Education is relational at its core. Not because relationship sounds warmer than method, but because understanding becomes exact only where someone can see another person thinking. What a teacher can do in a single hour one-on-one with a child, no curriculum, no classroom, and no machine can do. This text describes a practice that takes that simple truth seriously, and why it is needed now.
Premise
A child sits with an open textbook and does not understand what they are reading. They do not know that they do not understand. They read the words, the words make sentences, the sentences make pages — and nothing in the process gives the child any signal that they are taking nothing in. The decisive fact is not that the child does not understand. It is that this can remain invisible. Anyone who has ever spent an hour one-on-one with a student has seen it.
This is not the failure of a single child. It is the functioning of a system that was not made for that child and cannot be fully remade around any single child. Modern public schooling — in Switzerland and across industrial societies — took its present basic form in the nineteenth century. It was meant to give children a shared foundation, make democratic participation possible, and order a society becoming faster, more technical, and more specialised.11On the history of Swiss public schooling, see Lucien Criblez (ed.), Bildungsraum Schweiz: Historische Entwicklung und aktuelle Herausforderungen (Haupt, 2008), especially Lucien Criblez / Christina Huber, “Der Bildungsartikel der Bundesverfassung von 1874 und die Diskussion über den eidgenössischen ‘Schulvogt’,” pp. 87–130. See also EDK, “150 Jahre Bildungsartikel in der Bundesverfassung” (29 May 2024), on compulsory primary education under Art. 27 of the 1874 Federal Constitution. For the broader institutional-historical context: Ivan Illich, Deschooling Society (Harper & Row, 1971), and David F. Labaree, Someone Has to Fail: The Zero-Sum Game of Public Schooling (Harvard University Press, 2010). It succeeded. Public schooling has, in Switzerland and elsewhere, taught generations to read and write, enabled social mobility, and shaped citizens able to sustain a country. There is no reason to be ungrateful to it.
Gratitude does not change the fact that this form reaches its limit where the task is not merely to deliver material, but to understand a single act of thought. The more stable world of work to which school could long be oriented no longer exists in that form. Adults will need more: judgment in unclear situations, the capacity to work alongside machines without disappearing into them, and the willingness to keep learning when no one is producing a curriculum. For this, public schooling is not reliably built. A whole class, a curriculum, a timetable, an exam at the end: these are the tools it has. They are not the tools with which one reaches every student at their own pace, with their own question, in their own language.
The sharpest criticism of this form is not that school accomplishes too little. It is that school, especially when it works on its own terms, can produce a confusion it can hardly correct: learning becomes equated with progress through an institution.22Ivan Illich, Deschooling Society (Harper & Row, 1971), especially the critique of the institutionalisation of learning; David F. Labaree, Someone Has to Fail: The Zero-Sum Game of Public Schooling (Harvard University Press, 2010), on the logic of school sorting and credentialist expectation; Paul Goodman, Compulsory Mis-Education (Horizon Press, 1964), as an earlier American formulation of the same unease. Ken Robinson later popularised parts of this critique, most visibly in Out of Our Minds: The Power of Being Creative (Capstone, 3rd ed. 2017) and the TED talk Do schools kill creativity? (2006). Whoever completes tasks, passes exams, and finishes years counts, in the language of the system, as successfully learning. Whether understanding has arisen, whether a question of one’s own has appeared, whether what was learned remains alive outside the examination, is a different matter. For a single child, that confusion matters. It makes it possible for a child to be occupied by school and absent from learning.
The question of this text is therefore not how the system ought to be reformed. That question matters, but it is not the task here. The narrower and more honest question is this: what can happen beside school, without pretending to replace school? What is possible for a child now, in an hour where the system briefly recedes? What can a teacher do with an hour, a child, a table, a piece of paper, and nothing else?
The answer begins with what is available only in that hour: not the material, not the method, but the relationship itself.
The Relational Core
What a teacher can do in an hour with a child, no one and nothing else can do in that moment. This is not a claim against curriculum, classroom, or machine. It is a claim about precision: in the individual hour, what must remain blurred in the system becomes visible.
