Warsaw April 2003




WARSAW: 23-26 APRIL 2003


ED1a “Economy & development”: 24 April 2003


Time for a Paradigm Change: Development as a Complex Adaptive System


Dr Samir Rihani


Senior Research Fellow

University of Liverpool, United Kingdom

At the Threshold of a Paradigm Shift

Kuhn (1996: 10) demonstrated that disciplines operate within ‘paradigms’ that define what is and is not considered, implicitly, to be sensible and legitimate at any point in time.

The study and practice of development is no exception. Development is envisioned by tradition as a finite, primarily economic, process that follows an orderly path determined by laws of universal applicability. Some countries, it is asserted, discovered these laws and became fully ‘developed’. Others, at present ‘undeveloped’, are yet to go through the same process, guided and helped by the ‘developed’ countries.


Paradigms go through radical ‘paradigm shifts’ when it becomes clear that current assumptions are unable to cope with significant problems. Development is at present approaching such a transformation. Leading members of the development establishment, such as Joseph Stiglitz, past chief economist at the World Bank, agree with this conclusion. The title he chose for his presentation to the 1998 United Nations Conference on Trade and Development was ‘Towards a New Paradigm for Development’. Rihani (2002) proposed a new paradigm founded on a radically different concept of development, as described below.

Paradigm in Distress  

Development has been the focus of attention for commissions, summits, global agencies, and governments for over fifty years. According to Easterly (2001:33), “Western countries gave $1 Trillion (measured in 1985 dollars) in aid” between 1950 and 1995. There are some successes, but overall the development project, even when measured in narrow economic terms, has failed its main clients. As estimated by the World Bank (Policy and Research Bulletin, April-June 2001), the picture, now and in future, could not be bleaker. Well over one billion people have to live on less than one dollar a day (at 1993 purchasing power parity terms). This sorry picture might not change much by the year 2015; the ‘low case’ puts the figure at over one billion and the ‘base case’ at just under 0.8 billion. The worst regions affected in this way are sub-Saharan Africa and South Asia.

Similarly, the United Nations Development Programme (1999:38) reported that “world inequalities have been rising steadily for almost two centuries”. Specifically, the GDP per capita of the poorest nations hardly changed from its 1820 levels. Moreover, the achievements of the development movement from the 1950s have been marginal at best. This persistent record of failure is an intriguing paradox. Just as significant, experts, institutions and governments were unable to find substitutes for the prescriptions that have repeatedly failed previously. This is another intriguing paradox.

Locked Into an Inappropriate Mindset

The above two-sided paradox is a classic symptom of a paradigm in need of change. Even at this early stage, it is possible to see the absurdity of assuming that all of today’s ‘developed’ countries followed the same path, and the same policies, and that they have now reached an end-state of development. These beliefs are actually held by some to be sensible, such as Fukuyama (1992). Conversely, historical facts seem not to support any of these assumptions, as shown by Chang (2002) for example.

The present paradigm emerged in embryo at the 1944 Bretton Woods conference, but its political economic ideology was spelled out in Truman’s Inaugural Address of 20 January 1949. The doctrine made three assumptions that have endured:

  • Development is equated with economic growth in the main. Economic development must come first; human development would follow.
  • Economic development must be achieved by adherence to strict capitalist ideology, as the leading nations are presumed to have done on their way forward.
  • Economic development requires top-down assistance from and oversight by external (mainly Western) agencies: they know better than local actors.    

Based on the above premise, Rostow (1960), for instance, reduced economic development to a few distinct stages that could be undertaken in a finite period of time, involving ten to fifteen years of aid from abroad. Toye (1987: 11) reported, critically, that development has been commonly seen as a move ‘through a series of stages… derived essentially from the history of Europe, North America, and Japan…’

The paradigm, now known as the Washington consensus, has been elaborated, but not radically changed, over the decades. Hoogvelt (2001:248) described it as a “neo-liberal agenda imposed by the joint IMF stabilization programmes and the World Bank structural adjustment programmes…” Stiglitz (1998) argued that the Washington consensus “took privatisation and trade liberalization as ends in themselves, rather than as means to more sustainable, equitable, and democratic growth.”

