I went off to New York last week at the invitation of the UNDP Regional Center in Europe and Central Asia to discuss using complexity thinking to design a new ‘Finch Fund’ to support innovation and scaling up. Most scale-up exercises take successful pilots and just try and replicate them (one of the UNDP organizers, Millie Begovic, memorably likened it to trying to turn a particularly cute baby into a really enormous baby, rather than an adult). The Finch Fund is looking for a better way, and thinks complexity and systems thinking could be the answer, so it decided to try a bit of ‘interdisciplinarity’. I spent the day with a handpicked group of psychologists, ecologists, philosophers and economists all working on complex adaptive systems. It was horrible, but productive (I think).
Leaving your discipline is traumatic – akin to being suddenly infantilised and made vulnerable. From being a bit of a development know-all, I became a know-nothing, resentfully observing an incomprehensible exchange of references, books, jargon and gossip (oh, Bob’s at Yale now, is he?). I got angry, sullen, even contemplated walking out in protest.
But I didn’t flounce out, and instead (perhaps inspired by Robert Chambers‘ call for greater reflection in our work) started musing about what was going on in my head.
Lesson 1: ‘Interdisciplinarity’ is tough: be prepared for some serious mood swings.
Lesson 2: Beware geeks bearing gifts. What kinds of people do you want in the room? I would say the ability to empathise is probably more essential than being top of the field. Choose on the basis of emotional intelligence as well as the more academic kind. Some of this lot failed pretty heavily (maybe they were exacting revenge for being bullied at school) – if you ever see me give a furtive little smile of self satisfaction when someone says they can’t follow what I’m saying, please give me a slap.
Lesson 3: I found myself veering between ‘oh wow, this is a whole new way of seeing the world’ and ‘this is just describing what we do already in a new vocabulary’. Need to resist both temptations, and try and sift through the murk for the occasional nugget.
But how you do that depends on whether you are trying to immerse yourself in a whole new paradigm (painful, you have to leave aside your reference points, with no idea of where you are going to end up, or whether the journey will be worth the pain). The alternative, which I tend to prefer, is to enter enemy territory, briefly immerse yourself in ambiguity, uncertainty, alien vocabulary and ‘OMG I know nothing’ vertigo, then grab whatever is useful, and retreat back behind the development stockade to recover.
In real time, there is also a difficult choice between trying to follow every word (very difficult, and means you have to keep interrupting and asking for explanations, which gets humiliating after the first couple of times) or surfing the conversation, gliding over the bits you don’t understand and trying to grasp the overall shape of the thought process, but never quite knowing if you’ve missed the important stuff.
What, if anything, came out of my day of rage?
Actually it may prove quite productive. At a minimum, the Finch Fund (so named after the Galapagos finches that inspired Darwin’s theory of evolution) will look for projects that include:
- Multiple parallel experiments on a similar issue, where it’s OK (in fact required) to fail on some of them (a bit like the Tanzanian accountability project that I go on about a lot)
- Positive deviance: clear evidence that the teams have studied what is already happening in their countries and issue areas and identified particularly positive or negative outliers as part of the experiment.
- Fast feedback, especially from poor people, which determines how the set of interventions evolve and change.
- Don’t overdesign: built-in flexibility will allow any given intervention to morph and evolve in response to feedback.
If it just stops there, the Finch Fund will be pretty unique and I think could provide a really influential model for how to think more intelligently about ‘going to scale’. Some other, deeper ideas may have been lurking in the fog of jargon that could make it even more interesting. They revolve around how to identify systems that are particularly propitious for rapid evolutionary change. In general these have a steep gradient of difference between different parts of the system, a high level of ‘microdiversity’, and feedback systems with costs that fall as a change goes to scale. Not sure quite what such ideas would mean in practice – let’s see how the Fund gets on.
Two other dilemmas: what is the right response to a complex system; study it really hard to try and understand it better, or accept that you cannot fully understand it, so trial and error is a better approach? Reminiscent of the evolutionary culture wars between intelligent design and random mutation + natural selection!
Secondly, the toolkit temptation. If we want this approach to be adopted, we will have to set out guidelines for how it works, what questions to ask, what to avoid etc. i.e. a toolkit. But isn’t that contradictory, if a central point about complex systems is that they defy blueprints and best practice? Time for the padded cell………