Complexity Theory and Health: Viruses, Bat Attractors, Tractors, Cannibalism and Mad Cows

Written in 1999 as Chapter 3. Complexity and Human Health: Across the Health Hierarchy

Nick Higginbotham, Glenn Albrecht, Linda Connor (2001) Health Social Science: A Complexity and Transdisciplinary Perspective. Oxford University Press.

Draft 8/8/99           (Glenn Albrecht)



3.1.  Complexity theory: Ideas that transcend disciplines

Once we understand that human health is the outcome of complex processes that operate within and across physical, psychological, social and ecological systems, our way of thinking about health problems needs to reflect this interrelatedness. Detailed knowledge of complex, adaptive, dynamic systems cannot come from within the narrow focus of traditional disciplines. In order to understand complexity, we need the ‘nondiscipline’ of being able to work across disciplines (Nelson, quoted in Chu and Simpson, 1994, p.21).

Health problems exist in what we later call ‘transdisciplinary space’ (see Chapter 4). Within this space are interactions between humans, and between humans and their environment, that range from deeply subjective and intersubjective to the impact of global climate change. New developments that have taken place in complexity theory over the last decade have produced a theoretical framework that can attempt to explain the full scope of these interactions. Complexity theory is now applied across a great variety of disciplines to biophysical systems, human social systems and expressions of human culture and creativity in an attempt to provide greater insight into both the nature of disorder and the emergence of new forms of order (e.g., Hayles 1991; Goerner, 1994; Guastello, 1995; Masterpasqua and Perna  1997; Dean 1997; Ormerod 1998; Byrne, 1998).

We now describe the theoretical features of complexity theory in order to examine the way complexity theory can be used to understand and inform problems of human health[1]. In order to understand the impact and novelty of complexity theory, we first examine the earlier theories that attempted to describe complex systems and their characteristics.


3.2.1 Newtonian Reductionism and Determinism

The image of the world that Newton (1642-1727) and his followers described was essentially machine-like (mechanistic) and deterministic (rule governed). In a Newtonian world, a finite number of rules or laws govern the motion of material bodies and these laws are universal. All other aspects of the ‘world-machine’ can similarly be determined and a complete account of the workings of parts of the machine is thought possible. In mechanistically unified structures, such as those Newton described, it is entirely valid to break down the structure of large things into smaller things in order to study them, because a complete account of the parts will constitute an account of the whole. An important outcome of this perspective, which is characterised as ‘reductionism’ in the philosophy of science is the idea that ultimately all things can be explained through mechanistic and deterministic laws. The use of the machine metaphor implies that regularity and order is the norm for complex systems and that instability or unpredictable change is the exception.

Newton’s laws of motion worked well at one level of human experience. Newtonian laws have been successfully applied in producing technologies from sewing machines to jet aircraft. The relationships between the variables in mechanistically unified systems are described as linear. Within linear systems parts are related to each other in ways that do not change the nature of the whole. We can describe what will happen in linear systems and plot causally related variables on a graph. The relationship between cause and effect will be proportional, in that in normal circumstances a large causal force will produce a large effect. With detailed knowledge of initial states and the environment within which they operate, linear systems are capable of precise mathematical description and offer high degrees of predictability. For example, a large input of energy into an engine will produce a proportionately large output of work or horsepower.

However, at more theoretical levels Newton’s laws were counter-intuitive. In a Newtonian universe, if we had absolute knowledge of all laws and rules governing matter, then we could with equal plausibility reconstruct the past or predict the future. Time sequences within linear systems are in principle reversible. As Paul Davies explains:

Newtonian time derives from a very basic property of the laws of motion: they are reversible. That is, the laws do not distinguish ‘time forwards’ from ‘time backwards’; the arrow of time can point either way. From the standpoint of these laws, a movie played in reverse would be a perfectly acceptable sequence of events (Davies  1989, p.14).

Nineteenth century physicists such as Clausius (1865) and Boltzmann (1872) challenged the idea of time’s reversibility when they developed the idea that the universe is tending toward thermodynamic equilibrium. The first and second laws of thermodynamics suggest that:

  1. the energy of the universe is constant, and
  2. any closed system will tend spontaneously to a state of maximum possible disorder (entropy), that is the state of thermal equilibrium.

Entropy is a measure of ‘the degree of disorder’ of a system (Pais 1991, p.81). The idea that entropy increases in a closed system suggests an ‘inevitable element of irreversibility in mechanical systems in the course of time’ (Pais1991, p.82). The implications of this view, however, were even more alarming than the theoretical Newtonian reversal of temporal order. The physicist Von Helmholtz  (1854) pointed out that if entropy always increases, then the universe is moving toward its own destruction. Paul Davies gives a graphic summary of this gradual running-down of the universe:

The remorseless rise in entropy that accompanies any natural process could only lead in the end, said Helmholtz, to the cessation of all interesting activity throughout the universe, as the entire cosmos slides irreversibly into a state of thermodynamic equilibrium. Every day the universe depletes its stock of available potent energy, dissipating it into useless waste heat. This inexorable squandering of a finite resource implies that the universe is slowly but surely dying, choking in its own entropy (Davies 1989, p.19).

This depressing view of our planet’s fate was not seriously challenged until the  emergence of complexity theory in the 1970s. Complexity theory changed the nineteenth century view of entropy as a destructive force to one involved in the creation of order.

The way physicists and others have come to this view is through the observation that in certain circumstances (described as ‘far-from-equilibrium’), matter displays a tendency to self-organize in new and often unpredictable ways.

