This week, Owen Barder gave an excellent presentation on Complexity and Development and asked whether development is an “emergent property of a complex adaptive system.” After listening to his talk, I fully agree with this definition. On further reflection, however, I decided that development is more like toast. Yes, white, wheat or rye, crisped up with heat. Don’t you agree? Let me explain.

During his talk, Owen referred to Thomas Thwaites’ book “The Toaster Project” which describes his efforts to make his own toaster. In particular, Thwaites tried to recreate this …

By collecting and smelting his own iron ore, obtaining a gallon of petroleum and producing his own plastic, etc. and then assembling the whole thing.

He ended up with this

Which was marginally successful. It worked for about 12 seconds before the circuits shorted and failed.

The point that Owen took from this story is quite true: even a simple object produced in a modern society relies on the existence of multiple supply chains, technologies, products. These do not come into existence in isolation. They are all related and co-evolve in patterns that are best modeled as complex adaptive systems. No preplanned “Big Push” or “structural adjustment” is going to generate the inter-related markets, institutions, organizations, behaviors, and norms that make up a functional whole in a ‘developed’ society.

But I think the value of the toaster story in relation to development goes further than this. Thwaites’ approach is a great metaphor for the way typical development programs are designed. In particular, why did Thwaites start by trying to reproduce a toaster that he purchased in a London appliance store? Instead of imitating someone else’s solution (a graphic demonstration of “isomorphic mimicry”), he could have asked how to make toast with materials on hand. Rather than taking the end product of a complex process as a model and trying to reproduce its form, he could have started with a simple design and tried to experiment and adapt it (see problem-driven iterative adaptation),

Thwaites wasn’t trying to produce a toaster. He was really exploring questions of design and technology in our era. But imagine if his goal really had been to make toast. If he had been camping in the woods, he might have simply jammed a stick into a piece of bread and held it over a small fire. Over time, he might have taken a wire hanger and fashioned it into a more sophisticated bread holder. Or if, like me, he had visited a house on an island off the coast of Maine with a propane campstove, he might have found something like this in the cupboard.

I first encountered this “toaster” during a vacation in 1979. Not only did it make delicious toast, but by my best estimate it had lasted about 30 years with no maintenance. When I visited last fall – more than 30 years later – it was still working. In that context. For that purpose.

That’s why I think development is toast.



CGD blog posts reflect the views of the authors drawing on prior research and experience in their areas of expertise. CGD does not take institutional positions.


Views from the Center