There may be many definitions of Resilience. But from the literature I read this week, it seems that Holling's definition is accepted by most researchers in this area. To use this definition, first of all, the fundamental unstability of social-ecological systems must be acknowledged. The equilibrium or stable state can only be maintained near certain points. In most other areas of the state space, there are only domains of attraction or areas without equilibrium. Thus, while stability is defined by Holling as "the ability of the system to recover back to the
equilibrium state after disturbance" and is rare and hardly achievable, resilience is defined by him as "a measure of persistence of systems and of their ability of absorb change and disturbance and still maintain the same relationships between populations or state variables". The important thing, he said, is the system's ability to maintain relationships and structures within itself in the face of changes. Systems with more attraction domains and options are more resilient.
His followers developed the concept further. Carpenter et al. (2001) claims that there are three properties in Holling's resilience definition: (a) the amount of change the system can undergo (and therefore the amount of external force the system can sustain); (b) the degree to which the system is able to self-organise; and (c) the degree to which the system can build the capacity to learn and adapt.
Measurement:
Some years later, Carpenter et al. (2001) suggested a tool for analysis of resilience: the adaptive cycle. By defining four stages in a cycle of system configuration and services, the key variable influencing the system's ability to provide services and other slow-changing variables which have fundamental and intriguing effect on the system may be identified.
Another useful tool is bifurcation analysis, which serves to identify domains of attraction in the space of parameters (Anderies, 2000).
Combining these tools and social science / ecology study, with the help of stakeholder participation, it is possible to find out resilience of which factor in the system is concerned, how is it affected, and what is the threat, i.e. "resilience of what to what" (Carpenter et al.,2001). Walker et al. (2002) proposed a 4-step method to assess regional resilience: First, the stakeholders get together to decide the key properties of the study system and the range of trajectories the stakeholders want the system to follow (step 1 and 2); the information is then used in more specialised, quantitative analysis on where th resilience resides (step 3); finally, an integrated evaluation of management and policy implications is produced by both scientists and
stakeholders (step 4).
Pay attention to:
- Technical uncertainties vs. structural uncertainties. The former one is unavoidable and often addressed by researches; the latter one comes from unexamined assumptions which may be wrong in certain circumstances, and can lead to wrong discourse and solutions.
- Which configuration of the system is more desirable? In what time scale should resilience be considered? If sustainable development is to be considered, resilience at the cost of other area or era's collapse may not be a desirable goal.
- The strategy and institution to achieve resilience are very likely to be very system-specific. They also tend to change with system conditions and external factors, such as climate, legislation and neighbour systems.
- Difficulties in resilience measurement include (but not limited to): key drivers hard to find and predict; human actions and reactions change system's course in complicated ways, making prediction more difficult; systems may change too fast in sudden events.
- All these researches are based on Western perspective and literature. How to apply and adapt them to my research area?
Questions:
- If a system achieve Holling's resilience, what is the implication of this? What will be the effect of such resilience?
- How to address slow-changing variables in limited time for research?
- Among the 3 properties of resilient system, is anyone more desirable or important? Should all systems have all 3 of them? Should the 3 be considered together?