How Scandinavia built a childhood around controlled danger — and where that same region is quietly failing at the exact same lesson.
In 1943, in the middle of the German occupation of Denmark, a landscape architect named Carl Theodor Sørensen opened a playground in Emdrup, a housing estate on the edge of Copenhagen, that had almost nothing in it. No slides. No swings. No brightly painted climbing frames. Instead there was a fenced plot of dirt, bricks, scrap timber, old tools, and a supervisor named John Bertelsen, who mostly stayed out of the way. Sørensen had noticed something during his years designing conventional playgrounds: children ignored them. They preferred building sites, ditches, and rubbish heaps — anywhere they could shape the environment rather than just move through it. His answer was the skrammellegeplads, the junk playground, built on the idea that children develop competence by handling real materials and real consequences, not by being protected from both.
That single Copenhagen experiment is usually cited as the birth of what play researchers now call risky play, and it set the terms for a debate that Scandinavia has been running with itself, in public, for eighty years. On one side sits a body of child-development theory and everyday practice that treats supervised exposure to danger — heights, speed, tools, fire, traffic, the chance of getting lost — as something children need, not something to be engineered out of existence. On the other side sits a newer and less flattering pattern: evidence that the same Nordic countries known for handing five-year-olds saws are also raising a generation more anxious, more supervised indoors, and less independent than their reputation suggests. Both things are true at once, and reckoning with the region honestly means holding them side by side rather than picking the version that fits the postcard.
The model: risk as a teaching tool
The theoretical backbone of Scandinavian risky play comes from Ellen Beate Hansen Sandseter, a professor at Queen Maud University College in Trondheim, whose doctoral research involved watching Norwegian preschoolers at play and asking both the children and their teachers what felt dangerous. Her 2007 study sorted what she observed into six categories, which have since become the standard taxonomy in the field.
| Category (Sandseter, 2007) | What it looks like | Danish/Norwegian institutional example |
|---|---|---|
| Great heights | Climbing trees, high platforms | Tree climbing and tower structures in staffed nature playgrounds such as Valbyparken in Copenhagen |
| High speed | Running downhill, sledding, cycling fast | Steep slides and sledding hills; unsupervised cycling from an early age |
| Dangerous tools | Knives, saws, axes | Danish daycare “knife diplomas”; forest kindergartens where children whittle and split wood |
| Dangerous elements | Fire, deep water, cliffs | Supervised campfires at outdoor kindergartens (friluftsbarnehage) |
| Rough-and-tumble | Play fighting, wrestling | Generally tolerated rather than broken up by staff |
| Disappearing / getting lost | Hiding, exploring alone in tall grass or forest | Children given wide, often unfenced natural terrain to roam within |
Sandseter and her colleague Leif Kennair later argued, in a paper on the evolutionary function of risky play, that this kind of play works as a kind of inoculation. Children who are exposed, in manageable doses, to heights, speed, and separation from adults appear to build tolerance for the fear those situations naturally produce, rather than letting that fear calcify into anxiety. A more recent comparative study from Sandseter’s group found that Norwegian children showed significantly higher risk willingness than Canadian children in a playground task — and, notably, that this greater willingness to take physical risks did not translate into recklessness in traffic. The same children who climbed higher also crossed streets more carefully. The finding cuts against the instinctive assumption that a child comfortable with risk is a child in more danger; the data suggests risk-tolerant children are often the more competent risk assessors, not the less careful ones.
Denmark carries this furthest into policy and daily habit. Reporting from Quartz on Danish daycare practice describes staff handing out knife diplomas once children demonstrate they can use a blade safely, and forest playgrounds on Copenhagen’s outskirts where building fires is a normal afternoon activity. Play researcher Heidi Vikkelsø Nielsen, quoted in that reporting, chose a photograph of a small girl wielding a full-size saw as the image that best captured Danish childhood. Tim Gill, a British play advocate, has described Danish playground design as deliberately including “water, fallen trees, hillocks, ditches and other risky elements” as standard ingredients rather than hazards to be designed away. A comparative study of Danish and American mothers, shown identical photographs of children sledding, sawing wood, and climbing trees, found Danish mothers consistently more comfortable with their own children in those same situations.
