What kind of science do we need for the aid and post-2015 agenda?

Spent an intriguing evening last week speaking on a panel at the wonderful Royal Society (Isaac Newton and all that), on the links between the post-2015 agenda and science. The audience was from the government/science interface – people with job titles like ‘Head of Extreme Events’.

I talked (powerpoint here – keep clicking) about how science can help developmentistas by bringing them up to date with what science is actuallyCynefin for UKCDSabout. Less Newton more Darwin, in terms of moving from a 19th Century world of linear causal chains, static equilibria and reductionism, to ecological and complexity thinking. I also tried linking some of the stuff I’ve been reading on complexity thinking with the Cynefin framework. It seems to me we need different kinds of science for the different quadrants:

  • Complex: complexity theory, evolutionary/ecological approaches
  • Knowable but complicated: more traditional analytic research methods aimed at nailing down causation
  • Known: Just identify and roll out best practice
  • Chaos: no idea – any suggestions?

The reason I like this is that it helps clarify when we need to bash our brains on complexity theory, and when we can stick with the old fashioned stuff. Convinced?

My other point was to stress that science has to address issues of power and distributive impact – issues like intellectual property rights and the current efforts to restrict poor countries’ access to medicines, but also the impact of new technologies. Geoengineering seemed to resonate as an example: it’s no good thinking about it as a purely technological challenge, you also have to think about winners and losers from its implementation (if they dump a million tonnes of iron filings in the oceans to absorb carbon, it isn’t going to be off the shores of Europe…..).

But enough about me, what did other people say? David Cameron’s post-2015 czar, Michael Anderson, was strikingly interested in complexity and uncertainty – a theme which dominated the evening. He stressed the political obstacles to taking them seriously – politicians don’t want to know; the public switches channel. Correcting that needs an educational effort from scientists, but also finding good ‘proxy indicators.’

Wrong model, guys
Wrong model, guys

Proxy indicators are magical: they take the temperature of a complex system well enough to be useful, and they communicate directly with policy makers and public. According to Michael, the maternal mortality rate is the perfect example – an excellent proxy indicator for the overall state of health systems and a powerful means of communicating with a wider audience. Michael reckons we need such ‘canary in the mine’ indicators to help tackle complex processes such as climate change, conflict or the sustainability of oceans (apparently phytoplankton levels are the best guide to ocean health, but don’t cut it with Joe Public, so they went for fish stocks in the High Level Panel report).

The overall discussion on the role of science was a bit all over the place – I guess ‘Science’ is a very big thing. Perceptions of science are deeply split: policy-makers see it as a source of certainty – ‘what works’, ‘this we know’, ‘facts’ – that they can cling to in their daily swirling clouds of opinion and ideology. But scientists don’t agree – they are much more aware of the limitations of scientific knowledge and the messiness of the world.

Some of the conversation was more about the downstream application of science to implement policies and achieve whatever goals are agreed. For Ban Ki-Moon’s post-2015 special adviser Amina Mohammed, the issues were building scientific capacity in developing countries (entirely missing from the MDGs), linking science-blind parliaments and politicians with nascent scientific communities, and dealing with slow/bad data.

Over dinner (Chatham House rules), multi-disciplinarity got a hostile reception – people reckoned that sometimes you need a single discipline, sometimes a combo – it depends. Better, perhaps to try and adopt a ‘problem driven approach’. Identify the problem, and then see which disciplines jump to the task – shades of Matt Andrews’ ‘problem driven iterative adaptation’ again.

The conversation got heated (appropriately) on climate change, with scientists laying into the civil servants about the necessity of at least discussing the implications for growth (‘growth is exponential; the planet is finite – it doesn’t add up’), and the civil servants wearily explaining the nature of political realities – you can’t question the primacy of growth and keep your job.

And one lovely quote from Isaac Newton himself, nicked from Ben Ramalingam’s forthcoming book Aid on the Edge of Chaos: ‘I can calculate the movements of heavenly bodies, but not the madness of men.’ True that, judging by an evening with the boffins.

Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


11 Responses to “What kind of science do we need for the aid and post-2015 agenda?”
  1. Catherine Dom

    Hi Duncan

    Interesting. I’m afraid I’m half convinced only by the “canari in the mine” idea. Ok, it may well tell you a lot about the state of health systems at large but nothing about WHY health systems are like this, which is quite important and back to complexity.

    I also don’t buy any longer (never have actually) this argument that ‘Joe public’ switches off when talked about the complex world. This is not what e.g. ODI research on the UK public shows. I think it’s a convenient excuse for politicians who indeed are scarred by complexity.

    Finally another lovely quote which I think is worth bearing in mind when thinking about sciences and development… Einstein: “Theory is when you know everything but nothing works. Practice is when everything works, but nobody knows why. We have put together theory and practice: nothing is working, and nobody knows why.”