What is here called relationship is not a soft word. It describes a simple fact: a child learns differently when another person sees them clearly and thinks alongside them. The teacher sees what the child is doing, where they hesitate, what they do not understand, and what interests them. Her next question follows that perception. The child accompanied in this way learns differently from a child listening to a class. It is not only a different pace; it is a different kind of learning.
This insight has gained new weight through artificial intelligence. When machines can write texts, solve problems, and supply explanations, the question changes: what does education still have to do?33See Isabelle C. Hau, “Welcome to the Era of Relational Intelligence,” Stanford Social Innovation Review 24/2 (Spring 2026), DOI 10.48558/PY66-AH68. Hau develops the thesis in a broad social context. The pedagogical claim here is this text’s own inference from it: as machines take over tasks, the human capacities that require responsibility, relationship, and judgment become more valuable. One answer is this: machines can generate answers; they cannot answer for a child. They can recognise patterns, but they do not assume pedagogical responsibility for the child who is evading, giving up, or only pretending to understand. This is where relational intelligence begins: perceiving, responding, thinking with another person, without handing responsibility to a system.
The argument is not that the human, compared with the machine, retains some residual competence. It is this: the capacity to engage with other people so that thought, trust, and action can arise is becoming more important, not less. Whoever wishes to judge, collaborate, lead, teach, or care needs this capacity. It is not a consolation prize after automation. It is the main thing.
Historically, this is not a novelty. Education was relational long before it became administrative. The Greek paideia names a formation of the person that cannot be exhausted by curriculum.44The classical study is Werner Jaeger, Paideia: The Ideals of Greek Culture (3 vols., Oxford University Press, 1939–1944). For the pedagogical follow-through see also Hans-Georg Gadamer, “Education Is Self-Education,” Journal of Philosophy of Education 35/4 (2001), 529–538, based on an address delivered on 19 May 1999. Apprenticeship, study under a master, the scientific school: all bound learning to relationship. Mass schooling is a young form in this history, not the origin. The point is not a pedagogical fashion. It is something older that we have lost sight of for a while.
But if relationship is the medium of education, one has to ask what is awakened in it. The answer is not, first, information. It is curiosity. That is the subject of the next section.
The Developmental Ground
Children come into the world curious. This is not sentimentality but a well-supported developmental finding. Susan Engel has described how curiosity develops in childhood, how it becomes visible, and how little room it often gets in the ordinary business of school.55Susan Engel, The Hungry Mind: The Origins of Curiosity in Childhood (Harvard University Press, 2015), especially the chapters “Curiosity Goes to School” and “Cultivating Curiosity.” For the classroom observations cited here, see also Susan Engel, “Children’s Need to Know: Curiosity in Schools,” Harvard Educational Review 81/4 (2011), 625–645, DOI 10.17763/haer.81.4.h054131316473115. No one who has ever spoken with a four-year-old needs persuading that children live by questions. The more serious question is what happens to those questions once instruction becomes a system.
Engel’s important finding is not a romantic claim about children but a sober account of the environments they enter. In her classroom observations, visible episodes of curiosity were already rare in kindergarten and rarer still in fifth grade. The classroom is not a difficult place for curiosity because teachers forbid questions. It is difficult because children quickly learn that a question interrupts the movement of the lesson. Whoever asks holds up the class, risks saying something wrong, admits not knowing. There are reasonable grounds to stay quiet, and children learn them quickly.
The figure is familiar: a student who gives the right answer without being interested in the question. In academic high schools one meets such students too: young people who perform well and no longer know what they once wanted to know. They have not become stupid. They have become polite. Education as politeness.
The first task of a tutor is therefore not to begin by explaining. It is to protect the child’s curiosity. That sounds larger than it is. In practice it means: returning questions. Taking a question seriously, even when it sounds naive. Refusing to be the answer-machine. Pressing the child to say in their own words what they have been thinking before they receive an answer.