The IMF and the World Bank’s structural stabilisation and adjustment programmes provide the most striking examples of an orderly paradigm. Policies were based on a succession of economic growth theories derived exclusively from experience in the USA and Europe. William Easterly (2001: 103), who was a high-ranking economist at the World Bank for eighteen years, traced the history of failure that marked this search for the right theory of universal applicability, as shown in the figure below. In technical terms, the search accorded with an orderly (linear) view of development as a finite economic activity with a clear beginning (when the country is ‘undeveloped’) and an inevitable end (when the country is ‘developed’), and a path common to all nations that leads from start to finish.












Despite changing perceptions in the 1980s and 1990s that recognised development as a ‘multidimensional concept’ (Elliott 1996: 5-7), the current orderly paradigm has remained unchanged. However, doubts about its adequacy have gathered momentum recently (Rihani, 2002). These doubts were given added urgency by the realisation that allied disciplines, such as economics, were being analysed as Complex (nonlinear, less orderly and less predictable) phenomena.

Complex Adaptive Systems: Local Chaos and Global Order

Certain systems are described as Complex because their behaviour is governed by dynamic local interactions between their, typically numerous, components that produce unexpected emergent properties. The expression does not necessarily mean they are complicated, although they do require different methods of analysis and management from those suited to orderly systems. In addition, some are called adaptive because of their ability to evolve over time in response to changing internal and external conditions.

Complex Adaptive Systems present a mix of local chaos (the dynamic interactions of the component parts) and self-organised global order that is not imposed by external forces. Adam Smith’s concept of the invisible hand of the market is an example of self-organisation. Individual transactions present a chaotic picture, but the market shows a recognisable pattern that is reasonably stable.

Boolean networks have been used to study the behaviour of systems. Each network has many internal elements with two possible states: active or inactive, for instance light bulbs that could be turned on or off by some of the other bulbs depending on the connectivity of the network. By altering connectivity and the local rules that dictate how the elements interact patterns of order, self-organised Complexity, or chaos were created and observed (Kauffman 1993: 36).

In an orderly regime there is one unchanging state (the condition of the network at any instant), while an infinite number of states without a recognisable global pattern are involved in a chaotic regime (Kauffman 1993: 174). In self-organised Complexity the system scrolls through a large but finite number of almost similar states (Coveney and Highfield, 1996: 166 and Byrne, 1998: 26). The near-identical states differ but they do so within specific limits. In Complexity parlance they are said to fall into one attractor. The attractor in force at any time gives a Complex system its observable global pattern.

A simple way to envision an attractor is to think of an air-conditioning setup. A preferred room temperature is selected, but in practice the temperature oscillates within an acceptable range. At any given instant the exact temperature of the room defines the state of the system at that point, but all the states (the range around the chosen temperature) define the attractor for that particular air-conditioning arrangement.

Minor variations between the system’s states cause some change but this is normally contained within the prevailing attractor. Despite the frenetic internal activity, outwardly the system seems to be unchanging. Occasionally, however, minor variations could trigger a major shift into a different attractor that presents a new global pattern, but in most situations there is no way of knowing which initial impulse would shunt the system into a new pattern. There is a simple explanation for this feature. Complex Adaptive Systems are associated with positive feedback. Orderly systems, on the other hand, are normally governed by negative feedback that limits, rather than magnifies, disturbances.

Adaptation, Survival and Learning

Evolution is an open-ended process composed of small but effective adaptations by which the system improves its chances of survival. However, the system must be able to detect changes in ambient conditions, including activities by other coevolving systems. Intelligence helps, but is not necessary. Copious internal variety normally ensures that some elements would survive and prosper under the new conditions. In effect, as a Complex Adaptive System continues to evolve it develops what appears to be a key facility ‘for acquiring information about its environment…’ (Gell-Mann 1994: 17).

There is, hence, a cyclical process of survival, adaptation and ‘learning’ that unfolds over long periods of time: an uncertain exploration of possibilities rather than a rush to a chosen destination. As the process unfolds, “the greatest complexity represented has a tendency to grow larger…’ (Gell-Mann 1994: 244).

