Schools Never Taught Reading. The Digital World Just Exposed It.
At every education conference, the same argument surfaces at some point. Children today can’t handle texts. They used to read more. We need to return to tradition.
I’d like to agree. It would be simpler.
But I have a fundamental problem with this argument — and it has nothing to do with the digital world. It has to do with what we actually mean when we say “back then.”
Nostalgic statistics and what they don’t measure
Yes, more pages of books were read per student per year. That’s true. But what does it mean that someone “read”? That their eyes moved across lines of text? That they could summarize a plot? That they knew what the teacher expected from an interpretation?
Because that is not understanding. That is the performance of understanding.
For decades, school had a very effective tool for concealing the absence of comprehension: compulsion and fear. Students learned to simulate understanding — summary by template, quote in the right place, interpretation aligned with the answer key. This system worked brilliantly as a grading machine. As a tool for building genuine competence — it didn’t work at all.
Many of us truly learned language — really learned it — only at university. Not because we suddenly became smarter. Because for the first time there was context: the text was a tool for thinking about something we actually cared about. There was someone who read with you, not someone who checked whether you had read. There was safety that allowed admitting confusion without punishment.
The digital world didn’t destroy reading comprehension. It removed compulsion and fear — and revealed that genuine comprehension was never fully there to begin with.
Where does understanding actually come from?
This is not a technical question. It is a question about the architecture of learning.
Reading comprehension is not the result of exposure to text. It doesn’t follow from the number of pages read or the number of years spent in a classroom. It is the result of four things simultaneously:
A cognitive schema — to understand a text, you need something to attach it to. Knowledge of the world, experience, prior reading. A student who doesn’t understand a nineteenth-century novel often fails not because the text is too difficult, but because they don’t yet have the map of the world on which that novel makes sense.
Intrinsic motivation — active processing of text, which is what we colloquially call comprehension, only activates when the reader has a reason to read. Not an external reason — “because there’ll be a test” — but an internal one: curiosity, need, a question the text might answer.
Psychological safety — comprehension requires willingness to admit incomprehension. This is an act of cognitive vulnerability that is only possible in an environment without punishment for not knowing. Traditional school systematically destroyed this condition.
Modeling the process — someone has to show how to read. Not what to say about a text, but how to enter a text: to ask it questions, to lose the thread and find it again, to not understand a passage and keep searching. This was almost never taught.
Traditional language education provided at best one of these conditions — exposure. And it punished students for the absence of results from that exposure, which destroyed the third condition. This is a structural contradiction built into the system.
Text hasn’t disappeared. The teacher of text has.
The paradox of the contemporary student is that never in history has a teenager read more text than today. Comments, messages, posts, articles, podcast transcripts, product descriptions, terms and conditions clicked past without reading. Text is everywhere, all the time.
The problem is not a shortage of text. The problem is that the only teacher of reading in this environment is the algorithm. And the algorithm teaches one thing: click on whatever triggers the strongest emotional response.
This is a lesson in rhetoric — but the rhetoric of manipulation, not comprehension. And students absorb it brilliantly, because it is delivered in an environment that rewards immediately, doesn’t punish for misunderstanding, and requires no simulation of competence.
School lost to the algorithm not because the screen is stronger than the book. It lost because the algorithm understands the conditions for learning better than most educational systems: intrinsic motivation, immediate reward, personalization, no punishment for error. Of course — it uses this knowledge for purposes that have little to do with education. But the mechanism works.
What digital literacy is — and why it belongs to language teachers
Digital literacy is not how to use a computer. It is not technology. It is something that the humanities has been doing for centuries — just in a new environment.
It is understanding how text works: why a headline is constructed one way and not another, how a disinformation narrative is built, why some sentences stay in your head and others vanish. This is rhetoric. Stylistics. Text analysis — only using material the student encounters every day.
It is critical source evaluation: who wrote this, why, what they left out, who they cite and why. This is journalistic and historical craft — but above all, humanistic. Someone has to show students this. A math teacher? A physics teacher? No. This is at the core of language work.
It is awareness of algorithms as editors: students should understand that their feed is not a window on the world but a mirror of their previous clicks, amplified and radicalized by a system designed to maximize engagement. This is knowledge about communication, about the power of language — always the domain of the humanities.
It is writing in different registers: post, newsletter, essay, argumentative thread, podcast script. We teach one register. We examine one register. And then we wonder why students “can’t write” — because they can’t write this specific format that school recognizes as writing.
And finally: it is understanding what AI generates and why. Every student uses or will soon use generative tools. Analyzing AI-generated text — where is the author, where is the point of view, why does this text feel flat — is one of the best stylistic exercises available today.
Language education already has all the tools for this work. It has had them for decades. The problem is that it often prefers not to use them on new material.
The false dichotomy
Paper book versus screen is the wrong division. This is not a question of medium.
The real question is: are we teaching reading as an act of critical thinking, or merely as a technique for decoding symbols and producing answers that match the key?
If the latter — it doesn’t matter whether we do it with a nineteenth-century classic or an iPad. The result will be the same: a student who can reproduce but doesn’t understand. Who can write an essay but can’t write an email to an employer. Who knows the canon but doesn’t notice they’re being manipulated by the text they’re currently reading.
Retreating from screens will not solve the comprehension problem. The solution is teaching comprehension — on every text, in every medium, with a clear model of what that process looks like from the inside.
What can be done differently
A few concrete directions that require no systemic revolution — only a decision:
Introduce the analysis of digital texts as a legitimate genre. Fake news, viral posts, public debate commentary, narrative advertising — these are texts with authors, purposes, and structures. They can be analyzed with tools language education already possesses.
Model the reading process, not just its output. Read together with students, aloud, showing confusion, backtracking, asking questions of the text. This is a lesson in metacognition hidden inside a language lesson.
Teach writing in different digital registers. Post, newsletter, argumentative thread, script — this is not abandoning literature. It is expanding the craft to forms students will actually use.
Work with AI-generated text as text. Generating and critically analyzing AI output sharpens stylistic sensitivity better than many traditional exercises.
Build bridges between the canon and the present. Why is a nineteenth-century novel still relevant? Because its characters have imposter syndrome, because they live off identity performance, because they’re trapped in information bubbles. These connections are real and they engage students — but someone has to draw them.
This is not an attack on tradition
It is an invitation to see what tradition actually offers.
Language education holds a unique position in school. It is the only subject whose core is language, communication, narrative, and critical thinking. No other subject has such tools.
The problem is not that children are different. The problem is that the environment is different, and the method has stayed the same — a method that, in truth, never fully worked, but effectively concealed its failures.
Students don’t read not because the screen made them lazy. They don’t read because no one has shown them that reading is a tool of power over their own world. That understanding a text means not being manipulated. That writing well means existing in a way that carries weight.
That is the most profound lesson a language teacher can give. But to give it — they must first believe that the environment where their students actually live is worthy of that lesson.