Complexity – or the Governance of the Inherently Unpredictable

A core test of any international relations (IR) theory or model is how reality-inclusive it is. The impediments to getting such a theory or model right, however, are more than conceptual – the very language you use can act as blinders rather than a source of revelation. That is why beginning in the 1980s we saw social scientists and security analysts attempt, in earnest, to use complexity theory and chaos theory to tease out new, expanded meanings and insights in their fields.  Although their attempts have perhaps yielded more useful metaphors than conceptual breakthroughs, the off-kilter perspectives provided by these theories continue to provide added value to researchers. To illustrate this point and provide added context for today’s lead article, the following primer looks at how complexity theory is helping us to define and analyze today’s IR environment effectively.

One prominent narrative depicts the post-Cold War security environment as one characterized by diffuse and potentially unknowable risks instead of known threats. Another way to capture the dynamics of the current international environment is through the lens of complexity, a term that denotes the challenges associated with analyzing and understanding as well as managing a system. The complexity paradigm is concerned with situations that are unpredictable by nature and not just by virtue of the limitations of the observer.

 Chaos and Complexity Studies

The precursor of complexity thinking is the investigation of dynamic systems. Very simply put, a system is a set of mutually dependent components or variables. Each component in the system stands in interrelation with every other component in the set and also in interaction with the system environment. In other words, the components interact with each other within the system’s boundaries to function as a whole to perform a task. Three types of systems exist:

• Ordered systems are those that are structured around a set of rules or laws that contain clear patterns where reliable outcomes can be determined. They are characterized by stability and discernible cause-and-effect relationships.
• Complex (adaptive) systems incorporate a number of variables that simultaneously play many different roles in the system’s evolution and following many different laws of behavior. These systems are non-linear and potentially volatile. Their behavior can only be discovered by studying how these elements interact and how the system adapts and changes throughout time.
• Chaotic systems are turbulent complex systems in transition to a different order. The relationships between cause and effect are impossible to determine because they shift constantly and no manageable patterns exist. No medium to long-term prediction whatsoever is possible.

The study of chaotic and complex systems has created a large and growing field of complexity science. While a variety of disciplines have used this analytical and theoretical prism, it has been particularly popular within the natural and, increasingly, social sciences. This trend points to the fact that complexity studies are applicable to all kinds of systems, regardless of their size or nature, like technological (sub-) systems, societal phenomena or the international political system.

Understanding the Behavior of Complex (Adaptive) Systems

One of the fundamental issues of complexity is the nature of order and organization. In complexity theory, the appearance of new order is explained through the concept of self-organization, which connotes a spontaneous formation of a stable, but complex pattern out of seemingly random entities. Self-organization is triggered by internal variation processes, which are called “fluctuations”. Over time, complex systems become increasingly sensitive to internal and external fluctuations – and the more complex a system is, the more sensitive and vulnerable to fluctuations it becomes. At a certain point, these fluctuations pass a critical threshold, also called a bifurcation point, and then, following a transitional stage of chaotic fluctuations, completely reorganize the entire system.

Also, large systems with many components have the tendency to evolve into a poised and highly imbalanced “critical” state where minor disturbances may lead to events, called avalanches. This means that complex systems tend to adapt to, or are themselves on, the edge of chaos and most of the changes take place through catastrophic events rather than by following a smooth gradual path. The new path the system will take cannot be predicted and controlled before it is taken. Due to this emergent behavior, the complex system cannot be understood by reducing it to its parts; moreover, the behavior we are interested in evaporates when we try to reduce the system to a simpler, better-understood one.

