Direct and indirect effects, actor interactions, and complex systems are not additive

From Doing Bad by Doing Good: Why Humanitarian Action Fails by Christopher J. Coyne. Page 147.

Economic life, legal life, social life, and political life are characterized by complex systems.12 According to Robert Jervis, systems have two defining properties: “a set of units or elements is interconnected so that changes in some elements or their relations produce changes in other parts of the system,” and “the entire system exhibits properties and behaviors that are different from those of their parts.” Consequently, systems thinking stands in contrast to linear-type thinking, which holds that changes in outputs are directly proportional to changes in inputs, and that such changes take place in isolation, meaning they do not affect other aspects of the larger system.

Because the man of the humanitarian system thinks in linear technocratic terms, failure is viewed as being caused by a lack of inputs and a lack of adequate planning. Linear thinking is also evident in the common assumption that if a certain amount of humanitarian resources succeeds in accomplishing the desired ends in one context, that outcome can be replicated in other contexts by replicating the inputs. Such thinking assumes that inputs are proportional to outputs irrespective of the specific context in which they are employed.

The fundamental dilemma is that while policymakers and humanitarian practitioners recognize that unintended consequences, or system effects, exist, and are often negative, they fail to appreciate that these system effects are the direct result of the linear thinking that permeates all aspects of the state-led humanitarian system. Indeed, by their very nature, large-scale government bureaucracies are characterized by topdown planning grounded in linear thinking, but humanitarian action takes place within an array of interconnected complex systems (economic, legal, political, and social) that constitute a larger complex system. This implies, as starkly illustrated in Libya, that humanitarians “can never do merely one thing,” even if this is the intention, because there are a series of unpredictable consequences over time and space that emerge from any single intervention in a complex system.

There are three interrelated reasons why system effects occur. First, while interventions in a system do have direct effects, they also have indirect effects that are often long-term and variable and thus not immediately obvious or observable. The notion of “blowback,” the unintended consequence of a government’s covert foreign interventions in the form of violence against the intervening government’s citizens, serves as an illustration of this logic.17 Blowback is not an intended outcome of the initial intervention, and it occurs in future periods, well after the initial foreign intervention, making it difficult to link the violent retaliation to the initial intervention.

Second, because systems are characterized by interactions between multiple actors, the relationship between any two individual actors will be determined not only by their direct interactions but also by interactions with, and by, others in the system. For example, the delivery of humanitarian aid to those in need is not just a function of the direct relationship between humanitarian practitioners and the recipients but also the external relationships of these parties with government actors or other local power brokers.

Finally, systems are not additive, and thus system outcomes are fundamentally different from the sum of the individual elements. For example, it may appear that providing more humanitarian aid will result in helping more people in need, that more aid will equal less suffering. However, this is not necessarily the case, as illustrated by the numerous examples of useless humanitarian aid clogging ports and logistical hubs, which in turn prevents humanitarians from importing different types of assistance that actually could help those in need. In such cases, simply adding additional resources—sending more donated items—does not correspond with achieving more of the desired end of helping those in need and, in stark contrast, prevents this outcome from being accomplished.