Alex, I totally agree. Words like “resilience” and “robustness” are almost always used inappropriately, and without a clear understanding of networks and complex systems.
Resilient To What?
First off, systems can only be “resilient” to a particular kind of perturbation like random failures or targeted attacks. Moreover, different network topologies are robust to different things. A system that is robust to random failures may not be robust to targeted attacks that take down critical hubs. It all depends on how the elements of the system are connected.
Resilient In What Context?
Second, systems can only be resilient in a particular context or environment. When defending from attacks by wolves, a buffalo would turn sideways to present itself as very large and threatening — an effective response. However, when frontiersman showed up with guns, this previously successful adaptation became a maladaptive liability, since it present an easier target to the scores of buffalo hunters. This is not a minor maladaptation either, as their near extinction was the result. This process works the same way when inhabitants of an ecology are well adapted to each other, but a new species is introduced, which then subsequently disrupts the entire ecology.
Resilience Is Complex
Finally, as Stu Kauffman has pointed out, even the process of learning to be resilient changes over time. Systems can become better or worse at being resilient and learning to be resilient. In other words, adaptive systems can become better at adapting. Conversely a system that is supposedly “resilient” is always resilient only within a certain scope or context. A system can be resilient therein, but actually by vulnerable if that context changes. There are many overlapping contexts of resiliency operating all of the time.
Consequently, although I do use those words, I always seek to clarify them and encourage others to also be more clear as to what they mean when they use those words. Glad to see I’m not the only person who has noticed. ;-)