I WHAT IS COMPLEXITY? AN INTRODUCTION FOR EDUCATORS Sara Heinrich and Ella Jamsin, Ellen MacArthur Foundation HISTORICAL FOUNDATIONS The historical roots of complexity science are numerous and draw notably from thermodynamics and the theory of evolution. To mention a few: A range of related schools of thought emerged in the early 20th century, such as Smut s Holism, Bertalanffy s General Systems Theory and Ashby s Cybernetics. In natural sciences, Ilya Prigogine played a key role in studying complex systems in chemistry in the 1950s, uncovering key insights into path dependency and irreversibility. Around the same time, Jay Forrester pioneered system thinking in management, and created system dynamics, a methodology to visualise and analyse the behaviours of complex systems. His team at MIT was later joined by Donella Meadows, member of the Club of Rome and of the team that produced the global computer model supporting The Limits to Growth. In 1984, a group of physicists created the Santa Fe Institute, an independent theoretical research institute dedicated to the multidisciplinary study of complex adaptive systems. It is now recognised as a leading institution in complexity science research.
TO DIVE DEEPER, EXPLORE THE WORK OF SOME OF THE FOLLOWING AUTHORS: Jan Smuts (1870-1950) Holism and evolution Ludwig von Bertalanffy (1901-1972) General systems theory William Ross Ashby (1903-1972) Cybernetics Alan Turing (1912-1954) Morphogenesis (theory of how biological growth occurs) Ilya Prigogine (1917-2003) System boundaries, open systems, thermodynamics Jay Forrester (1918-2016) Organisational learning Elinor Ostrom (1933-2012) Political economy and tragedy of the commons Donella Meadows (1941-2001) Thinking in systems Brian Arthur (1946-) Complexity economics Peter Senge (1947-) Organisational management Marten Scheffer (1958-) Ecological resilience and tipping points CONCEPT AND RELEVANCE Most real-world systems are complex. They cannot be understood by analysing their separate parts, often called agents, because these agents are strongly linked by feedback relationships: the agents behaviour influences the whole and the whole influences the behaviour of agents. As a consequence, you can t accurately predict the future state of the system: a very small change in initial conditions can have surprisingly large effects. The weather is a typical example of a complex system. Complex systems are often described as adaptive (and known as complex adaptive systems ) because they can adapt to changes in their environment. Learning is a particular case of adaptation, particularly present in systems of animals, where the experience of a previous change in conditions is stored in the system s memory and influences future behaviours. Complexity has become a highly interdisciplinary topic today, building bridges across for example biology, physics and social sciences. It relies on computer modelling and machine learning, and finds applications in education and management. This reflects the fact that most real-world systems are complex and adaptive, and so increasingly are our technologies. Through the study of flocks of birds, social media and artificial intelligence, to name but a few, complexity researchers have uncovered a number of principles that can help understand the dynamics of all complex systems (see box on the next page). All organisations, including schools and workplaces, are complex adaptive systems of people. Their level of (de) centralisation of decision making, learning capacity, nature of hierarchy and self-organisation, and feedback mechanisms all influence their dynamics and success. The history of an organisation shapes its future development or, in other words, the evolution of an organisation is path dependent. The economy is another complex adaptive system, strongly coupled
with ecological systems. This fact is taken into account for example in the model of circular economy, which emphasises the importance of feedback loops and draws from the notion of resilience the ability of the system to maintain its core function in the presence of changes in its environment. In a circular economy stocks and flows of resources, such as money, materials, information and energy, are acknowledged to interact with each other. Designing a product or service to fit into such an economy demands considering its interactions with economic and ecological systems along its entire lifecycle. Likewise, any organisation active in the transition to a circular economy needs to consider its interactions with the wider system and pay attention to emergent behaviours system level effects that cannot simply be explained or predicted from the actions of individuals and organisations. The science of complex systems can also shed light on how systems change and therefore provide critical insights on how to create better economic, environmental and social outcomes. Change does not always take place gradually, but sometimes through a critical transition after