Copenhagen’s traffic playgrounds — miniature towns with working stoplights, kid-sized bike lanes, and scaled-down street furniture — extend the same logic to the road. Rather than keeping children off streets until an arbitrary age, the city gives them a consequence-free replica of the street grid to make mistakes in first. Norway builds a parallel structure into early childhood education itself: more than 400 naturbarnehager, or nature kindergartens, now operate across the country, a number that has grown sharply since the 2000s, built on the concept of friluftsliv — open-air life — as something to be practiced from infancy rather than discovered later. One account from an early-childhood researcher who grew up in a Norwegian city and now studies risky play notes that Norwegian children are typically given the freedom to walk to school or play unsupervised by ages five to eight — compared with eleven to thirteen in Greece, and an age many American nine- to eleven-year-olds never reach at all, according to the same cross-national mobility comparison. She points to one structural factor rarely mentioned in the romanticized version of this story: Norway’s universal health coverage means a broken arm is a medical event, not a financial one, which removes a great deal of the liability anxiety that shapes playground design in places like the United States.
None of this is presented, in Scandinavian pedagogical literature, as recklessness. Sørensen’s own description of playgrounds as “the city’s most important form of public plantation” reflects an argument that risk-managed environments are civic infrastructure, on par with schools or clinics, not an indulgence. The premise running through all of it is straightforward: children who are never allowed to encounter danger under supervision do not become safer. They become less able to recognize danger the one time it matters, unsupervised.
The counter-model: the same countries, overprotected in different directions
The trouble with treating Scandinavia as a single coherent philosophy is that Scandinavians themselves have spent two decades arguing that their own parenting culture drifted the other way. In January 2004, journalist Ivar Ekman published a series of articles in the Swedish daily Svenska Dagbladet profiling Danish psychologist Bent Hougaard, who had coined a term for parents who clear every obstacle from a child’s path: curling forældre, curling parents, after the sport in which players sweep the ice smooth ahead of a sliding stone. Hougaard’s description of the resulting childhood, quoted in Ekman’s later account for Psychology Today, was blunt: children raised this way live “like the yolk in an egg, completely protected from all evil.”
Ekman has said the response “ruined dinner parties” for months, because everyone recognized the pattern in themselves or their friends, and a decade later Merriam-Webster was citing the same reporting to explain the term to English speakers, noting that the idea found real traction because so many Swedes saw their own habits reflected back at them.
That the term originated in Scandinavia at all is the tell. Overprotective parenting is not an import that arrived alongside American helicopter parenting; Danes and Swedes built their own vocabulary for it because they recognized it in their own households, running in parallel with the forest kindergartens and knife diplomas the same countries export as a national brand. Even the risky-play infrastructure has drifted toward caution over time. Norway’s own outdoor-education researchers acknowledge, in the same account cited above, that increasing media coverage of childhood injuries and stricter playground safety regulation have produced “a slow drift toward caution” even in a country built around friluftsliv — outdoor play is described as being in decline there too, just from a much higher starting point than most of the world. European playground certification under EN 1176, the continent-wide safety standard, now governs equipment height, fall zones, and surfacing with enough specificity that manufacturers market compliance as a selling point in itself — a bureaucratic layer the original junk-playground designers would have found bewildering, even if the standard technically exempts staffed adventure playgrounds from its full scope.
The clearest failure of self-protection in Scandinavian childhood, though, is not a drift toward caution at all. It is the opposite: a domain where almost no supervised on-ramp was ever built in the first place. Every institution described above exists to give children a scaled-down, watched version of a real danger before they meet the full-size version alone — the traffic playground before the actual street, the knife diploma before the unsupervised blade, the staffed campfire before the one nobody is tending. Nothing plays that role for the phone. There is no Danish equivalent of a trafiklegeplads that teaches a nine-year-old to read a recommendation algorithm the way a mini traffic town teaches her to read a crosswalk; no staged introduction to a feed engineered to hold attention, the way a climbing frame is staged by height for a five-year-old. Children who spent their toddler years handling a saw under a watchful adult were, a few years later, handed a smartphone with none of that scaffolding, at precisely the age Sandseter’s own research treats risk assessment as still under construction. The method that works reliably offline — controlled exposure, an adult nearby, consequences kept survivable — was never adapted for the version of danger that actually spends the most hours with children today.