In summary, therefore, a Complex Adaptive System presents the following features:

  • A large number of interacting elements; human beings in social systems.
  • A tendency for self-organisation when the elements interact at an appropriate level of connectivity and in accordance with suitable local rules.
  • Added layers of elaboration lead to greater Complexity (e.g. the USA has more Complexity than Afghanistan).
  • The current pattern of the system is the product of its history up to that point, but what happens next is less certain.
  • The system evolves through punctuated equilibrium (Coveney 1995: 232); large upheavals separated by long periods of relative global stability.
  • Evidence of positive feedback. A state of criticality enables the system to explore its fitness landscape to the full (Coveney and Highfiled, 1996: 107).

Evidence of Complexity in Development

Support for the contention that social systems behave as Complex Adaptive Systems, rather than orderly systems, has been documented by several authors; including Byrne (1998), Coveney and Highfield (1996), Day (1994), Gell Mann (1994), Jervis (1999) Kauffman (1993 and 1996) Nicolis and Prigogine (1989) and Ormerod (1994 and 1998).

Rihani (2002) focused his arguments on parallels between the way nations evolve and Complex Adaptive Systems. For that purpose, he traced the history of the international political economic theories (mercantilism, liberalism, Marxism, etc.) and analysed topics that are an integral part of the development scene, such as trade, debt and aid. Space does not allow for a detailed exposition of the evidence, but the method of analysis was straightforward. Signs of Complexity (as listed in the previous section) were looked for in the inputs and outputs of the development process. However, the following examples drawn from today’s ‘developed’ and ‘developing’ economies will augment the case for a Complexity view of development.

The ‘developed’ nations present a stable common pattern typified by a pragmatic mix of market economy, social welfare, and liberal democracy. Within the deceptive uniformity of that global pattern there is massive variety and flexibility between and within these nations, which afforded them stability that has overcome many crises and challenges for several centuries. This setup is reminiscent of the healthy variety provided by near-similar states within one stable attractor.

Second, the ‘developed’ countries appreciated at an early stage the value of ‘social capital’. The government of the USA, for example, is ‘the world’s biggest enterprise (Samuelson and Nordhaus, 1995: 299-304). Its high spending on nutrition, health, education, and income protection, is dictated by the wish to endow citizens with the freedom and ability to interact and pursue their varied interests.

Third, as shown below, today’s leading countries followed an evolutionary path based on steady accumulation of modest growth over very long periods. They have been exceedingly rich for a long time and their development is not at an end. Furthermore, their open-ended evolution exhibited signs of punctuated equilibrium and gateway events, such as the Industrial Revolution, and the invention of the steam engine and computers.

Gross Domestic Product/ Head in 1990 international dollars (Chandler, 1997: 6)


























































Fourth, development in these countries stemmed from uncoordinated efforts by individuals and groups concerned mainly with their particular businesses, intellectual pursuits, and hobbies. No one planned the British Empire or the Enlightenment! Significantly, their development was never seen as simply a matter of economics.

Finally, today’s ‘developed’ countries are becoming richer, and the gap between them and poorer nations is inexorably widening. This is precisely what one would expect if the development of nations proceeded along typically Complex Adaptive lines: average Complexity increases and the highest Complexity stands to gain the most.

It is not claimed here that these nations set out to follow a path to development based on Complex Systems theory. Basically, they selected, mainly through trial and error, practices that optimised their performance. Recent discoveries in the field of Complexity provide theoretical, scientific if you wish, explanations of why these particular practices proved to be more effective and sustainable than others.

The key point to note here is that the ‘developed’ countries followed various routes and adopted different policies that were changed pragmatically to suit conditions. Hence, for most of its history “the United States has been a high-tariff nation.” (Samuelson and Nordhaus, 1995: 696). Britain, on the other hand, was by tradition economically liberal, but it went through mercantilist and socialist phases as well. In recent decades, the USA altered course and became the ultimate in capitalist free-market ideology, but that could change again if the need arises. Rigidity was hardly ever part of the recipe for progress.

The Situation Within the ‘Developing’ Countries

It is possible to consider the analogy between development and Complexity from the so-called developing countries’ perspective as well. Assuming that a Complexity view of development were correct, conditions here are in general the exact reverse of what is needed for sustainable evolution to occur. This accords closely with the persistent failure of development efforts. Fundamentally, it would be more accurate to describe these nations as ‘non-developing’ rather than ‘developing’.