An example makes the point visible. A child says they don’t understand fractions. The easy answer would be to explain fractions again. The better beginning is the question: “Well — what do you think a fraction is?” The child is surprised that this question is being asked, and has to think for a moment. They say something — half right, half wrong. Here, in this half-right, half-wrong sentence, the lesson begins. Not at the curriculum, but at the point where the child actually stands.
It would be a mistake to treat curiosity as a method for making the child take in the curriculum more obediently. Curiosity is not a trick for smuggling the material in more effectively. It is the force by which material can become thought of one’s own. What a child takes from an hour is more likely to endure when it is tied to a question of their own. What is taken in without that tie often disappears after the test. Many adults know this from their own school years.
But curiosity must lead somewhere. It leads, when it does not run off into the arbitrary, to understanding. And understanding has a test that is hard to fake. That is the subject of the next section.
The Test of Understanding
Under Feynman’s name, a simple test of understanding has become famous: explain a thing so that a beginner can understand it.66The phrasing is widely attributed to Feynman; it does not appear in this exact form in his writings, but is a fair compression of what he set out at several places. See The Pleasure of Finding Things Out (Perseus, 1999), particularly the essays on teaching, and the prefaces to The Feynman Lectures on Physics (Addison-Wesley, 1963–1965). That sounds simple but is not banal. It is one of the most honest tests of learning, and it is often bypassed in school.
It is bypassed because school can test memorisation more easily than understanding. A whole class can be examined on the one more easily than on the other. The student who solves a quadratic equation by following a memorised procedure produces a correct answer. The student who does the same and can also explain in their own words why the procedure works has accomplished something else. The first has memorised. The second has understood. If both are rewarded equally, the child quickly learns that the one is enough.
This test is so effective because it forces memorisation to show itself. When one explains to a beginner what one has learned, all the places stand out where one has only the words and not the matter. One hesitates. One reaches for terms that themselves require explanation. One cannot answer a follow-up because one never had the answer oneself, only the right vocabulary. In this hesitation what one has not understood becomes visible — and here the next lesson begins.
The individual hour is the ideal site for this test, because the tutor can play the beginner without being the beginner. She knows what a quadratic equation is. But she can ask as if she did not, and the student must explain as if speaking of it for the first time. In this play — hardly possible in a whole class — the gaps come to light. They come to light without shame, because they appear at a point at which hesitation is in order. In explaining one may hesitate; in being examined one may not. That difference is the gift the tutor can place beside school.
And then there is a phenomenon that is hard to measure but known to every practitioner. When a child has truly understood something, their face changes. The eyes settle, sometimes the child laughs briefly, sometimes says “ah.” This response is not politeness, and it is not proof. It is a sign that something has rearranged itself. Joy, in this sober sense, is not the bonus of learning. It is one of its best marks. Whoever has seen this face once knows what a successful hour is; much of what grades measure looks coarse beside it.
What remains is the question: what does this look like in practice? What does the tutor do in a single hour with a single child? That is the subject of the last section.
The Practice
An hour. A child. A table. A piece of paper. That is all that is needed for something essential to happen.
An hour begins with the child arriving — usually with a task they have failed at or are avoiding. That task is the occasion; it is not the content of the hour. Before tutor and child bend to it, a question comes first: how the week was, what happened at school, what stayed with them. The question is not cosmetic. It is the first condition of the child thinking today rather than running through one more programme.
Then the work turns to the matter at hand. The matter is always concrete: a math problem, a German text, an English translation, a passage to memorise. But the tutor does not go through the matter by explaining while the child listens. The tutor asks, and the child explains. What do you think is being asked here? If you were explaining it to a third-grader, where would you start? Where, would you say, do you stop understanding it? These questions are the actual work. If the child can answer them, they have already half-understood the matter, and the other half often comes in the conversation. If they cannot, then the lesson begins — and begins not at the curriculum but where the child actually stands.
