The following terms refer to the work done over the past three decades by scientists in the fields of physics, chemistry, biology, economics and mathematics. Many of these scientists are associated with the Santa Fe Institute in New Mexico, but also include scientists based in Europe, such as Prigogine, Sengers, Nicolis, Allen and Goodwin. There is no one theory of complexity, but many such theories arising from the various sciences studying complex adaptive systems (CAS).

As many of these concepts are borrowed from the natural sciences, their use within an organisational and management context, can lead to misapplication and misunderstanding. The research project will attempt to clarify the concepts and principles which are common to all complex systems and to explore their application to organisations. A series of seminars on Strategy & Complexity and a Study Group on Organisational Complexity, will be held at the LSE and will endeavour to stimulate the dialogue between academics from a variety of disciplines and business executives, through the discussion of work in progress as well as the application of the principles of complexity in organisations.


There are many definitions of complexity and all are context-dependent. For the purposes of this proposal, complexity is defined as organisational complexity and is associated with the intricate inter-relationships of individuals, of individuals with artefacts (such as IT), and with the effects of inter-actions within the organisation and between organisations and their ‘environment’ which includes related businesses. Complexity arises through connectivity and the inter-relationships of a system’s constituent elements. The way these inter-relations arise, the way they help maintain and create new patterns and structures which enable an organisation to evolve, is not well understood. Complexity in this context is also associated with the characteristics of non-linearity*, self-organisation*, emergent properties*, far-from-equilibrium* operation and sensitivity to initial conditions*.

Complex evolving systems (CES) refers to those systems which are able to learn and which change their internal structure and organisation over time, thus changing the behaviour of individual elements. The term complex adaptive systems is used by the Santa Fe scientists to describe complex systems which adapt through a process of self-organisation* and selection. However, physical, chemical and biological systems are not conscious and do not ‘learn’ in the sense that humans learn. Hence the term complex evolving system [Allen, 1996] is used in this proposal to distinguish human from other complex systems. Both CAS and CES are subject to change through mutation or totally unexpected change, which is then subject to adaptation. Characteristics of complex systems: The study of natural complex systems has shown that all complex systems share certain generic characteristics. Some of these characteristics are included in this terminology and the research project will explore their application to social systems. Such application, however, questions long held assumptions and has profound implications for management, methods of work, the shape of organisations, and the development and use of information technology.

Non-linearity & multiple outcomes

Modelling of aggregate behaviour in organisations is usually based on the assumption that all individuals exhibit average and thus predictable behaviour, when organisations are entities made up of individuals who interact, are mutually inter-dependent and exhibit non-average behaviour. Through multiple inter-actions, organisations are capable of many possible responses; that is, they are complex, unpredictable, non-linear systems, producing multiple outcomes. Yet they are studied as if they were simple, linear systems guaranteed to produce a single, predictable outcome. Another aspect which is often ignored, is that any outcome is influenced by a number of contributing factors. These factor cannot all be taken into account for various reasons; they may not be known, may not be quantifiable or they may be ignored as relatively insignificant, yet these factors may be subject to the phenomenon known as sensitivity to initial conditions*, which could lead to unforeseen and often undesirable consequences.

Sensitivity to initial conditions

When a small change in the initial conditions produces major and unpredictable qualitative changes. Traditional approaches implicitly assume that events occur at an average rate (there are exceptions, and Robust Planning for example does not make that assumption) and that they can be adjusted if they deviate from the desired plan by employing the appropriate adjustment mechanism. But events do not unfold with average regularity and adjustments rarely produce the desired effect. No planning mechanism can take all initial and influencing conditions into account, and at times a small change in the initial conditions produces major and unpredictable qualitative changes. This coupled with positive feedback or increasing returns [Arthur 1990, 1995], makes accurate forecasting and the planning of specific outcomes extremely difficult.

Innovation as exploration of the space of possibilities

Traditional approaches also ignore an organisation’s capacity to learn and change and to maintain diverse and varied strategies, assuming that a single ‘optimum’ strategy is both possible and desirable. The sciences of complexity have shown that for an entity such an organisation to survive and thrive it needs to explore its space of possibilities and to encourage variety. When markets were stable and growth was a constant, single optimum strategies based on extrapolation from historical data, were thought to be feasible. But unstable environments and rapidly changing markets require flexible approaches based on variety. [Ashby, 1956]


Economic models often assume that a state of equilibrium is a desirable condition, but the sciences of complexity show that systems which survive and thrive, do so when they are pushed away from equilibrium, while if they remain at equilibrium they die. When far-from-equilibrium, systems are forced to experiment and explore their space of possibilities and this exploration helps them discover and create new patterns of relationships, different structures and innovative ways of working. [Prigogine, Nicolis] Non-linear dynamics (or chaos theory) may be used to explain the emergence of these new patterns as analogous to the transition phase of bounded instability, between stability and instability which is a state of creativity and innovation. [Gleick 1990, Parker & Stacey 1994] Although non-linear dynamics is an integral part of the theories of complexity it is only one aspect and needs to be balanced by the broader realm of understanding offered by complexity. Analogies based on the ‘edge of chaos’ need to be made applicable to social systems and organisations. One key question which will be addressed in both phases of the research project is the balance between stability and instability necessary to encourage innovation while avoiding both instability and stagnation. At the transition state between stability and instability, order and organisation may arise spontaneously out of disorder through a process of “self-organisation”.


The spontaneous organisation of the system’s elements into coherent new patterns, structures and behaviours. Change in human organisations may be brought about by the spontaneous self-organisation of individuals. These new patterns are not decreed, designed or imposed by any specific individual. They simply happen. They may subsequently dissolve and leave little trace or they may have a longer lasting effect and change the structure of the organisation. In the latter case true evolution has taken place and the internal structure of the organisation has changed. We need to understand how to encourage self-organisation as a means of creating new innovative patterns of behaviour as well as a means of devolving the strategy process throughout the organisation. If the organisation of the future is to work on a different basis, that pattern or shape will need to emerge and evolve from a given set of simple principles. [Allen, Bovaird, Goodwin, Holland, Kauffman, Lane, Nicolis, Varela]

Dissipative structures

Dissipative structures (Prigogine & Stengers 1985) are open systems exchanging energy, matter or information with their environment. In Prigoginian terms, all systems contain subsystems which are continually “fluctuating”. When one or more fluctuations become so powerful, as a result of positive feedback, that they shatter the pre-existing organisation, the system has been forced into a far-from-equilibrium condition and has reached a point of bifurcation. It is inherently impossible to determine in advance which direction change will take. The system may disintegrate into instability or leap to a new level of order or organisation called a “dissipative structure”. It is given that name because it requires more energy (or information) to sustain it than the simpler structure it replaced. In terms of the flow of information, a stable system can be sustained with a sluggish flow, but a much more vigorous and richer flow is necessary for a system operating far-from-equilibrium. If the flow of information becomes too fast, however, then the system may disintegrate.