Marc Armitage - Thought Crime

Loose Parts and Serendipity (2 of 2)

Loose Parts and Serendipity (2 of 2)

a \‘shȯrt-rēd’\ piece

The first of this two-part piece explored whether loose parts are ‘open ended’ or not (you will have to check that one to find the answer to that) and it ended by suggesting that loose parts work best when we leave them somewhere and then walk away. Part two explores exactly where that ‘somewhere’ should be.

It’s largely about the word serendipity and a long dead German biologist.


The dead German was Jakob von Uexküll (1864-1944) a biologist who was particularly interested in how living organisms perceive the physical environment in which they find themselves. His conclusion was broadly that all organisms strive to become familiar with their surroundings by actively exploring their environment and making physical contact with it.

When they discover something that they ‘don’t know’ they apply what they ‘do know’ in order to conclude what the form and function of this new feature is. Gradually, the organism begins to develop a detailed perception of their environment and so its ability to interact with it increases. He called this developing known-environment the Unwelt.

It’s more complex than that, obviously, but the key takeaway parts are active exploration, physical contact, and the application of knowns to unknowns. You might already be able to see the connection between Uexküll’s ideas and play in general but maybe not in relation to loose parts yet.

The general connection is that his concept adds some explanation to the way that children interact with the environment they find themselves in, being that they are also apparently living organisms. It explains, for example, why children seen to be drawn to physically touching anything they come into contact with; and also why children can reach different perceptions from each other on the same finding: it’s to do with their respective stock of previous knowns.

There are other less obvious connections too

Despite the fact that you may think you have never heard of Uexküll I bet you twenty-pence you have been influenced by him as his notion can be found at the heart of the work of other names you may be more familiar with including the perceptual ideas of the psychoanalyst Karl Jung, the cognitive theories of psychologist Jean Piaget, and the concept of affordance from James J Gibson.

Most significantly, Uexküll’s work can be seen in the ideas of the mathematician and computer pioneer Seymour Papert. Why is that important? Because Papert is as significant in the development of loose parts theory as is the better-known Simon Nicholson. That’s why.

Which brings us to the serendipity bit

Serendipity (the occurrence and development of events by chance in a happy or beneficial way) is not a word you hear often in the teaching world. In fact, it can often be seen as the complete opposite of teaching. After all, to ‘teach’ implies planning and direction which means that we know where things are heading once we have started them off and that thing is usually some form of learning outcome.

To imply, therefore, that learning might take off in some unknown, unplannable direction is something that just does not make itself felt in the training of many educators and so there is a temptation to actively guard against serendipity kicking in. For some, it can be seen as a distraction from the carefully planned learning experience we are in the process of delivering but for others the deviation from planned learning can be absolutely terrifying.

Papert, on the other hand, was a pioneer in the ‘discovery method’ of learning (as was Simon Nicholson) and was a firm proponent in the idea that learning could, and should, be largely self-directed. This was centred in his belief that, “You can't teach people everything they need to know. The best you can do is position them where they can find what they need to know when they need to know it.”

That’s serendipity, see?

The trouble is

Although this idea has made itself felt in the training of many educators it is not the case in all. Modern teaching methods and the training that promotes it (especially in the United States) are still often a highly structured affair and the delivery of learning is often seen as a time-restricted activity (which is the real enemy of play and self-discovery).

The result can be that despite a recent explosion in the interest of loose parts in a learning context it leads some to consider them as just a substitute for the more typical materials we find in education catalogues. That leads to thinking of them as requiring the same degree of structure or at the very least as needing a ‘starting point’ for effective learning to occur.

So, we set them up as ‘provocations’ or invitations to play on a tabletop or in a ‘station’. This is completely at odds with the ideas of Nicholson and Papert, even if we see the setup as only a begining'.  


This is because at the root of the concept of loose parts is not only that children should have access to as broad a variety of materials as possible but that they should be able to choose how those materials can be combined.

So, whether we acknowledge it or not, when we adults gather specific materials together and place them in a context that we have pre-decided, like the tabletop, then we have both restricted the variety of available materials and any possible combinations available. Worse, we have removed the element of serendipity – that chance discovery of a previously ‘unknown’ combination.

The solution is to go back to the ideas of a long dead German and embrace the idea that we don’t always know what the ‘knowns’ are for the playing children that we work with. Even they don't know. New understanding comes down to them making physical contact with the environment around them, applying their existing knowns to newly found unknowns, and creating their own connections.  

The conclusion

As said in part one, we should just leave loose parts somewhere and walk away. But leave them where?

The answer is not to set them out in a corner or on a table in a pre-conceived way. Instead they support children best when hidden or apparently dumped randomly in places where children have to actively seek for them or come across them by accident. This gives them the opportunity to make their own discoveries, come across serendipitous combinations and use them in a way that they choose to so that they can learn what they need to learn when they need to learn it.  

It’s not neat. It’s not tidy. Yet it is incredibly powerful.

Marc Armitage


Photo a bunch of loose parts tucked around the corner, in Meriden Adventure Playground UK taken by Marc

For more information and to follow Marc on other platforms head over to