What remains at the end of a good hour is not a list of solved problems. It is a child who sees a problem differently today than an hour ago — and at best, the moment in which the child notices this themselves. Not every hour produces such a moment. Some are slow, resistant, apparently small. That is precisely why the work does not need a grand pedagogical gesture; it needs a discipline of precision: not helping too early, not helping too late, and not rescuing the child from their own thinking.
Parents want grades and progress. That is understandable, and the two often go together. But the real value of this work is first that moment, repeated over weeks until it becomes a habit.
To the parents reading this: what is offered is not homework help. Homework help is honourable work, but a different kind of work. What is offered is an hour in which your child uses the school’s material to learn what school cannot reliably teach: that learning is a process that belongs to oneself, and that a question, honestly asked, is the beginning of something rather than its end. Grades often follow; that is the unromantic truth. But they follow as a side-effect, not as a programme.
What this practice does not do: promise grade jumps. Deliver weekly progress reports. Take over the parenting. The child comes for an hour a week — not because everything happens in that hour, but so that what begins there can echo through the rest of the week.
Parents who want to test whether this form of tutoring fits should send a short note to team@jonashertner.com: the child’s age and school year, the subject or situation, and what currently feels stuck. Then we talk, the child comes by once, and together we decide whether the work makes sense.
On the history of Swiss public schooling, see Lucien Criblez (ed.), Bildungsraum Schweiz: Historische Entwicklung und aktuelle Herausforderungen (Haupt, 2008), especially Lucien Criblez / Christina Huber, “Der Bildungsartikel der Bundesverfassung von 1874 und die Diskussion über den eidgenössischen ‘Schulvogt’,” pp. 87–130. See also EDK, “150 Jahre Bildungsartikel in der Bundesverfassung” (29 May 2024), on compulsory primary education under Art. 27 of the 1874 Federal Constitution. For the broader institutional-historical context: Ivan Illich, Deschooling Society (Harper & Row, 1971), and David F. Labaree, Someone Has to Fail: The Zero-Sum Game of Public Schooling (Harvard University Press, 2010). ↩︎
Ivan Illich, Deschooling Society (Harper & Row, 1971), especially the critique of the institutionalisation of learning; David F. Labaree, Someone Has to Fail: The Zero-Sum Game of Public Schooling (Harvard University Press, 2010), on the logic of school sorting and credentialist expectation; Paul Goodman, Compulsory Mis-Education (Horizon Press, 1964), as an earlier American formulation of the same unease. Ken Robinson later popularised parts of this critique, most visibly in Out of Our Minds: The Power of Being Creative (Capstone, 3rd ed. 2017) and the TED talk Do schools kill creativity? (2006). ↩︎
See Isabelle C. Hau, “Welcome to the Era of Relational Intelligence,” Stanford Social Innovation Review 24/2 (Spring 2026), DOI 10.48558/PY66-AH68. Hau develops the thesis in a broad social context. The pedagogical claim here is this text’s own inference from it: as machines take over tasks, the human capacities that require responsibility, relationship, and judgment become more valuable. ↩︎
The classical study is Werner Jaeger, Paideia: The Ideals of Greek Culture (3 vols., Oxford University Press, 1939–1944). For the pedagogical follow-through see also Hans-Georg Gadamer, “Education Is Self-Education,” Journal of Philosophy of Education 35/4 (2001), 529–538, based on an address delivered on 19 May 1999. ↩︎
Susan Engel, The Hungry Mind: The Origins of Curiosity in Childhood (Harvard University Press, 2015), especially the chapters “Curiosity Goes to School” and “Cultivating Curiosity.” For the classroom observations cited here, see also Susan Engel, “Children’s Need to Know: Curiosity in Schools,” Harvard Educational Review 81/4 (2011), 625–645, DOI 10.17763/haer.81.4.h054131316473115. ↩︎
The phrasing is widely attributed to Feynman; it does not appear in this exact form in his writings, but is a fair compression of what he set out at several places. See The Pleasure of Finding Things Out (Perseus, 1999), particularly the essays on teaching, and the prefaces to The Feynman Lectures on Physics (Addison-Wesley, 1963–1965). ↩︎