Unschooling and self-directed education

· 12 min read

Holt's disillusionment arc

John Holt began as an educational reformer, working in private schools in the 1950s and 1960s and writing books that were initially understood as advice for teachers. How Children Fail (1964) is a journal of his classroom observations, recording in detail the moves children make to manage fear and please adults — guessing strategies, mumbling answers, going blank — and the way these moves substitute for thinking. The book was widely read by teachers. Holt's slow conclusion, over the next decade, was that these moves were not symptoms of bad teaching but of the structural situation of being a student. He stopped recommending school reform and started recommending school exit. By Instead of Education (1976) he was using the word "deschooling," borrowed from Ivan Illich. The arc matters because it documents a thoughtful person changing his mind against his own professional interest, which is rare in any field.

How children learn when nobody is teaching them

Holt's How Children Learn (1967) is the constructive companion to How Children Fail. It records his observations of small children — his nieces and nephews and the children of friends — figuring out how to talk, count, draw, and navigate a household. The book's force is cumulative. By the fifth or sixth example, the reader has noticed that children pursue understanding with relentless precision when nobody is grading them, and stop pursuing it almost immediately when grading begins. Holt did not invent this observation; Piaget and others were making related claims. But Holt's prose made the observation usable for parents. The book remains the single best introductory text for anyone wondering whether children would learn without being made to.

Gray's evolutionary frame

Peter Gray's contribution was to embed Holt's observation in evolutionary psychology. The hunter-gatherer societies that anthropologists have studied — !Kung, Hadza, Aka, Agta — universally allow children long unsupervised play in mixed-age groups, with minimal direct instruction. Children acquire the full adult skill set this way, including hunting, tool-making, plant identification, and social negotiation. Gray's argument is not that we should return to hunter-gatherer life, which is impossible and undesirable, but that the cognitive architecture children bring to learning was shaped in that context and still expects it. Schooling is a 200-year-old patch on a 200,000-year-old system. The patch works imperfectly, and the system, given any chance, reasserts itself.

The Sudbury model

Sudbury Valley School, founded in Framingham, Massachusetts in 1968, is the longest-running institutional experiment in self-directed education. Children ages four to nineteen mix freely, choose their own activities, govern the school through a weekly meeting where every student and staff member has one vote, and graduate when they convince their peers they are ready to take on adult life. There is no curriculum and no required instruction. Sudbury graduates, tracked in follow-up studies by Daniel Greenberg and others, go to college at roughly the rate of the surrounding population and report unusually high career satisfaction. The school is small and self-selected, so the data should be read carefully, but the existence of fifty years of continuous operation refutes the strongest version of the claim that this approach cannot work.

Radical versus pragmatic

Within the unschooling community there is a long-running argument between radical unschoolers, who extend the principle to all domains of family life (no bedtimes, no screen limits, no required chores), and pragmatic unschoolers, who keep some household structure while removing the academic structure. The radical position is internally coherent — if children are competent self-regulators in learning, why not in sleep and food? — but the pragmatic position has more durable households. The collective lesson from twenty years of community discussion is that the absence of academic coercion is the load-bearing piece. The other rules can stay or go without changing the educational outcome. Families burn out faster trying to abolish everything at once.

The reading question

The single most asked question of unschooling parents is: when will the child read? The answer is: between four and ten, with most clustering at six or seven, and a handful as late as nine or ten with no long-term consequence. The late readers catch up within a year. This claim, which sounds extraordinary, has empirical support — see studies by Peter Gray and Gina Riley, and the older Plowden Report data from Britain — but it is hard for parents to believe in advance. The child who is not reading at seven is the test case. The parents who hold their nerve discover the model works. The parents who panic and import a phonics curriculum produce a child who reads at seven, which would have happened anyway.

The Taylor critique

Astra Taylor, the filmmaker and writer, was unschooled in Georgia in the 1980s and 1990s and has written ambivalently about it. Her position is not a repudiation — she credits unschooling with her intellectual independence and the absence of school-trained credential-seeking — but she names the costs. Isolation. The pressure of being a one-family experiment. The way unschooling without a community becomes a parent-imposed identity the child cannot opt out of. Her critique is more useful than the standard outside critique because it comes from inside and identifies specific failure modes that the movement can address: more community density, more humility from parents, more honest accounting of which children the model serves and which it doesn't.

What an unschooling day actually looks like

The honest description is uninteresting from the outside. The child wakes when they wake. They eat. They do something — read, build, draw, watch a video, talk to a parent, play with a sibling, go to a co-op, take a lesson they chose. They eat again. They do something else. Periodically they go through phases where they appear to do nothing of value for days or weeks, and periodically they go through phases where they consume an entire field — astronomy, knitting, the French Revolution — at an alarming rate. The variance is the model. The flat middle that schools produce, in which every child does the same thing for the same hours, is precisely what unschoolers are trading away. Some weeks the trade looks brilliant. Some weeks it looks like failure. Over years it usually looks fine.

