Article / Drylands / DRYLANDSCRP / Environment / ILRI / Resilience / SLS

Resilience of what, to what and for whom? Contextualizing resilience and development in drylands

One of the focus areas of the Livestock Systems and Environment program at the International Livestock Research Institute (ILRI) through the CGIAR Research Program on Dryland Systems is on adaptation and resilience. Here, the objective is to build capacity for adaptation to environmental change and resilient development processes. Over the next few weeks, we shall publish a blog series featuring newly-published work by scientists in the program, focusing on the subject resilience and issues around development resilience in the context of drylands.

This second entry in the blog series is written by Mike Jones, who leads the Resilience Thematic Group at the International Union for Conservation of Nature (IUCN) under the Commission on Ecosystem Management.

A member of the Muungano Makaror Farming group in Wajir feed their livestock with fodder harvested from their farm

This blog series arose as a consequence of my informal review of the Davies, Robinson and Ericksen (2015) paper just prior to its publication. I approach the debate from the perspective of an ecologist who has been managing the complexity of conservation and development challenges for 40 years. I found Holling’s panarchy and adaptive cycle (Holling 2001) to be useful tools for understanding complexity, systemic change, and the role that resilience plays in change processes.

This ‘Holling perspective’ affects the way in which I see environmental management problems and think about what might be done to improve the manner in which we seek the ever elusive goal of sustainability.

My other reason for using Holling’s metaphors is the finding of Mooney, Duraiappah and Larigauderie (2013): ‘Thus, the resilience approach [developed by the Resilience Alliance] is an important approach to sustainability and has the same goal as sustainability science, but it is built on an overarching theory that sustainability science per se lacks.’

I share the concerns of Davies, Robinson and Ericksen (2015) with regard to the manner in which resilience-building has become a mainstream development activity and agree that we need to focus the meaning of ‘resilience’ in a development context, so that aid dollars are spent in ways that effectively address the vulnerability of impoverished communities to undesirable change. A narrow focus on food security may not build resilience or do much to lift people out of poverty traps because these are complex problems that require an understanding of the drivers and interactions across scales that create poverty traps and lead to food insecurity. Enhancing resilience must include socio-economic development, and the primary aim of development should be to create systemic change that addresses the drivers keeping people locked in poverty.

What I have difficulty with is deriving some meaning from ‘resilience of the development process’. Holling’s models provide a method for defining a change process, in which resilience only plays a part. ‘Resilience of the development process’ seems to be tautological. Resilience is a characteristic of complex systems that affects the way that systems change. Too much resilience – the system becomes rigid and risks collapse when environmental conditions change. Too little resilience – the system is unable to recover from disturbance and becomes locked in a poverty trap. This view of poverty as a maladaptive system state characterized by low resilience stands in contrast to the view that poverty traps are resilient. Poverty traps are not resilient but they are difficult to spring because of feedbacks that prevent the accumulation of resources necessary for developmental growth. In the drylands where I have worked in Africa, the resources necessary to sustain pastoralist livelihoods have been expropriated by others to satisfy the growing demand for access to land.

In contrast to the ‘resilience of the development process’ the Barrett and Constas (2013) definition of development resilience is consistent with the meaning of resilience in complex systems. Their model applies to the lower end of the growth curve of Holling’s adaptive cycle and is an attempt to define what it takes to escape from a poverty trap and continue to grow. This is useful, and when combined with consideration of the interactions that occur across local, regional, national and global scales will enhance our ability to understand and address poverty traps.

Another part of the Davies, Robinson and Ericksen paper that leaves me searching for meaning is the notion of risks associated with resilience, when resilience assessment based on Holling’s sense-making models, is a method for assessing the risk of changing to undesirable system state.

Apart from my quibbles about the meaning of some of the statements made by Davies, Robinson and Ericksen, their paper raises important issues to be addressed if the mainstream is to be successful in its attempts to build resilience. In this regard it is worth considering that the concept of ecosystem services has been oversimplified to the point where people are unaware of the underlying complexity of the systems that they are managing (Norgard, 2010). Let us hope that the concept of resilience does not suffer the same fate and that a useful tool for helping us to understand and manage change in complex systems is not reduced to a simplistic device of little value.