Lack of rudimentary levels of democracy and security ensures that individuals are not free to interact. In addition, most ‘developing’ countries do not have a framework of rules and regulations that commands popular support. There is either wasteful chaos; brought about by constantly shifting, unclear, and often conflicting rules, or stifling order; created by arbitrary rigid rules that hinder interactions (De Soto 2000: 15-20, 63). To make matters worse, global institutions, such as the World Bank and the IMF, insist that these countries should conform to rigid global rules and ideologies. Any deviation is interpreted as a challenge to the ‘world order’ that merits a robust response. In effect, poorer and weaker nations are not allowed to explore their fitness landscape effectively.

The factors that enhance individuals’ capability to interact are also in short supply. The priorities were defined at the 1995 World Summit for Social Development as nutrition, primary healthcare, clean water and safe sanitation, basic education, and family planning. UNICEF identified these factors, in The Progress of Nations 1997, as the “foundation for sustainable human development.”

Summers (1999), former chief economist at the World Bank and at the time of speaking deputy secretary at the U. S. Department of the Treasury, said, concerning the failure of some countries to develop, ‘I am convinced that there is a common element of a lack of social connection- a lack of links between people because governments have pre-empted not just all the political and economic space, but also much of the social space.’ The language is different but the meaning in Complexity terms is unmistakeable: progress relies on free interactions between capable people. The key conclusion is also evident: there can be no economic development in the absence of human development.

Not Only What To Do But How

In the main, useful interventions within a paradigm that treats development as a Complex Adaptive process are restricted to those that would enable interactions to proceed in a manner that produces self-organised stable patterns rather than rigid order or chaos. Control involves continual monitoring of outputs and encouragement for the elements to interact in ways that move the system towards desirable ends. Under these circumstances, patience, variety, flexibility and pragmatism are indispensable.

Management of Complex Adaptive Systems is, therefore, a delicate reiterative activity that deals with the systems as a whole and one that accepts progress for what it really is: slow and uncertain evolution. Command-and-control methods and detailed long-term forecasts and elaborate plans, effective for orderly systems, are inappropriate. The same could be said of universal recipes, to fit all nations, concocted by experts thousands of miles away. In particular, top-down management through remote control from abroad is nonsensical for an activity largely driven by local dynamics, affected by positive feedback, and involving vast numbers of individuals and groups.

What Would a Paradigm Shift Mean in Practice?

Reasons for a paradigm shift in development were outlined and an alternative paradigm was described in the previous sections. It would be useful to sketch out a few interrelated consequences that would arise from adoption of such a framework. A list of detailed recommendations that promise better results than those achieved in the past, as is the case in orderly situations, would be inappropriate within a paradigm based on Complexity. Ideas put forward can only be of a general, holistic, nature concentrating in particular on attitudes and practices, as described below. Beyond this, each nation has to chart its own path within the (changing) circumstances that exist at any point.

First, world powers, governments of the ‘developing’ nations, and global institutions would have to accept development as a lengthy, messy and uncertain affair, driven in the main by local actors. It comes mainly in the form of modest local improvements that accumulate over long periods of time. This, apparently innocuous, proposition has far-reaching consequences. For instance, a World Bank housed in a building owned by the US State Department, with a president appointed by the US government, inevitably leads to undue pressure on the developing countries to conform with preset ideas and ideologies. More to the point, a World Bank with 8,000 employees in Washington and only 2,000 in the field is an obsolete model. No wonder the World Bank is accused of being remote and ineffective: it is difficult for it to be anything else.

Second, nations follow different paths defined in part by frozen accidentsinherited from the past, such as location, history, religion, traditions and pure chance. The future, on the other hand, is conditioned by local opportunities and constraints as well as the activities of other coevolving nations. Both are in continual flux. Pragmatism and variety are essential for survival. Regimentation imposed by external or internal forces, such as the World Trade Organisation and local leaders intent on copying ‘the West’, is detrimental to the proper evolution of a nation. Indigenous knowledge, local coping strategies, and domestic actors have to take a clear lead in driving the development process.  