    That said, I do like the ‘what can be done’ bullets depending on the type of problem. No idea for chaos…

  2. Thanks for your interesting blog, Duncan. What an interesting meeting that must have been. The cynefin model is exciting and a worry at the same time. How easy is it in practice to separate out contexts or problems in this way? For example, using drones to kill certain leaders of terrorist organisations is a simple problem with a simple measurable outcome. But the wider and longer-term effects may not be so simple. Conversely, the hunger safety net project I visited in Turkana in North Kenya (see http://www.embracingcomplexity.com/claremont/blog/?p=69&gi_sn=51b70b8db65fb%7C0 ) was a wonderful example, I think, of embracing the complexity of the situation – connecting humanitarian and development approaches, linking the core objective of providing cash to people in remote areas during a drought with improving supply chains for local fishermen. It is a complex intervention which has, in some ways, relatively straightforward outcomes. So simple projects may have complex outcomes and complex projects may have relatively simple outcomes.
    To give another example, fragility in certain nation states and regions is created, in part, because of entrenched, long-standing hostilities between factions. These locked-in factors can be known and knowable (to use the terms in the cynefin model) but lead to brittleness and lack of resilience in societies. The existence of entrenched hostilities between factions is one of the determinants of fragility, and can contribute to tipping a society into some sort of chaos. There is (clearly) a difference between stability (Syria was not regarded as fragile, did not score highly on measures of fragility) and resilience. But what is chaos? Chaos, such as we see presently in Syria, still contains within it these strong, instituted, historically-shaped oppositions between groups. So is Syria a chaotic regime or not? Do we ever find ‘pure’ chaos (there are no patterns and nothing can be known) or, if we think that way, do we miss how certain strong patterns of behaviour sustain and shape the future?
    So, are these regimes – of chaos, complexity and so on – really separable or, if we overplay the use of such a model, does it fuel our desire for more certainty than we can have? Does it, paradoxically, stop us looking at and engaging with the complexity on the ground, the complexity of what is? To quote Einstein, “Everything should be made as simple as possible, but not simpler.”
    On another tack, the interest in using ideas from the natural sciences to apply to the human world does create problems Adopting scientific methods for topics for which they was not developed is not a scientific thing to do, does not make it science. Traditional management is based on Newton’s mechanics, economists, have based much of their thinking on assumptions of equilibrium. As Mark Buchanan has argued http://www.amazon.co.uk/Forecast-Physics-Meteorology-Sciences-Economics/dp/1408827379 it is a shame economists did not follow the lead of the meteorologists and develop a science of disequilibrium and instability. Mark quotes Will Hutton … “Economics is a discipline for quiet times. The profession has no grip on how the abnormal grows out of the normal…. like weather forecasters don’t understand storms.”

  3. fred carden

    I am with Amina Mohammed on the need for enhancing scientific capacities- and science of all kinds. I think it useful here to refer back to Pasteur’s quadrant. Simply put, Stokes made the case that there are s different kinds of science with somewhat different methods. And this would apply both to the research and its assessment. Pasteur’s quadrant is about use-inspired basic research, or research that is solution oriented but includes basic research because we are far from solution. Bohr’s quadrant is pure basic research and Edison’s quadrant is pure applied research. One quadrant is left blank in this model, I guess because Stokes could not identify anyone who was interested in research that was both low on relevance to application and low on relevance to the advancement of knowledge. The more important point he highlighted is that there is more than one science. So we cannot treat its conduct or assessment as unitary.

  4. We hope to contribute one scientific solution to the development conundrum and that is a super low cost form of air transportation for people with very little cash available for transportation.

    Our wind powered aircraft (www.windpoweredaircraft.com) hold the promise of allowing people to leap-frog ground based transportation hurdles to move crops and goods to distant markets without using any fuel or emitting any greenhouse gasses.

  5. I do agree with Michael Anderson that there is something about being able to communicate a goal clearly and convincingly that helps win support and backing. Maybe it isn’t “joe public” who switches off, but politicians). Can we be creative and find more ways to express the challenges we face in emotionally compelling but scientifically robust ways?

  6. Hi Duncan,

    I have two generic answers. First of all we need a science that respects the facts that sereral sciences are needed. One could call that holistic, but maybe that word is correct but resonates less well lately.

    Secondly we need a science that pushes the idea that “leaving A” is more important than “defining B”. With that I’m not proposing to wander off in any direction, but IMHO >95% of effort to analyse the A->B route is waste and ‘d better be spend on leaving A to find A’.

  7. jerrold

    Exciting conversations can’t take away the trend in development where the phrase “armchair development’ is a reoccurring in social media lately.
    The field reality in sharp contrast with scientific approach , lack of capacity and political momentum is hampering sustainable action.

  8. Georg Lennkh

    On chaos:
    At a meeting organized by the French-led World Policy Conference, last december, a workshop dealt with major threats. Among many, three were discussed in more detail, climate, cyber-attacks and old age. we concluded, that there is no satisfying answer to any of them, and that we just have to be prepared to deal with consequences as they turn up (to note: this workshop was chaired by the French insurance group AXA)