Do fragile states evolve like forests? Insights from complexity thinking

Oxfam’s engagement with physics-trained complexity enthusiast Jean Boulton is starting to generate some really interesting ideas. Jean has been helping us think through our work in fragile states – the big challenge for a lot of aid organizations over the next few years. Just before Christmas, she came in to tell us where her thinking has got to on the different kinds of fragility.

Drawing on Buzz Holling’s work researching the life-cycle of forests (I love this kind of disciplinary boundary-hopping), Jean looked at how complex ecological systems evolve over time, how diversity leads to resilience and how even natural systems can become rigid and brittle as they mature (see slide).

Boulton fragility cycleWhilst societies and nation states are subject to much more interference and influence from the wider world than is typical for your average forest, the work is important as it is based on observation, and cuts through the seemingly endless (and not always fruitful) discussions as to what complexity words and theoretical concepts actually mean.

Turning Jean’s cycle into a much more boring table

Stage Characteristics Typical aid interventions Countries that may look a bit like this
Birth Emerging out of chaos; no fixed patterns; lots of growing (and vulnerable) shoots; power and control fluid Be prepared for sectarian conflict and potential to fall back into chaos; nurture new shoots as they appear; support ‘structuring’ and emphasise the voice of civil society in influencing new institutions and in power and resource-sharing; recognise the potential for large corporations and others  to exploit the power vacuum South Sudan


Youth Diversity; resilience; stable due to multiple relationships; power sharing Act to strengthen capacity and livelihoods, maintain  and expand economic and political diversity ?
Maturity Efficient, but with falling diversity; relationships are reducing in number and becoming fixed; power static and concentrating in fewer hands Support minorities and the less powerful; defend diversity; build livelihoods, education and strengthen civic capacity and voice to combat potential for concentration of power in hands of elites Mali
Old Age Locked in, rigid relationships, autocratic power Support citizens movements in breaking/disrupting rigid power relationships. OPTI
Potential for collapse Brittle to the point of failure; power battles re-emerge Protect ‘islands of effectiveness’, strengthen the voice of civil society, be cognizant of future potential economic or geo-political ‘tipping points’ Yemen
Chaos Many structures collapsed, fast-changing dynamics and shifting power bases, great potential for violent conflict and taking advantage of power vacuum Humanitarian response; supporting civil society and non-violence movements, ensure the voice of the citizen is heard in the international arena, continuing to envision post-conflict solutions Syria

The reason I like this typology is both that it creates a dynamic sense of how fragility evolves over time, and the implications for would-be ‘change agents’, whether local or foreign. For example, when institutions are new and malleable and vulnerable, there is a lot more sense in devoting time to influencing them and protecting them from co-option by elites, than when they have become sclerotic and autocratic. It also creates a sense of ‘seizing the moment’ – institutions, once co-opted by the powerful, become much more resistant to ‘democratisation’; equally, when institutions collapse, the potential for various groups – ideological or economic – to exploit the power vacuum, is high.

I’d be interested in your additions/improvements, as I have a feeling I will be using this in the future.

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.


19 Responses to “Do fragile states evolve like forests? Insights from complexity thinking”
  1. Catherine Dom

    Two quick thoughts.

    First, ‘merge’ the column on “typical aid interventions” as you have it with Jean’s earlier and somewhat ‘looser’ (not at all in a negative sense, maybe I mean less prescriptive) typology which is at her page, almost at the bottom – that is
    Self-regulating – Experiment, Encourage diversity, keep some slack, Allow some variation in approach
    Lock-in – Explore what will break the lock, allow innovation and new entrants
    Chaos – How to cohere emerging structure, build stability.

    Second, ‘merge’ with power analysis – which may indicate that in some circumstances there are voices other than civil society which would be useful to support.

    Haven’t thought through the implications of trying to do this, I leave it to you.

    More generally, I like the image as well, but wonder where its relevance stops. One difference I think is that forests perhaps always go through this life-cycle (do they? I’m not a specialist – or are they managed in such a way that they don’t go all the way to collapsing?) whereas the different stages are different countries and presumably we don’t want to see South Sudan becoming Yemen in a few decades?

  2. Dear Jean,
    This is very interesting and is possibly a new framework to think about evaluating aid in conflict which is the focus of one of our work streams. (An example of ‘youth’ stage state may be Nepal?) Please let me know if there are references that you can suggest. I’d encourage you to take this forward and think about how this framework might be exploited to think about both – classes of policy options, and, possibilities for evaluation methods. Best wishes, Jo

  3. Aldo Matteucci


    good try, try again. As Jean says:”Such mechanical thinking can lead to blindness to change and difficulty in adapting to shocks and fast changes.” Anything that smacks of “cycle” implies inevitability, dharma, or what have you.