Brand, F.S. and Jax, K. 2007. Focusing the meaning(s) of resilience: Resilience as a descriptive concept and a Boundary Object. Ecology and Society 12(1):23.

Barrett, C.B. and Constas, M.A. 2013. Toward a theory of resilience for international development applications. Proceedings of the National Academy of Sciences. 111(40):14625–14630.

Davies, J., Robinson, L.W. and Ericksen, P.J. 2015. Development process resilience and sustainable development: Insights from the drylands of eastern Africa. Society and Natural Resources. 28(3): 328–343.

Holling, C.S. 2001. Understanding the complexity of economic, ecological, and social systems.  Ecosystems. 4(5):390–405.

Mooney, H.A., Duraiappah, A. and Larigauderie, A. 2013. Evolution of natural and social science interactions in global change research programs. Proceedings of the National Academy of Sciences.

Norgaard, R.B. 2010. Ecosystem services: From eye-opening metaphor to complexity blinder. Ecological Economics, 69(6):1219–1227.

4 thoughts on “Resilience of what, to what and for whom? Contextualizing resilience and development in drylands

  1. I very much welcome Mike’s contribution here.

    I can’t speak for the co-authors of our recent paper (, but I, like you Mike, think that the ‘Holling perspective’ and the concepts coming from the social-ecological resilience community have much to offer to sustainability science and indeed to those of us now working on ‘development resilience’.

    For instance, I find that resilience thinking is highly useful for making sense of the interactions amongst phenomena such as poverty, corruption, environmental degradation and conflict. And while we agree on much, here is one area where it seems we disagree—I think that many kinds of poverty traps are a manifestation of social-ecological resilience, albeit perverse, undesirable resilience. System resilience is the capacity of a system “to tolerate disturbance without collapsing into a qualitatively different state that is controlled by a different set of processes” (, defined by characteristics of being able to undergo change and still retain the same controls on function and structure, of being capable of self-organization, and of having the ability to build and increase the capacity for learning and adaptation. It seems to me, this definition fits like a glove: interwoven causal dynamics among particular forms of resource use, particular power relations, and particularly forms of poverty and vulnerability often create a mutually reinforcing situation in which a social-ecological system is able to absorb the impact of changes in national government, the efforts of NGOs, and so on. In resilience terminology, they absorb disturbance, and in self-organizing they maintain essentially the same, and maintain essentially the same processes.

    We are interested, as you say, in “effectively addressing the vulnerability of impoverished communities to undesirable change”; but we are also interested in effectively addressing the vulnerability of impoverished communities TO STAYING THE SAME.

    I’m concerned that the development and DRR communities have subtly lost sight of the hopeful, transformative goal of human development. And so in our treatment of development resilience we didn’t distance ourselves from social-ecological resilience because the latter is misguided, but rather because it is not the right conceptual tool for capturing what we hope development resilience really means. Development resilience for us an ongoing and continual (if not necessarily smooth) trajectory of ever-improving human well-being.

    Let’s keep this conversation going. We in the development, DRR, and environmental communities definitely need, as you say, a deeper appreciation “of the underlying complexity of the systems” we are dealing with.

  2. Lance points to something that may have caused a great deal of confusion about the relationship between resilience and poverty with his reference to this statement posted on the RA website:

    “System resilience is the capacity of a system to tolerate disturbance without collapsing into a qualitatively different state that is controlled by a different set of processes.” (

    What is missing from this statement is an explanation of what happens after the collapse phase.

    When disturbed, a resilient system, collapses and either reorganizes and returns to the same state (i.e. it bounces back) or it reorganizes in a different way (i.e., it is transformed) and grows as a new system. The processes of collapse and renewal compared with collapse and transform are illustrated in Figs 1 and 2 of Cumming & Collier (2005). The relationships between resilience, potential for change and connectedness are clearly shown in Allison & Hobbs (2004). The statement on the RA web page is misleading. In a resilient system, collapse is just one relatively short phase of the change cycle. A poverty trap occurs when for some reason, the system has collapsed to the point where it has lost the potential for change, connectedness and resilience, and feedback within the system maintains it in this impoverished state.

