Because resilience thinking has increasingly been used in various disciplines and domains and extended with a broader scope of concepts, it is difficult to find a unified and encompassing definition by which it can be accurately referred to. Furthermore, to elucidate and synthesize all resilience theories and conceptual frameworks that have been put forward will require volumes of written work. We therefore argue in this paper that research works pursuing the common strategies of system resilience require a language that can help describe the specific contexts in which resilience is applied. We propose here a taxonomy for general resilience that consists of four orthogonal dimensions, namely, type of shock or perturbation, target system, phase of concern, and type of recovery. Furthermore, it has also been observed that despite its domain-dependency, there exist resilience strategies that cut across multiple disciplines and domains, specifically, redundancy, diversity and adaptability. There is another, however, which we argue here in detail that is equally compelling, i.e., a strategy that can break the rigid stability that leads to greater fragility and a more severe collapse with prolonged period of severity. Specifically, the strategy is to deliberately inject or induce regularly small 'controlled' shocks into the system to regulate the build-up of complexity and rigidity among its components and their connections. Doing so will not only prolong the period of stability of the system, but also shorten the period of severity in the wake of a severe shock.