Scandinavia’s exceptionalism disappears entirely here. Social psychologist Jonathan Haidt’s comparative research on Nordic adolescent mental health found that self-reported anxiety and depression among teenage girls rose sharply across all five Nordic countries starting in the early 2010s, alongside a 144.8 percent increase in inpatient self-harm diagnoses among ten- to fourteen-year-old girls between 2010 and 2021. Haidt’s explanation — that smartphones and social media colonized adolescent social life between 2010 and 2015, faster in the Nordics and the Anglosphere than almost anywhere else — has drawn substantive pushback from researchers who point out that the same international survey data he cites shows psychological distress climbing more gradually since the 1980s, well before smartphones existed, and that broader cross-national datasets show far smaller and less consistent effects than his headline graphs suggest. Whichever side of that argument holds up better, the underlying fact both camps accept is not in dispute: Nordic teenagers, by self-report, are considerably more anxious than they were fifteen years ago, and it happened at the same moment their physical world was, if anything, getting freer. A country that built a whole pedagogy around handing a six-year-old a saw has built nothing resembling it for what a thirteen-year-old does with a phone — no supervised version, no scaled-down introduction, just an unguarded handoff at the one age the rest of the system was designed never to leave unguarded.
Which model is actually correct
Framed as a single yes-or-no question — is the risky-play model right, or is the overprotective one — the honest answer is that the question is stacking two different phenomena on top of each other and asking them to cancel out. They don’t measure the same thing, and Scandinavia doesn’t run one experiment. It runs two, in different domains, with opposite results.
On physical risk, the model holds up well and the evidence is not close to ambiguous. Sandseter’s category system is now the standard reference point in play research internationally, not just in Norway, and the studies built on it — including her own finding that risk-willing children showed more efficient risk assessment in traffic, not less — point in one direction. A comparative study of London and American playgrounds, cited by the Danish playground manufacturer Monstrum, found that playgrounds designed for risky play attracted more visitors, more physical activity, and a wider age range of users than conventional litigation-averse equipment, and other research in the same literature has linked risk-taking in free play to fewer injuries later in organized sport. Denmark did not stumble into knife diplomas and traffic playgrounds as a quirk; the practice has an evidence base, and Denmark, Norway, and Sweden have simply been more willing than most countries to build institutions around it rather than treat it as an eccentric parenting choice.
On everything downstream of a screen, the picture reverses, and Scandinavia’s own vocabulary — curling parents, coined at home, not borrowed — already conceded the point twenty years before Haidt gave it an international audience. The paradox is not that Scandinavia got risk wrong. It’s that a region built genuinely serious infrastructure for managing physical risk in childhood, and then let a form of risk it hadn’t built any infrastructure for walk straight past all of it. A child who has learned to assess a fall from a tree has not, by that fact, learned anything about assessing an algorithm designed to hold her attention. Those are different competencies, taught by different environments, and Scandinavia only invested seriously in one of them.
The more useful reading, then, is not that one Scandinavian model is correct and the other is a national failure of nerve. It’s that risk tolerance and risk literacy don’t transfer automatically between domains, physical and digital, no matter how thoroughly a culture masters one of them. Denmark’s playgrounds are a genuine achievement worth exporting, and have been, from Lady Allen’s post-war adventure playgrounds in London to nature-play installations now built in Illinois. But the same countries that solved the broken-arm problem are currently living through the consequences of not having solved the broken-attention problem, and no amount of forest kindergarten changes that.
Sources
- Wikipedia, “Emdrup Junk Playground“
- The Copenhagen Post, “The junk playground – Denmark’s eco-contribution to outdoor school education,” 2017
- Helle Nebelong, “Nature playground in Valbyparken“
- Danish Architecture Center, “Danish Playgrounds Move Into Children’s Imaginations Around the World“
- Sandseter, E.B.H., “Categorising risky play — how can we identify risk-taking in children’s play?,” European Early Childhood Education Research Journal, 2007
- Sandseter, E.B.H. and Kennair, L.E.O., “Children’s Risky Play from an Evolutionary Perspective,” 2011
- Ellen Beate Hansen Sandseter, research site and recent risk-willingness study
- Quartz, “Why the Danes encourage their kids to swing axes, play with fire, and ride bikes in traffic,” 2018
- The Conversation, “Denmark’s ‘hands-off’ approach to parenting“
- The Local Norway, “Everything you need to know about Norway’s ‘outdoor’ kindergartens“
- After Babel, “A Norwegian Approach to Supporting Children’s Risky Play“
- Merriam-Webster, “Snowplows, Dolphins, and Other Words for Parenting Types“
- Psychology Today, “A Second Take on Swedish Parents“
- en-standard.eu, “EN 1176 / EN 1177 playground standards“
- After Babel, “The Teen Mental Illness Epidemic is International, Part 2: The Nordic Nations“
- LSE Parenting for a Digital Future, “Do smartphones really cause mental illness among adolescents? Ten problems with Jonathan Haidt’s book,” 2024
- Monstrum, “The Extinction of Risky Play?“
English