    In drylands and semi-arid rangeland this condition is often created by the continual “press” disturbance of expanding human populations on a finite land base leading to declining per capita income, loss of plant cover, soil and water. A reinforcing feedback is created that pushes the social and ecological components of the system into deepening poverty due to a decline in primary production as a consequence of soil and water loss. This is evident in some rangelands where rain-fed crops are now grown in areas that were dry season grazing reserves required to keep stock alive during drought years. Grazing reserves for drought years in semi-arid systems are an essential part of the resilience of pastoralist systems.

    Maintaining the “slow variables” like forests, soils, and hydrology is one of the attributes of a resilient ecological system. Where these have been eroded, restoration is a long-term process that requires alternative kinds of land use and livelihood, i.e. a substantial transformational change in pastoralist livelihoods so that they can lift themselves out of poverty while the land upon which they depend recovers. Degraded rangeland systems are not stuck in poverty because they are resilient; they are stuck in poverty because they have lost the wherewithal to be resilient. That is why poverty traps are so difficult to spring; why development needs to be a long slow process; and why funding agents need to be prepared for high failure rates.

    Degraded rangelands are living systems of people and nature that have been broken by colonization and simplistic economic development thinking aimed at maximizing production with little understanding of how the system works or what is required to maintain its capacity to produce. These systems cannot be mended like so much machinery with short-term development projects. They require long term support while their resilience is restored.

    Allison, H.E. & Hobbs, R.J., 2004. Resilience, adaptive capacity, and the lock-in trap of the Western Australian agricultural region. Ecology and Society, 9(1).
    Cumming, G.S. & Collier, J., 2005. Change and identity in complex systems. Ecology and Society, 10(1), p.29.

  3. Another view in the resilience literature is that there can be two or more stability domains – e.g., the ball in cup diagrams. And in many versions of the adaptive cycle heuristic (Holling’s “lazy 8”), the collapse can sometimes be followed by an “exit” — i.e., the ball passes a threshold and rolls down into a different cup that may be less desirable than the previous stability domain. “Perverse resilience” was one of the most useful concepts to come out of resilience thinking; I hope it’s not being abandoned.

  4. The ball and basin metaphor is a way of understanding how systems can cross thresholds and enter a different system state. For example, savanna systems can exist as grasslands, woodlands or some intermediate mixture of grasses and woody species, i.e., three different states with different assemblages of species and interactions between them. The system will move between these states depending on climate (a larger scale stabilizing system), fire and herbivory. Each one of these states is resilient and a shift in state will depend on the outcome of complex interactions within the system, interactions between the system and climate, and the occurrence of one or more events that shock the system to the point that it shifts from one state to another. For example, a combination of high rainfall, high primary productivity in the grass layer, low levels of herbivory, winter frost and hot dry season fire after a run of “no fire” years can shift a wooded savanna to a grass savanna.

    There is nothing perverse about the resilience of any of these system states but any one of them may be undesirable from the perspective of the people who use them. Graziers would prefer grasslands, people who use wood products would prefer woodlands. There is no “exit” from the “lazy -8” involved in these transitions.

    Changes in system state are different from permanent systemic collapse into a poverty trap. A poverty trap occurs when the system is degraded by constant over-use that permanently removes species, erodes soil and soil nutrients, and reduces the ability of the system to retain water for primary and and higher levels of production. Systems in poverty traps have lost their resilience. This is where the “exit” from the “lazy 8” occurs.

    The only time that resilience might be considered perverse is when a system has accumulated and locked up the resources needed for change, and there are high levels of connectedness between all the parts of the system. For example old growth forests in ecological systems and the global financial system in social systems. These systems are “over-resilient”, or to use Holling’s word, they are in a “rigidity trap”. From the perspective of old growth forest, and the perspective of the financial services industry, these over-resilient-states are highly desirable.

    Resilience is no more than an abstraction to help us understand change processes in complex systems. Resilience is neither good nor bad in itself, but system states can be good or bad depending on who or what is benefiting from use of the resources (the social and ecological “capitals”) that exist within the system.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s