the system reaches a tipping point. This also brings in the concept of resilience, but this time as something we might want to break in order to move to a different, more effective model. We can accelerate transitions by creating the right conditions - the rules of the game - in which an adaptive system can spontaneously evolve towards a more positive outcome. We might for example leverage the high (and increasing) level of connectedness between people in the modern world to spread new ideas and practices. Developing better models of the economy and how it changes, sharpening our ability to think in systems, and leveraging concepts of complexity in the way we work are just a few of the practical implications of complex systems for individuals and organisations. A SUMMARY OF COMPLEX SYSTEMS PRINCIPLES 1 Basic principles A system is more than the sum of its parts. Many of the interconnections in systems operate through the flow of information. A system s structure influences its behaviour. Stocks and flows A stock is the memory of the history of changing flows within the system. A stock can be increased by decreasing its outflow rate or increasing its inflow rate. Stocks allow inflows and outflows to be de-coupled and independent as they act as delays, buffers and shock absorbers in systems. Feedback loops A feedback loop is a chain of causal connections that runs from a stock through a set of decisions, rules or actions dependent on the level of that stock to a flow, which once it changes will alter the level of that stock. 1 This list is an adapted version of an overview provided in Donella H. Meadows, 2008, Thinking in Systems A primer (White River Junction: Chelsea Green), pp. 188
. Balancing feedback loops push a system towards an equilibrium or goal and are therefore both sources of stability and resistance to change (for example, a price increase causes a reduction in demand, which in turn causes a decrease in price). Reinforcing feedback loops are self-enhancing and lead to exponential growth or runaway collapse over time (for example, technological innovation provides the means to develop further technologies). The information delivered by a feedback loop even nonphysical feedback can affect only future behaviour: it can t deliver a signal fast enough to correct the behaviour that led to the current feedback. Systems with similar feedback structures produce similar dynamic behaviours. Dominance, delays, and oscillations A system often exhibits complex behaviour as the relative strengths of its feedback loops shift, causing first one loop and then another to dominate. A delay in a balancing feedback loop makes a system likely to oscillate. Changing the length of a delay may drive a large change in the behaviour of a system. Constraints In physical, growing systems, there must be at least one reinforcing loop driving the growth and at least one balancing loop constraining it, because no system can grow forever in a finite environment. Non-renewable resources are stock-limited while renewable resources are flow-limited. Resilience, tipping points and selforganisation The resilience of a system describes the ability of a system to persist and maintain its core function and/or purpose in the presence of disturbances, stress or other changes in its environment. 2 There are always limits to resilience. 3 Complex systems can transition from one phase to another suddenly and unexpectedly after reaching a tipping point. In the presence of strong reinforcing feedback loops, a tipping point can occur when the resilience of the system is low. Complex adaptive systems often have the property of selforganisation, i.e. the ability to structure and re-structure themselves, to learn, diversify, and increase their complexity. Path dependency The state of a complex system at a point in time depends on the sequence of events and decisions that preceded that point. Source of surprises Many relationships in complex systems are nonlinear, meaning that a small change in the system can lead to disproportionate effects. 2 Complexity Explorer glossary https:// www.complexityexplorer.org/explore/glossary 3 See for example the concept of planetary boundaries, i.e. thresholds beyond which the structure of the planetary ecosystem could tip into a new regime, developed by Johan Rockström and 28 other scientists in a 2009 article Planetary Boundaries: Exploring the Safe Operating Space for Humanity. To learn more about it, Johan Rockström TED talk is a good starting point: http://www.ted.com/talks/ johan_rockstrom_let_the_environment_guide_ our_development?language=en
