Play is an essential part of learning. The young of many species play in order to test their physical limits, form bonds with others, explore the environment, practice hunting behaviours and generally mimic their elders. Human children are no different in this respect: we play in order to learn about ourselves and our environment. It’s probably true to say that the instinct to play is ‘hardwired’ into us and, short of locking children in a box, there’s no way to prevent them from playing. Social learning is the basis for the transmission of human culture and play is an unavoidable component of social learning.

There’s an irony that things human being find easy to learn are fiendishly difficult to teach machines. This is Moravec’s paradox: contrary to the initial expectations of artificial intelligence researchers, robots and computers are excellent at learning how to do many of the things we’re bad at, but – thus far – have struggled to learn those things we find easy. The instinct to play is a fantastically useful adaptation which accelerates our ability to learn the knowledge we need to survive and thrive in a hostile environment.

But, whilst play is great way to learn some things, it’s an ineffective way to learn others. The basic rule of thumb is that if a behaviour or skill is a universal constant possessed, as far as we can be sure, by all human cultures since homo-sapiens first emerged, then it will probably develop in children through immersion, emulation and modest amounts of coaching and modelling. But, if a way of acting or understanding the world is culturally specific – i.e. it has emerged since the dawn of recorded history – then it probably requires an element of formal instruction in order to learn. Most people will pick up how to speak, manipulate simple tools and predict the behaviour of animals and other humans without much direction; play may well be enough.

However, most people will need a fair bit of guidance and support to learn to read, solve algebraic equations, and operate complex tools. The more complex and culturally specific something is, the greater the requirement for formal instruction. No one is going to work out to physics formulas or how to make a Gâteau St-Honoré without a lot of fairly formal instruction from a relative expert.

This is why schools have emerged. As Steven Pinker says in The Blank Slate, “Education is a technology that tries to make up for what the human mind is innately bad at.” Children don’t need to go to school to learn how to walk, talk, recognise objects or remember the personalities of their friends, even though these things are much harder than reading, adding or remembering dates in history. The culturally specific knowledge taught in schools is unlikely to be acquired without formal instruction. Schools – institutions for providing formal instruction – have only existed in literate cultures where cultural knowledge has out- stripped universal folk knowledge. If we look at ancient Sumer where cuneiform, the earliest known writing system developed, we also find the first archaeological evidence of schools, which were set up to induct budding clerks into the mystery of scribing. These edubas (scribe schools) were probably fairly informal, usually taking place in private homes, but some sites, where particularly large numbers of school tablets have been unearthed, are considered by archaeologists to be ‘school houses’.

Where schools are not imposed, they emerge. In the poorest slums and most remote villages of India, Nigeria, Ghana, Kenya and China, low-cost private schools have sprouted up spontaneously. In The Beautiful Tree, James Tooley first recorded this phenomenon in the Indian city of Hyderabad which contains over 500 private schools catering to the children of day labourers and rickshaw pullers. People everywhere seem to recognise that where these sorts of institutions don’t exist (or where they are so corrupt they may as well not) adults can gather children in rooms and teach them culturally accumulated wisdom. Obviously, these schools are unlikely to conform to our expectations of what schools in developed nations should look like, but the way in which they operate is essentially the same as the Sumerian eduba. The school – with classrooms, teachers, and formal instruction – has emerged as the simplest, most effective way to handle the business of education.

So what about those who want instruction to be more ‘play based’ and seek to make the environment in schools more similar to the ‘natural’ environments’ where children learn the basics? Schools may need to make sure that children’s environments are conducive to acquiring the folk knowledge we all take for granted. Just because we have an evolved predisposition to attend to and rapidly learn this stuff, it doesn’t follow that we will automatically do so. Luckily, we’re highly motivated to learn these things and, just so long as we encounter them in our environment, we almost certainly will. This might provide an argument in favour of coaching and modelling approaches in the early years of education to ensure all children are immersed in the kind of environment in which they pick up speech, group cooperation and a sense of self. But if we’re tempted to teach these kinds of things explicitly later on, then we will almost certainly be wasting our time and that of our students.

Beyond that, one authority advocates of play-based instruction cite in support of their preference is the late Dr Karyn Purvis.

Her claim that “Scientists have recently discovered that it takes approximately 400 repetitions to create a new synapse in the brain – unless it is done with play, in which case, it takes between 10-20 repetitions” is extraordinary. If true this provides much needed evidence to support the whole-scale introduction of play-based instruction into school systems. Unfortunately, it doesn’t seem to be true. Jen Taylor, an advocate for play-based instruction tried to trace the source of the quote and wrote about her efforts here. When approached for clarification, the Karen Purvis Institute of Child Development offered this response:

This was indeed said by Dr. Purvis. Sadly, before she gave us the reference she became terribly ill and passed away. We have not been able to locate this, so the source went to the grave with Dr. Purvis.

So who are the ‘scientists’ who made this incredible discovery? No one seems to know. We’re asked to accept that they chose only to tell Dr Purvis about their findings before her untimely death and then, for some reason, decided not to publish their work. If any neuroscientists out there are looking for a claim to test, it should be easy enough to design an experiment that establishes beyond doubt whether play is an up-to-40-times quicker way to acquire knowledge.

As outlined above, there are excellent reasons for acknowledging the crucial role of play in our development and reasonable evidence to allocate a part of the early years of schooling to structured play. Claims beyond that are not well-supported and benefits no one to repeat spurious ‘evidence’ in our pursuit of ideology. We all want the best for children, but good intentions are not enough. As the economist and social theorist, Thomas Sowell has said, “It is so easy to be wrong – and to persist in being wrong – when the costs of being wrong are paid by others.”

This post is adapted from Chapter 2 ‘Built by Culture’ of Making Kids Cleverer.