    If a state evolves in “life-cycles,” intervention will do what? Hasten it along? Make the inevitable happen? Break out of the cycle – into what?

    Evolution has no direction – contingencies make it up as it goes along. Cultural constructs – like states – are even more based on contingencies, and an ever changing political and social context. Your “cycle” simply implies that the future is past in the making.

    To wean you off :=)) such ideas so early in the year, check the cover of CONWAY et als.: All yesterdays. It shows Protoceratops (a kind of dinasaur) living in trees: “it is not necessarily because that’s what they are good for, but simply because they can.” (pg. 40)

  4. Just wanted to add that I am not meaning to suggest that a nation state inexorably moves around such a life-cycle. Indeed the potential to collapse partly or fully into chaos is a threat that is very real for all the countries listed. Rather, the idea that fragility is not just one ‘state’ nor a fixed ‘state’ points to what sort of interventions are most helpful to increase resilience.

  5. Peter

    This implies that there is an endless cycle for violence and fragility. Or is it meant to portray “maturity” as a kind of aimed for steady state?

    I don’t find such typologies helpful unless they work historically and across geographical boundaries. Where do rich countries fit in? At the least you should be able to map in and track the development of say South Korea in 1954. Or what about Chile under Pinochet. Not to mention, say, US civil war or Europe in 1945. And what do they tell us about sustainable ways out of this cycle?

  6. Fantastic analysis. The complex-system process map shows that only after you fully understand a problem, can you acquire the capacity to solve it. And in order to understand the problem, you need the right conceptual tools (For example, tables?).

  7. The usefulness of these images for me is that they get us to explore the idea that fragility is not one thing and that there is a dynamic about it – things can get more or less ‘fragile’, using the word in the widest sense. It takes us away from seeing fragility as a definite thing and helps to consider how different situations need different interventions. I’m not suggesting there is an inevitable trot round this cycle … but there is a difference between considerations of ‘locked-in’ factors which are very difficult to shift and ‘chaotic’ factors where there is little to hang on to, and these images of forests help to give some meaning to this, I think

  8. Aldo Matteucci

    There is no historical evidence for the Tytler cycle, nor for yours. At best you get modern versions of Malthus: TURCHIN- NEFEDOV, but these instances relate to extractive empires, which have gone by the board with industrialization.

    Two principle remarks: (a) a complex ecological system will tend toward a steady state – the climax – and then fluctuate within a narrow band: it is a homeostatic system. (b) if you want to get evolutionary change, look up “niche creation” (LEWONTIN, then ODLING-SMEE): this concept has been spelled out about 1988 and has yet to come over into the social sciences. The core point is that life transforms the environment as it is shaped by it. This applies also and in particular to culture. So you’ll never have a “cycle”, but a drift.

    South Korea: until past 1960 (until Sigmam Rhee was gone) South-Korea developed far less than the North. Then the North involved. Now the South is 15 times richer than the North (by comparison West-East Germany: 3:1 at reunification). LANKOV is a good read.

  9. David Jacobstein

    Biggest value-added I see is the characteristics of maturity – that regular relationships may also imply a degree of static and loss of dynamism – and how that spills over into cycles of renewal.

    I agree with others that mapping countries on this is a bit too cyclical. I would rather pull out concepts from it that apply within countries, to certain sectors, or time period. Helps to identify likelihood of critical junctures in some respects (you could view the age categories as related to frequency/significance of such junctures, and somewhat predictive in that vein).

  10. Cynthia Neudoerffer

    Holling’s cycle is not a circle but a figure 8 – with the front loop being quite differentiated from the back loop – I am not sure that flatening it into a circle is helpful. Also in their book “Panarchy” Gunderson and Holling describe resilience as another (3rd)dimension of the adaptive cycle – not just a part of the front loop (see Panarchy book Figure 2-2 on page 41.) Although I think their definition of resilience in this context is more appropriate for ecological systems and is too narrow when applied to social-ecological systems – in particular where power and agency need to be clearly taken into consideration. Resilience in the ecological system context is about innovation and experimentation in the back loop in order to negotitate the backloop and stay on the same attractor (not flip onto a new attractor state but start a new r->K front loop on the same attractor) – in work I have done on transfomtation in social-ecological systems – specifically community-based resources management systems – I have found it also helful to think of resilience as having the ability and power to innovate and actually ‘engineer’ if you will a back loop in advance of an unexpected crash – and that part of social-ecological resilience is being able to innovate and foster transformative change that will lead to moving to a more desirable new attractor (though of, course who gets to decide what we call ‘more desirable’ is one place where questions of power come in …)

  11. David Hoehner

    Nations would be confined to this cycle only if there were no internal means by which new and diverse political and economic “species” could enter the ecosystem. This model implicitly endorses liberal political and economic orders, which allow for such regeneration through voting and creative destruction, respectively. Their power to foster regeneration, of course, is threatened when political or economic compeitition are weakened.