. When there are long delays in feedback loops, foresight is essential. Rational optimising decisions made by each actor in a system may lead to suboptimal results for the system as a whole. Mindsets and models Everything we think we know about the world and its subsystems is a model. Strictly speaking there are no separate systems the universe is a continuum so where to draw boundaries depends on the purpose of the discussion and therefore what we want to model. Our models of complex systems can have strong congruence with their true nature, but fall far short of representing them fully. Our mental models underpin and drive our beliefs and actions. 4 The usefulness of a model depends not on whether it makes good predictions (since no such certainty can be obtained), but on whether it exhibits realistic patterns of behaviour. INTERCULTURAL LINKS Early Western philosophers and Eastern philosophical traditions developed worldviews in the 6th century that are almost identical to complexity, focusing on flow, interdependency and the emergence of structures and patterns. Why then is the world today mostly governed by a mechanistic worldview? While these concepts continued to be important to Eastern philosophies, in the West a transition started with the dialogues of Plato and Socrates, who introduced discussion and reasoning the dialectic method and decided that form does not emerge but is governed by rules. Such ideas led, a thousand years later, to Newton s physics, a remarkable scientific breakthrough that nonetheless sowed the seeds of modern worldviews in which the natural and social worlds are seen as measurable and predictable. Many modern scientific disciplines developed since are reductionist: they attempt to describe systems in terms of their constituent parts. While this method has given rise to very significant scientific progress and new discoveries, when it comes to complex systems, such an approach may underplay the higher-level phenomena emerging from the interactions between the parts. This way of thinking culminated in the Industrial Revolution where machines - reductionist by design - drove unprecedented economic growth and development. This reinforced the mechanistic worldview and its underlying assumption of control: we could predict what would come out of the man-made system, provided we guaranteed consistency of feedstock. 5 Thinking of processes and systems surrounding us as mechanisms with cogs and wheels and simple causalities proved useful and, as machines entered everyone s lives, so did the mechanistic mindset and linear thought processes. Techniques however continued evolving and the advent of digital technology and computing power 4 George Lakoff s work is a good starting point to further explore this concept and includes the books Don t think of an Elephant and The Metaphors we live by. 5 Ellen MacArthur, Only a circular economy will lead to prosperity for all, http:// circulatenews.org/2016/04/only-a-circulareconomy-will-lead-to-prosperity-for-all/
. allowed models to be developed that better reflect the real world and its complexity. An illustration of that evolution is agent-based modelling, which takes into account feedback between the whole and its parts and can demonstrate emerging phenomena. These new tools have boosted the development of complexity science. In many areas of work, such as education, management, and the environment, thought leaders are increasingly recognising the complex adaptive nature of real-world systems and are therefore using a systems thinking approach to study and influence them. As an example, today s dominant economic thinking is the legacy of a reductionist model developed in the 19th century in an attempt to mimic the physical sciences. This approach enabled great progress in economic theory but led to the development of models that fail to reflect key macroeconomic patterns, most notably financial crises. Complexity economics, a movement in its infancy, is trying to address this shortcoming by applying the framework of complexity to the economy, thereby seeing it not as a system in equilibrium but as one in motion, perpetually computing itself - perpetually constructing itself anew. 6 REFERENCES AND FURTHER READING The following sources were consulted during the preparation of this paper. They also provide good references to learn more about complexity. De Rosnay, J. 1979. The Macroscope. (New York: Harper & Row) Meadows, Donella H. 2008. Thinking in Systems A primer (White River Junction: Chelsea Green) Mitchell, M. 2011. Complexity: a Guided Tour (Oxford University Press) Arthur, W.B. 2013. Complexity Economics: A Different Framework for Economic thought (Santa Fe Institute) Colander, D. and Kupers, R. 2014. Complexity and the Art of Public Policy (Princeton: Princeton University Press) Webster, K. 2015. A Wealth of Flows (Isle of Wight: Ellen MacArthur Foundation Publishing) Boulton, J. Allen, P. and Bowman, C. 2015. Embracing Complexity Strategic Perspectives for an Age of Turbulence (Oxford: Oxford University Press) Complexity Explorer. Various courses and resources, including Introduction to Complexity by Melanie Mitchell and a glossary of complexity terms, are available on the website complexityexplorer.org developed by the Santa Fe Institute. 6 Arthur, W.B. 2013. Complexity Economics: A Different Framework for Economic thought (Santa Fe Institute)