The parent's actual job

Unschooling does not mean parents do nothing. It means parents do different things. They keep the house stocked with books, materials, tools. They notice what the child is interested in this week and find one more thing in that direction. They drive a lot — to co-ops, lessons, libraries, friends. They have long conversations, especially at meals. They model adult work the child can see and ask about. They restrain the impulse to teach when the child has not asked. This restraint is the hard part. A parent who has internalized school logic has to actively un-learn the reflex to turn every observation into a lesson. Unschooling parents who do this well report it took them a year or two to fully relax.

The cost question

Unschooling is sometimes presented as cheap because there is no tuition. The reality is that it usually requires one parent at home or working flexibly, which is an income choice with real costs. It also requires the family to be located somewhere with library access, community resources, and other unschooling families, which constrains where one can live. The cash outlay can be small — co-op fees, occasional classes, materials — but the opportunity cost in foregone income is large. Honest accounting of unschooling has to include this. The model is more accessible to families with one high-earning parent and one flexible parent than to single parents or two-earner households at lower incomes.

Self-direction without ideology

A useful distinction inside the movement is between self-directed education as an empirical claim — children can do this, here is the evidence — and self-directed education as an ideological position about the illegitimacy of all schooling. The empirical claim is defensible. The ideological one is overreach. Some children, in some families, do well in some schools, and pretending otherwise alienates allies and embarrasses the movement. The strongest version of the unschooling argument is the modest one: this works for the families who choose it, the children who emerge are not damaged and are often distinctive, and the broader culture would benefit from taking the model seriously as one option among several rather than as a fringe lifestyle.

What happens after

The follow-up question is what unschooled children do as adults. The data is small but consistent. They go to college at slightly lower rates than their peers, often choosing community college or unconventional paths. They cluster in self-employment, the arts, technology, and trades. They report higher career satisfaction and stronger identification with their work. They tend to keep learning new fields throughout adulthood, which is the trait the original unschooling promise centered on. They sometimes struggle with bureaucratic environments, which is unsurprising. None of this is destiny — it is descriptive of a small self-selected sample — but it is enough to refute the strongest version of the worry that unschooled children will be unable to function in adult life.

The household-scale takeaway

For families not ready to unschool fully, the relevant move is to import unschooling principles into the hours the school does not occupy. Protect long unstructured blocks. Resist the reflex to fill every afternoon with lessons. Let boredom run its course; it is usually the precursor to a self-directed project. Keep the household stocked with materials the child can use without permission. Have real conversations about real things. Model adult work that the child can see and ask about. These moves cost nothing, do not require leaving the school system, and reproduce most of the unschooling effect at lower stakes. The collective version is what a culture would look like if a critical mass of families practiced this on weekends and summers. It would look, gradually, less like school and more like the kind of childhood Holt thought was possible.

Citations

1. Holt, John. How Children Fail. Rev. ed. New York: Da Capo Press, 1995. 2. Holt, John. How Children Learn. Rev. ed. Cambridge, MA: Da Capo Press, 1995. 3. Holt, John. Teach Your Own: The John Holt Book of Homeschooling. Rev. ed., with Patrick Farenga. Cambridge, MA: Da Capo Press, 2003. 4. Holt, John. Instead of Education: Ways to Help People Do Things Better. Boulder, CO: Sentient Publications, 2004. 5. Gray, Peter. Free to Learn: Why Unleashing the Instinct to Play Will Make Our Children Happier, More Self-Reliant, and Better Students for Life. New York: Basic Books, 2013. 6. Gray, Peter, and Gina Riley. "The Challenges and Benefits of Unschooling, According to 232 Families Who Have Chosen That Route." Journal of Unschooling and Alternative Learning 7, no. 14 (2013): 1–27. 7. Taylor, Astra. "Unschooling." n+1 no. 13 (Winter 2012). 8. Greenberg, Daniel. Free at Last: The Sudbury Valley School. Framingham, MA: Sudbury Valley School Press, 1995. 9. Illich, Ivan. Deschooling Society. New York: Harper & Row, 1971. 10. Gatto, John Taylor. Dumbing Us Down: The Hidden Curriculum of Compulsory Schooling. Gabriola Island, BC: New Society Publishers, 1992. 11. Llewellyn, Grace. The Teenage Liberation Handbook: How to Quit School and Get a Real Life and Education. Eugene, OR: Lowry House, 1991. 12. Gaither, Milton. Homeschool: An American History. 2nd ed. New York: Palgrave Macmillan, 2017.

◆

Cite this:

View edit history

← PreviousTracking apps and the self-as-data Continue →The homeschool movement — left, right, and otherwise

Comments

Be the first to share how this landed.