Third, human development: improving standards of health, nutrition, literacy, security, human and property rights, and good governance, could not wait for economic development to take place. The horse must come before the cart. There is basically no choice in the matter. Illiterate, diseased, malnutritioned and disenfranchised people could not take an active part in moving the national economy, or anything else, forward. This is not an easy proposition to put across. Not unexpectedly, it would seem odd that a poor country could find the resources to fund basic programmes. Nonetheless, several authors have already shown this to be feasible, such as Timberlake (1991) and Rihani (2002: 203). Fundamentally, appropriate remedies are not necessarily costly. An example, from WHO (2000: 10) illustrates the concept. It reported, “many deaths of children under 5 years of age could be averted for $10 or less…but the average actual expenditure in poor countries per death prevented …is $50,000 or more.”

Fourth, people, not experts, must come first. This is not a slogan but an essential policy decision dictated by a shift to a Complexity framework for development. WHO (2000: 14) described the difficulties of putting this objective into practice in the context of health, when it criticised the experts ‘for giving too little attention to people’s demand for health care, which is greatly influenced by perceived quality and responsiveness, and instead concentrating almost exclusively on their needs.’ This oversight could lead to a massive difference in projects implemented and the costs incurred (e.g. public health promotion augmented by health workers in mobile clinics as opposed to highly qualified consultants working in remote High-Tec hospitals). Similar views have been expressed about other services, such as the supply of safe water.

Finally, evolutionary change does not promise equality. Average Complexity increases over time, but those with the highest Complexity stand to make the greatest gain. Nations should seek to optimise their performance within their opportunities and constraints without being diverted by how others are doing. International comparative statistics are useful to an extent, but ultimately, monitoring of the performance of a nation from year to year is the most pertinent measure.

Impediments on the Way

The above examples illustrate the kind of transformations that would flow from adoption of the paradigm described in this paper. It is readily evident that the changes are substantial and would be unpalatable to some. Leaving aside parochial sentiments and institutional inertia, other hindrances would affect the speed of the shift already in train. These include, for example, bogus theories and ideologies advanced by powerful institutions, and obsessive focus on globalisation based on liberalisation at any cost.

However, intervention by leading powers and their multinational companies and institutions is the most potent obstacle to development. This is especially troublesome at a time when the USA, as the reigning hegemonic power, is in long-term relative decline (Kennedy, 1987: 514). Hegemonic powers at this stage become unstable and aggressive. The rise of the European Union and China poses a challenge that the USA could not ignore. Sadly, rivalries between the leading economic powers translate themselves into problems for the poorer and weaker countries. This is especially the case for marginalised regions, such as sub-Saharan Africa, that do not fall clearly within the ambit of any of the three ‘regional’ hegemonic powers (Stallings, 1995: 355). The most evident negative impact on development resulting from this factor is incessant strife, civil wars, cross-border conflicts, increasing militarisation and dictatorships. None of these features are conducive to steady evolution based on a multiplicity of free local actors.    


The main purpose of this paper was to put forward a synthesis for the diverse ideas now emerging haphazardly throughout the development industry. Complexity, it was argued, offers a sound basis for seeing the process of development for what it really is. People, as individuals, governments and institutions, are the masters of their own destinies. If they are free and capable to interact within a helpful institutional framework then they will enter the open-ended development process, and vice versa. In short, there are only two types of nations: developing and non-developing.

Efforts in recent years have shifted focus onto poverty, basic needs, institutional reforms, sustainable development, participation, and respect for indigenous knowledge. These trends are now firmly established within the mainstream of development work. Adoption of a Complexity view of development simply advances cogent reasons why these ideas would stand a better chance of success.

Several questions might be asked in response to the ideas advanced here. One concerns the issue of Complexity itself. The development community is already moving along the lines described, why bring Complexity, and a paradigm shift, into the discussion? There is a ready answer to this question. Recent innovations in development were adopted as technical or moral responses to the consequences of past policy failures, but they are not yet integrated within a unified theoretical framework. Complex Systems theory provides a synthesis that would enable a consistent approach to be adopted. This would also furnish a plausible substitute for the ‘science’ of economics that currently underlies most development thinking.

