literacy teachers working across more than 1,240 schools from 2026.
From year three onwards, students’ progress is checked twice a year in reading, writing and maths, including higher-level skills such as comprehension.
Around this core sit several supporting measures. Schools are expected to give an hour a day to reading, writing and maths, ensuring the basics receive sustained time.
A knowledge-rich curriculum aims to give children the background knowledge and vocabulary that comprehension depends on.
And cellphones have been banned during the school day since 2024, removing one of the main competitors for children’s attention in the classroom.
Is This The Answer?
The strength in the new approach is that it responds to the causes in the decline and not just the symptoms. It aims to make sure that every child gets the foundations that make reading sufficiently rewarding to compete with smartphones.
And let me repeat an important point. Reading does not come naturally. It has to be learned. It has to be taught. It is a skill that the brain must be trained to perform and it cannot be left to chance, to the home or to the whims and preferences of individual schools.
Structured literacy treats it accordingly, teaching every child the code explicitly and giving them daily practice. This matters most for the children the Atlantic and the Herald are most worried about.
When only a tiny share of adults read to children at home, the school becomes the only reliable place where a child learns to read well.
The Herald’s sources note that children from poorer homes have too often been given lower-level opportunities to learn; a single, consistent, explicit method gives them the same chance as everyone else. It is no accident that the largest early gains have come among Māori, Pacific and low-decile students.
The reforms also address the gap between decoding and understanding that sits at the heart of the postliterate problem.
It might seem that phonics, a skill about sounding out words, has little to do with the comprehension failures the articles describe. In fact the two are closely linked. A child who has to labour over every word has no mental capacity left to follow the meaning of a sentence, still less an argument across several pages. Fluent, automatic decoding frees up that capacity.
Comprehension also depends heavily on what a reader already knows. The Kansas students struggled with Bleak House partly because they did not know what the Lord Chancellor or Michaelmas term were, and did not think to find out.
That is why the knowledge-rich curriculum is an essential partner to structured literacy, and why progress checks from year three test comprehension as well as word reading.
Timing is another reason the approach makes sense. The Atlantic article describes reading as a kind of workout. The more you do it, the easier and more pleasurable it becomes, and the less you do it, the harder it gets.
Small early differences therefore grow into large later ones. A child who reads fluently at six reads more at eight and more again at 12, steadily building vocabulary and stamina.
A child who struggles at six learns to avoid reading and falls further behind each year. By the time problems surface at secondary school or university they are very hard to reverse. The phonics checks at 20 and 40 weeks are designed to intervene at the point of greatest leverage, before the gap opens.
There is good reason to believe this works. Structured literacy rests on decades of research into how children learn to read, and support for it crosses New Zealand’s political divide.
Labour says it began the work but the current government made it universal.
England offers a useful precedent. After it introduced a phonics screening check for six-year-olds in 2012, the share meeting the expected standard rose from 58 per cent to 82 per cent by 2019.
In the 2021 PIRLS study of primary reading, England ranked fourth of 43 countries and was one of only 11 to avoid a significant fall in scores during the pandemic period.
Finally, the reforms recognise the very real problem of distraction. The Atlantic article reports that after Texas banned phones in schools, one Dallas school district saw 200,000 more library books borrowed in a year, an increase of nearly 25 per cent.
New Zealand’s 2024 ban anticipated that finding, giving children time in class free from the devices that compete most fiercely for their attention.
It Won’t Happen Overnight
Early results are encouraging. In the first year of structured literacy, the share of new entrants at or above expectations on the 20-week phonics check rose from 36 per cent in term one to 58 per cent in term three, an increase of 22 percentage points in little more than half a school year.
The improvement went beyond simply getting more children over the line. The share of children exceeding expectations more than doubled, to 43 per cent, and the proportion needing targeted support fell from 52 per cent to 33 per cent.
Just as importantly, the gains were shared. Māori students at or above expectations rose from 25 per cent to 43 per cent, Pacific students from 27 per cent to 43 per cent, and children in low-decile schools from 18 per cent to 35 per cent.
But the results are leading indicators rather than a final verdict. They demonstrate that children are learning the foundational skills that the reforms target and are learning them earlier.
But the real issue is whether that is going to translate to a stronger 15-year-old reader and the answer to that is going to take time and not only time but a willingness to stay the course. The first cohort to be taught entirely under the new approach started school in 2025 and will not turn 15 until the mid-2030s.
England’s experience shows how long such a lag can be. Its phonics check began in 2012, but the international payoff became visible only in later rounds of the PIRLS study, years after the first children had been through the system.
Minister of Education Erica Stanford, herself, has acknowledged that New Zealand’s results remain well below where they were 15 years ago and that there is a long way to go.
This is the strongest argument for patience. A reform aimed at five-year-olds cannot show up in the scores of 15-year-olds for a decade. If it is judged prematurely on PISA results, or abandoned at the next change of government, New Zealand would repeat the very inconsistency that allowed the decline to happen in the first place.
The problem is that this programme could be abandoned – a worst case scenario – or modified to the point that it is tantamount to abandonment.
Some educators and the Labour Party see the checks as standardised testing by another name, warning of a one-size-fits-all system that echoes the old National Standards.
The answer is that the checks are designed as a diagnostic tool to trigger support for individual children, not to rank schools.
Labour itself backed structured literacy. Any disagreement is about the pace and universality of the change rather than its direction.
Conclusion
Rose Horowitch concluded her article with the image of the Library of Alexandria where knowledge was not so much lost to fire as it was to neglect and a loss of the will to preserve it. Today’s Library of Alexandria exists on servers. What is at risk is the ability and inclination to read them.
New Zealand cannot change the global attention economy, but it can make sure every child leaves their first years of school able to read fluently. That is what the Stanford reforms are built to do.
The early-years data is encouraging, the international precedent is supportive, and the logic is sound. The test now is patience: holding a consistent course for the decade it takes a five-year-old reader to become a 15-year-old one.
Post Script
I don’t usually set out a list of sources for my writing but in this piece I have made reference to the work of others. Rather than footnote each reference I have compiled a list of readings that provided material to which I have referred directly or have synthesised into this piece.
I especially recommend the Atlantic article by Rose Horowitch and the Herald article by Derek Cheng. It was those two articles that started me thinking about what has become this article
Sources
• Rose Horowitch, “The End of Reading Is Here”, the Atlantic, August 2026
• Derek Cheng, “What shorter attention spans have to do with global education results plummeting to records lows”, NZ Herald, 2026
• Ministry of Education, Terms 1 to 3 Phonics Checks results, October 2025
• 1News, Minister hails structured literacy for boosting new entrant reading, October 2025
• 1News, Phonics checks for students from first year at school, July 2024
• Inside Government, PISA results show room for improvement, September 2026
• Inside Government, New schools data shows student progress in maths and reading
• OECD, PISA 2025 country note: New Zealand
• UK Department for Education, England moves to fourth in international rankings for reading, May 2023
• Education Business UK, PIRLS and phonics screening check results
• The Conversation, Phonics screening check: why this test isn’t needed
This article was originally published by A Halfling’s View.
