More books on complexity than you can shake a stick at—plus a mea culpa

Last month, Duncan Green was kind enough to post my overly ambitious multi-book review on complexity thinking in development on his From Poverty to Power blog. It covered three books: Ben Ramalingam’s Aid on the Edge of Chaos; Jean Boulton, Peter Allen, and Cliff Bowman’s Embracing Complexity; and Danny Burns and Stuart Worsley’s Navigating Complexity in International Development.

It was a bit much to cover three substantive works in a single review, so naturally I left a fair bit out. This follow-up post is my attempt to do slightly more justice to the topic.

But first, a mea culpa: I ended the review post with the observation that only a few of the examples in these books stem from complexity thinking; rather, the approaches developed separately and then their effectiveness was explained in complexity terms. I’ve floated the same point before, distinguishing between complexity-relevant practice (i.e. where we can explain methods in terms of complexity) vs. complexity-informed practice (i.e. where we proactively design work based on complexity thinking). I see a lot of the former, but not enough (yet) of the latter.

In responding to my post, Ramalingam made the case that this was a bit of an unfair characterization of the practice to date—and in retrospect, I think he’s right. More of the examples from his book were designed with complexity in mind than I gave credit. In fact, several of the other authors (Burns, Worsley, and Boulton) also chimed in to note that they appreciated the review overall, but that I had somewhat missed the mark on that point. See the comments sections on the post for more.

This is a happy mea culpa though, as the practical upshot is to say that complexity thinking is shifting development practice somewhat more than I had previously thought. Perhaps the authors and I can still agree that we’d still like to see it shift more?

In any case, let’s turn to a few of the interesting stray points from the books that didn’t make it into the larger review.

Navigating Complexity (Burns/Worsley)

I referred to this book as the most practical guide to complexity in development, and a good pick for program designers, evaluators, and managers (as opposed to the other two books, which are a bit more strategically oriented). Burns/Worsley move quickly through a history of other approaches to development (big pushes, technical focus, institutions, rights-based, etc.) and complexity thinking itself, in order to focus on the practical implications.

They describe successful approaches to change as relying on participation, learning, and relationship and network building, which together allow us to understand the system dynamics better and generate appropriate interventions with ownership by stakeholders; ownership then leads to sustainability and scale. (Everything in italics is a concept that they discuss in much greater depth, so if any piqued your interest, you should read the book.) I like this framework because—although it might make Burns and Worsley cringe—you can turn this into a quick rubric for assessing a project or approach, based on how well it incorporates each of those elements.

The core of the book focuses on three approaches that build on these concepts: participatory systemic inquiry, as a planning tool; systemic action research, as a structured process that integrates learning and action; and nurtured emergent development, as an organic approach to building a social movement for change (explained with examples like community-led total sanitation).

They also have a great discussion of the nature of power in development and participatory processes, and a conclusion section that nods toward the political nature of development and the need for adaptive management.

Embracing Complexity (Boulton, Allen, Bowman)

This book was much more wide-ranging, in a way that makes it hard to give full due in any review. As mentioned in my other post, this book was written by a team that deeply understands the science of complexity, as well as the debates within that field. The result is an great journey through the ideas, easily followed by non-specialists, but which still may be more detailed than some readers want.

For those who embrace the journey, the chapters on complexity in the social world, management, strategy, international development, and economics are chock-full of insights and advice. For example, I took the following away on complexity-informed research processes: trace situations over time; include multiple-perspectives and opportunities for meaning-making; follow emergent, unexpected phenomena; explore causes; explore around issues; stay free from the need to define initial hypothesis; balance uniqueness and general applicability.

Similarly, on complexity-informed change management: understand the context and history; ensure broad stakeholder engagement; agree on a broad approach to change and create a general plan, but keep a contingency budget for experimentation; design pilots within the broad approach (small bets, anyone?); refine and iterate; allow a degree of local customization.

The chapter on development draws on work that Boulton did with Oxfam around programs in Kenya. It advocates for considering how interventions take account of the systemic/interconnected aspects of the context, how they pay attention to path-dependency, how they customize and experiment within context, and how they look for and respond to tipping points in the wider environment. One of my favorite takeaways was the idea of a new definition of “baseline” assessments: paying better attention to the starting point in a richer way, identifying contextual and historical factors and how each might relate to the intervention.

Aid on the Edge (Ramalingam)

I won’t share as much on this book, as it came out a few years before the others and has been subject of more reviews. This one is somewhat heftier than the others, though those who are familiar with the sector may be able to skim over the critique of how aid currently works (the first third of the book), instead focusing on Ramalingam’s description of complexity (second third) and its implications for the sector (final third).

What I think Ramalingam’s book does very well is make the case for changing the sector’s broader thinking, as well its accompanying approaches and toolkits. He notes the need to: “move from ‘experiments’ as a tool to ‘experimentation’ as a mindset.” The idea is echoed in points made in Boulton et al. about complexity being as much a worldview as a science.

So now what?

The question facing an interested practitioner is: okay, I’ve got the mindset, I’m in the zone; now what tools do I use? As J. asked in response to my previous post:

Tomorrow morning, when I sit down at my desk to manage a team of unruly aid workers, tasked with implementing a diverse portfolio of grant-funded activities in an ambiguous legal context, HQ riding my ass for an updated strategy document that no one will ever read… and knowing everything in those books: Which email do I send first?

The answer, unfortunately, is a big fat: “it depends.” Increasingly, that toolkit is being built out under a variety of headings: PDIA, Doing Development Differently, adaptive management, adapting development, adaptive learning, “collaborating, learning, adapting” (CLA—USAID’s preferred acronym), “science of delivery” (the World Bank’s framing, though it’s approached more as craft than science), and on and on. The differences are often shaped by the differing institutional contexts from which they emerge. Not all of the practice is complexity-informed, but all of it is complexity-relevant.

From where I sit—as a consultant who gets to be connected to a variety of these efforts, but often in small ways—these different approaches to fundamentally the same questions are starting to converge. People from different institutions and sub-sectors are starting to talk to one another. The changes are bubbling up into bureaucratic rules and operational procedures and hiring practices and more. I remain, as always, cautiously optimistic…