Applying complexity to international relations and security

Using the complexity prism can help to understand the dynamics of today’s world which has become increasingly characterized by the non-linearity between cause and effect. Within the last 20 years alone, we have seen how globalization has brought online new relationships, influences, exchanges, and advances. Transformative changes in technology coupled with rampant growth in interconnectivity have created a world where societal and technical networks are continuously multiplying, increasing interactions across geographical space, time, and systems. With this, the role of non-state actors within the international system has grown markedly. This is evident when examining the proliferation and characteristics of dark networks. Here violent non-state actors operate in a complex micro-geography devoid of state boundaries, selling and trading illicit materials such as drugs, guns and humans. Contrary to the traditional state-centric structures, today’s groups operate as small dispersed bodies that could be defined as a complex adaptive systems with little hierarchy and self-organizing tendencies that allow them to rapidly adapt to shifting conditions. Coupled with the structural changes found within non-state groups, the butterfly effect of asymmetric attacks bring to fore the complexity surfacing in the global environment where small events can have broad disproportionate effects.

From a governance perspective, the management of multi-level complexity imparts considerable challenges. Overall, this period has dispersed power, most notably away from nation states. Embedded in a world complete with interdependencies, transnational phenomena and accelerated complexity, nation states have diminished capacity to mobilize and control physical (and virtual) borders, communication and financial systems, and the movement of goods and people. In sum, static, state-centric models of government are poorly equipped to handle this environment as they are predominantly inflexible – hindered by hierarchy and bureaucratic structures that limit the flow of information, engagement of multiple actors, and ability to change and adapt quickly.

In addition, system behavior is most troubling for the field of security if cascades and surprise effects are combined. Cascade effects are those that produce a chain of events that cross geography, time, and systems. Such effects are common in more interconnected, interdependent (referred to as tightly coupled) systems. Surprise effects are unexpected, but also truly unknowable events that arise out of interactions between agents and the negative and positive feedback loops produced through this interaction.

Governing in a complex environment

Rightful questions about the applicability of physics and mathematics to social systems aside, complexity studies has placed within our grasp a set of very powerful intellectual tools and concepts.. New concepts, such as emergence, become conceivable, and new methods, such as nonlinear computer modeling, suggest themselves as fruitful modes of study. This is particularly important when compared to the still dominant ways of seeing and analyzing the international “system”.

Traditional approaches to the study of society based on such concepts as equilibrium, stability, predictability, centralization, and one-way causality are generally at odds with complex, nonlinear systems that most social scientists are interested in. The complexity paradigm focuses the attention on the concept of the inherently unpredictable situation. Rather than understanding this as a call to end all foresight efforts due to their clear limits in a complex world, learning to recognize and appreciate complexity, ambiguity, and uncertainty is the task ahead. This means that analysts need to start focusing on different methods that might work well in situations where the assumption of order and linearity does not hold. The aim should not be to reduce uncertainty, but to accept it for what it is and learn how to manage and govern within it.

This in turn means that we should invest in learning about type of system we are confronted with in a specific situation: ordered, complex, or even chaotic? As well as the interactions between these spaces. Most of the old tools and mindsets are geared towards managing ordered systems: best practices, standard rules of engagements, etc. The complex domain is characterized by inherent, non-reducible ambiguity. Therefore, understanding how to facilitate collaboration and encourage organic management without trying to re-establish hierarchical control becomes crucial. The goal of governance in complex environments is to achieve a balance between two thresholds: the first of control (where there is too little connectivity) and the second of total autonomy (where there is too much connectivity). This space in between is referred to as an opportunity for governance bodies to ‘explore and exploit’.

For example, as mentioned in Dave Snowden’s influential Cynefin framework for leading in complexity, complex contexts require truly interactive – and specifically more democratic and multidirectional – communication in order to generate the most innovative ideas. Also, dissent and diversity should be encouraged, because they advance the emergence of well-forged patterns and ideas. Within limits, barriers can be set to limit or delineate behavior of a system, which can then self-regulate within those boundaries. In addition, so-called “attractors” can be stimulated – they arise when small stimuli and probes resonate with people. As these attractors gain momentum, they provide structure and coherence to the system. As the outcomes are unpredictable in a complex context, policy makers should focus on creating an environment from which positive things can emerge, rather than trying to bring about predetermined results and possibly missing opportunities that arise unexpectedly.

In case you have missed any of our previous content on Future Forecasting and its Challenges, you can catch up here on: The Political Problems of Forecasting Structural Change and Changing International Structures