  12. Just to say that most people’s concerns about this blog seems to be about the implied suggestion that fragile states have a life-cycle and I would fully agree with this concern. It is not what the talk I gave focused on and not what I am suggesting.

    The interest for me in Holling’s work on forests is that it make tangible what a lot of words about complexity – resilience, non-linearities, critical junctures etc do not. It shows, in fact, (if you follow the link Duncan gave to what Holling said) the importance of diversity and of variation and shows how positive feedback loops can lead to the powerful becoming more powerful even in nature – where balance is not always permanent. For me, it brings alive what complexity is about.

    The project of seeking correspondence between worlds (between physics and economics or ecology and humans) can turn into a scientific mechanical project and I have come to feel that working with complexity theory, which is derived from differing mathematical models with differing assumptions, can, with all its language, just turn into another positivist project, giving the illusion that we will find new shiny tools and methods, but still within the same toolkit that held the mechanical methods. I have spent many years working with these concepts and have my own views on what things mean and what matters, but in the end it can inhibit people from actually engaging with the complexities of the world and seeing what is really there and learning through experience.

    The interest for me in this notion of fragility as a dynamic is that it helps me to think about fragility, or lack of resilience, as not being a certain ‘state’ that can be measured but being something that changes and also that different nation states or regions will have different types of characteristics and may thus need approaching differently. Some may be moving out of chaos (although they might move back in) and some may be brittle, at risk of collapsing into chaos. Together with a systemic and wide-ranging context analysis, which takes into account wider and more local factors, looks at the past and into the future, these notions of differing and dynamic types of fragility I think helps to loosen perspectives. Hope this helps to explain from where I am coming.

  13. This is really interesting. Might also be interesting to consider the (somewhat clumsily termed) notion of anti-fragility coined by Nassim Nicholas Taleb, author of The Black Swan. I’ve heard of it in climate adaptation and resilience circles. Rather than focus on regaining as much lost ground as possible (resilience), the notion is that stress provides an opportunity to bounce back stronger, with the metaphor of physiology (think of training adaptations for VO2Max for endurance athletes or broken bones healing stronger than their pre-break state). Applied to vulnerable communities, populations, and states, it’s quite compelling in my opinion. With climate change exacerbating poverty and resource degradation, especially in the tropical developing world, finding ways to harness the hardscrabble ingenuity of communities would be awesome.

  14. Simone

    Just a comment about ecology.

    Aldo said: “Two principle remarks: (a) a complex ecological system will tend toward a steady state – the climax – and then fluctuate within a narrow band: it is a homeostatic system.”

    Even if I agree with his other opinions, the idea of a climax for ecosystems it’s not entirely true.

    Some systems may seems to tend toward it and reach a steady state (with many, many gaps of changes and disturbances inside), until a major disturbance will arrive (and it will arrive sooner or later). Some other systems crave for dramatic disasters that will help them regenerate (i.e. mediterranean vegeration adapted to flash fires).

    Just my two cents.

  15. I’m with Simone on this. The idea of a single steady state is one thing that is opposed by complexity thinking in its focus on open systems, nonlinear interactions and variation.

    Homeostasis, ecological balance can occur when feedback loops are negative, but when feedback loops are positive then there is at best multiple possible equilibria and in many cases, as conditions are not in general stable, systems do not necessarily reach equilibrium. Brian Arthur picked up this point in looking at economic systems and Holling demonstrates this for ecological systems too. In ecological systems a new entrant into the ecology may replace existing species, and may in turn lead to the destruction of its habitat. No room to develop this here but this has resonances with how minority groups can end up taking control of countries, how corporations grow and dominate their supply chains etc etc.

    Someone earlier mentioned niche creation. This has been developed in economic thinking by Frank Geels.

    Anyway, I see myself sinking into the theory debate so will resist further temptations!

  16. Paul Chatterton

    As an environmentalist, I like the elegance of this theory – but it may be a little of a stretch to apply the metaphor to forests as it draws from an old fashioned theory of forest ecology. The approach is partialy correct if you apply it to a single tree. Trees increase in ecological connections as they age and then senesce and die. However they also tend to reach greatest ecological complexity at oldest age when they have most epiphytes and holes where birds and animals can nest. By contrast, the forests as ecological systems does not senesce (assuming it is not disturbed) but rather reaches a steady state of maximum diversity, maximum relationships and for instance maximum carbon retention which is stable over time. It is this uninterrupted interrelation of rich parts that could perhaps be a goal of good governance theory. The ideas of forest overmaturity and chaos were used as an excuse by logging interests to destroy his steady state. If you want to use them in this model, you would need to consider the chaos as a break from the steady state.