Another question could be framed as follows: everyone knows development is complex so what is new? Complexity as used here denotes a description of a system that has certain distinctive properties. It does not concern features relating to how difficult or complicated development might be. The aim is to define a radically different framework within which development could be pursued more effectively.

The third question might be: the Complexity paradigm seems reasonable, but would those involved in development accept such a radical departure? The answer is definitely yes, but it will take time. The study and practice of development is already undergoing change. As mentioned earlier, the new ideas being tried were arrived at through experience and intuition. However, not unexpectedly perhaps, they sit very comfortably within the paradigm shift proposed in this paper.

It is perhaps useful to end by emphasising the fact that paradigm shifts take a long time, and they are only adopted after much discussion and reflection. This paper is essentially a part of this elaborate process of consideration.


*Arthur, W. B. (1990) ‘Positive Feedback in the Economy’, Scientific American, February: 80-85.

*Arthur, W. B. (1994) Increasing Returns and Path Dependence in the Economy, Ann Arbor: University of Michigan Press.

*Byrne, D. (1998) Complexity Theory and the Social Sciences, London: Routledge.

*Caufield, C. (1996) Masters of Illusion, London: Macmillan.

*Chandler, A. D. (1977) The Visible Hand, Cambridge: Harvard University Press.

*Chang, H. (2002) Kicking Away the Ladder, London: Anthem Press.

*Coveney, P. and R. Highfield (1996) Frontiers of Complexity, London: Faber and Faber.

*Day, R. H. (1994) Complex Economic Dynamics: An Introduction to Dynamical Systems and Market Mechanisms, Massachusetts: MIT Press.

*De Soto, H. (2000) The Mystery of Capital, London: Bantam Press.

*Easterly, W. (2001) The Ellusive Quest for Growth, Cambridge, Massachusetts: MIT Press.

*Elliott, J. A. (1996) An Introduction to Sustainable Development, London: Routledge.

*Fukuyama, F. (1992) The End of History and the Last Man, Harmondsworth: Penguin.

*Gell-Mann, M. (1994) The Quark and the Jaguar: Adventures in the Simple and Complex, Boston: Little, Brown and Company.

*Hoogvelt, A. (2001) Globalization and the Postcolonial World, Basingstoke: Palgrave.

*Jervis, R. (1999) Systems Effects: Complexity in Political and Social Life, Princeton: Princeton University Press.

*Kauffman, S. (1996) At Home in the Universe: the Search for Laws of Complexity, Oxford: Oxford University Press.

*Kauffman, S. (1993) The Origins of Order, Oxford: Oxford University press.

*Kennedy, P. (1989) The Rise and Fall of Great Powers, London: Vintage.

*Kuhn, T. S. (1996) The Structure of Scientific Revolutions, Chicago and London: University of Chicago.

*Nicolis, G. and Prigogine I. (1989) Exploring Complexity: An Introduction, New York: Freeman.

*Ormerod, P. (1994) The Death of Economics, London: Faber and Faber.

*Ormerod, P. (1998) Butterfly Economics, London: Faber and Faber.

*Rihani, S. (2002) Complex Systems Theory and Development Practice, London: Zed Books.

*Rostow, W. (1960) The Stages of Economic Growth, Cambridge: Cambridge University Press.

*Samuelson, P. A. and W. D. Nordhaus (1995) Economics, Fifteenth International Edition, New York: McGraw-Hill, Inc.

*Stallings, B. (1995) Global Change, Regional Response: the New International Context of Development, Cambridge: Cambridge University Press.

*Stiglitz, J. (1998) Towards a New Paradigm for Development: Strategies, Policies, and Processes, Prebisch Lecture, UNCTAD, 19 October.

*Summers, L. H. (1999) ‘Summers questions development policy research’, Policy and Research Bulletin, April-June, Washington D C: World Bank.

*Timberlake, L. (1991) Africa in Crisis, London: Earthscan.

*Toye, J (1987) Dilemmas of Development, Oxford: Blackwell.

*UNDP (1999) Human Development Report 1999, New York: Oxford University Press.

*WHO (2000) The World Health Report 2000, Geneva: World Health Organisation.