Fragmentation of the Thinking and Working Politically agenda: Should we worry?
29th August 2016
Recently, I’ve read many articles and heard from many colleagues who are concerned about the apparent competition between the Doing Development Differently (DDD) network and the Thinking and Working Politically (TWP) community of practice.
Dave Algoso and Alan Hudson have made the case on Duncan Green’s blog that there are in fact nine separate initiatives all pushing in roughly the same direction. In a hilarious comparison, they argue that international development reform advocates may be heading for a Monty Python “People’s Front of Judea” scenario, where rebel factions are too engrossed in their disagreements with each other to combine in mutual opposition to Rome.
Both [TWP and DDD] share a sharp critique of rigid, pre-planned program design and have strong words for the standard ‘good governance’ agenda.
Should we be worried? TWP and DDD are part of a much wider response to fundamental problems in the assumptions guiding development assistance. The people involved in the TWP and DDD movements have responded to those problems with similar conclusions, though they have taken quite different paths and are appealing to different audiences. While there is some overlap in the people involved, the DDD community was co-led by a group of researchers at Harvard who were focused on large-scale programs designed to build state capacity, particularly those funded by the World Bank and other multilateral development banks. The TWP group was largely driven by people from bilateral donors and INGOs who were reacting to the internal factors that discouraged aid programs from grappling with political challenges. Both initiatives share a sharp critique of traditional aid programs’ reliance on external, best practice ‘solutions’ and rigid, pre-planned program designs. Both groups have strong words for the standard ‘good governance’ agenda.
I would argue that the creation of two separate groups is absolutely natural, and probably useful. DDD is appealing to a set of actors who have more difficulty talking about politics, including the World Bank. TWP is appealing to donor officials and practitioners who may have more scope for engaging with politics, but who need to challenge the simplistic and technocratic assumptions that tend to underpin development assistance.
They are closely related, of course – in fact, I would say that 80-90% of the recommendations from DDD and TWP are the same. But their differences are quite helpful at times. For example, it is much more effective to frame the challenges of large-scale technical assistance projects (especially those supporting government ministries) by using the DDD framework, to encourage more iterative and entrepreneurial approaches to reform. However, when we’re trying to support multi-stakeholder reform efforts around important economic or social issues, the TWP approach is much more useful. It helps us think about obstacles to reform and who we should support in the reform process.
So I disagree with the fragmentation narrative – but there is an important point to be made about definitions. A major, ongoing challenge for both communities has been defining what we mean by politically-smart programming and, similarly, iterative and adaptive programming. It’s very common to hear program teams and partners talk about how they have used TWP approaches and then offer examples that sound very traditional.
In the past few years, for instance, I have read several proposals for building multi-stakeholder coalitions for reform that really look a lot like old-school NGO advocacy projects. But what exactly is the difference? Civil society strengthening and advocacy programs have long supported actors to engage in political dialogue and to pressure elites and governments to support reforms. Yet these approaches often fail because the ‘coalition’ is mostly made up of like-minded actors from marginalised communities and NGOs. They have no connections to people in positions of power and overwhelmingly use confrontational approaches to influence policy. We know that this approach on its own is rarely sufficient, especially when there is no one on the ‘inside’ who can translate outside concerns into politically feasible proposals for change.
In their FP2P blog post, Dave and Alan list nine initiatives, including DDD and TWP, currently being used in the push to reform international development practice. Some seem to be apples and oranges. For example, the Smart Rules initiative is an internal DFID effort to open up more space for programs to operate in flexible, politically informed ways. Similarly, the Collaborating, Learning, and Adapting initiative at USAID is an internal process to carve out more space for iterative program designs, and to link US-supported programs to other development initiatives. It is difficult to draw comparisons between them and initiatives such as DDD and TWP, which have created open networks of donors, researchers, and aid practitioners who support broad aid and development reform agendas.
…there is certainly scope to get better at defining what we mean by politically smart, iterative programming – but there may not be one definition we can all use.
Some seem even further removed from TWP and DDD. For example, the Global Delivery Initiative is closely linked to the Science of Delivery agenda at the World Bank. GDI advocates experimental, iterative approaches for development programs – but within the broader objective of finding universal solutions for development challenges that can be applied anywhere. This initiative seems to me to be moving away from aid delivery that responds to local country contexts and the political and governance challenges that are unique to particular countries. It’s very hard to reconcile universal best practice concepts with iterative, politically smart programming – basically, they are pushing in opposite directions. These differences go well beyond the People’s Front of Judea vs. the Judean People’s Front vs. the Popular Front of Judea.
In conclusion, there is certainly scope to get better at defining what we mean by politically smart, iterative programming. But I don’t think we should assume that there will be one definition that everyone can use. This is a broad movement with many audiences, so we need to tolerate some separate tracks and slightly varying approaches. My suggestion for development practitioners and donors would be to consider the subtle differences between them, and use them to your advantage.
Image: Fragments (Photo: Antii Kyllonen, Flikr)
09th September 2016 at 08:49
Briefly, a comment from a practitioner organization looking to enhance our own adaptive and politically-savvy development work. It's optimal to have choices, and I don't mind at all if what's on offer is a mixture (not only apples and oranges, but give me aubergines too), so long as there is plenty of credible information about the drivers & purpose of each of the different approaches and initiatives, their comparative strengths, etc. And, importantly some notion of what the application of these initiatives resulted in: who is tracking the 'success rate' of these? I very much welcome Alan & Dave's attempt to name them together in one place, and categorize them, as a starting point to articulating their similarities and differences. The analogy with various Judea fronts focuses on ideological differences - but from our perspective, we want a bit of ideology (looking for broad alignment) but a whole lot of practical and tactical know-how and experience. Optimally, we would create a marketplace of ideas and approaches to adaptive development and learning, where we also ought to be comfortable with the fact that some of those ideas might be better or more widely applicable than others, and some might end up fading, or being absorbed by others. So, who else is peddling their goods?
07th September 2016 at 22:58
Thanks for picking up the thread of the blogpost that Dave and I put together, Thomas. Three quick points in response:
1) We agree that the existence of multiple conversations about related things is not necessarily a problem (and perhaps the reference to the People's Front of Judea suggests an antagonism that isn't there!). But we do think that there is a risk of duplication and fragmentation, and that there is value in maintaining the connections amongst the various conversations to guard against that.
2) We agree that the TWP and DDD conversations have significant overlap, but would maintain that all 9 conversations that our blogpost mentions have a common core - they are all fruit, even if not all apples - which is adaptive learning (see our Learning Plan for more on this). Linking up the conversations can help with efforts to clarify that common core and can also capitalize on the complementarities e.g. DDD is a bit quiet on politics, something that the TWP community might help with. We've recently set up the AdaptDev Google Group to help connect the conversations.
3) On our reading, the Global Delivery Initiative is not - despite the language around 'science of delivery' - about finding one-size fits-all solutions. Or, put differently, while the Global Delivery Initiative is about identifying common functional challenges, and gathering evidence about how they've been addressed, it's not about finding or promoting blueprint solutions.
Here's a blogpost on our learning plan:
And here's some info about the AdaptDev Google Group
Thanks again for continuing the conversation!
The views expressed in Opinions posts are those of the author(s) and do not necessarily represent those of DLP, the Australian Government or DLP's partner organisations.