3.2.2. complex dynamic systems

A central claim of complexity theory is that spontaneous order and organization can come from what appears to be flux and disorder in natural systems. The most recent form of complexity theory suggests that complex, open systems have distinctive properties that cannot be reduced to constituent parts. The physicist Paul Davies explains:

In the traditional approach one regards complex systems as complicated collections of simple systems. That is, complex or irregular systems are in principle analysable into their simple constituents, and the behavior of the whole is believed to be reducible to the behavior of the constituent parts. The new approach treats complex or irregular systems as primary in their own right. They simply cannot be ‘chopped up’ into lots of simple bits and still retain their distinctive qualities (Davies 1989, p.22).

Complexity theory advances perspectives on systems that run directly counter to all forms of reductionism and determinism. One strand of complexity theory, ‘chaos theory’, suggests that small, random changes to the parts of a system can give rise to large-scale or global changes to that system. Another strand of complexity thinking emphasises the spontaneous generation of an emergent order in complex systems.  (see Lewin 1993 p.12 for further explanation of the two strands).  It is this emergent order that provides an alternative perspective on systems within the Newtonian paradigm. The production of greater complexity, new types of diversity and novel spatial and temporal patterns and sequences in a variety of physical contexts suggests system behaviour more like that with which we are familiar in the biological realm. Hence, terms like ‘adaptive’, ‘evolutionary’ and ‘self-organisation’ can be applied to the understanding of matter and physical systems in general. The new view of systems suggests that there is a class of systems that can be described as complex adaptive systems in that they actively change and evolve over time. ‘Adaptive’ is the term used to describe the way complex systems change in response to changes in their environment. They are open to modify their behaviour in the face of pressure and perturbations from the wider environment within which they exist.

3.2.3 Self-Organisation and dissipative structures

The innovation of contemporary complexity theory resides in the claim that all the sciences, ranging from mathematics to biology and the social sciences, can now support the general hypothesis about evolution in life from the simple to the complex. The idea of increasing levels of complexity in an evolutionary view of the world is complemented by claims in complexity theory that all types of open systems have the potential to evolve of their own accord toward greater complexity. A constant theme that emerges in the literature is that new forms of complexity are achieved through spontaneous self-organisation. This means that there are no agents acting from outside the system that can account for the re-arrangement of the internal structure of a system to create greater order. The method by which this complexity evolves or self-organises is substantially explained through the idea of dissipative structures, or ‘new dynamic states of matter’ (Prigogine and Stengers 1994, p. 143).

Pioneering theorists of complexity, Prigogine and Stengers, describe this fundamental concept:

We now know that far from equilibrium, new types of structures may originate spontaneously. In far-from-equilibrium conditions we may have transformation from disorder, from thermal chaos, into order. New dynamic states of matter may originate, states that reflect the interaction of a given system with its surroundings. We have called these new structures dissipative structures to emphasise the constructive role of dissipative processes in their formation (Prigogine and Stengers 1984, p.12).          Dissipative Structures

           The most important discovery associated with complexity theory is that the second law of thermodynamics, the entropy law, while still having universal relevance, can be negated at local levels. By ‘exporting entropy into its environment’ (Davies 1989, p.85), a dissipative structure can maintain its structural integrity and evade an increase in entropy. In order to maintain or increase coherence and complexity, a dissipative structure must exist in relationship with an open system where exchange of energy can take place. A dissipative structure creates an internal order that ‘is far more efficient in utilizing energy for organization and maintenance than the background system within which the primary flux occurs’ (Dyke in Weber et al.1988, p.360).

The first law of thermodynamics states that the amount of energy in a closed system is constant. However, the second law also suggests that in a closed system (and open systems in a state distant from equilibrium) the amount of useable energy (ranging from high to low) available for any process will always decrease. Within such systems there is an unequal distribution of energy available for use, ranging from high level energy to useless waste heat. Left to its own devices, the system will inevitably move from a far-from-equilibrium situation to one that is termed ‘equilibrium’. There is movement from high level energy to low level energy, much like that involved in weather systems when the wind is caused by the movement of air from high to low pressure systems. We can term the difference between high and low pressure a pressure gradient and this term is also applied to the differences between highly usable energy in a non –equilibrium system and energy which is highly degraded and in a system close to equilibrium.

At local levels dissipative structures in non-living and living systems seem to defy the movement towards equilibrium. Indeed, they increase complexity and diversity at local levels while entropy continues to increase at the global level. A reformulated second law of thermodynamics makes this apparent contradiction understandable.  As Schneider and Kay explain, a restated second law suggests that ‘… the more a system is moved from equilibrium, the more sophisticated are its mechanisms for resisting being moved from equilibrium’ (Schneider and Kay 1995, p. 165). As a result, any system self-organisation that assists in the process of eliminating the applied gradient will be an ‘expected response of a system’ (Schneider and Kay 1995, p. 165).

We can anticipate then that open complex systems should adapt and self-organise to counteract the influence of an imposed energy gradient. The earth as a whole could be considered just such an open system with the sun applying a large energy gradient upon it. Schneider and Kay argue that this energy challenge to the state of equilibrium is a major reason why the earth has its great diversity of life present in self-organising ecosystems. The diversity of life and ecosystem complexity is mainly a response to the imperative to dissipate energy in the gradient from high quality (useable) to lower quality (difficult to use) energy.

As evidence supporting such an hypothesis, Schneider and Kay point out that at the earth’s equator, where 5/6 of the earth’s solar radiation occurs, the greatest species diversity is present. They argue that:

… food chains are based on photosynthetic fixed material and further dissipate these gradients by making more highly ordered structures. Thus we would expect more species diversity to occur where there is more available exergy (energy available to perform useful work) (Schneider and Kay 1995, p.168).

Dissipative processes have been studied in many types of physical and biological systems and they are now being identified in human social systems. Examples of dissipative structures in physical systems include cyclones, twisters, autocatalytic chemical reactions and convection cells. All types of biological systems can be considered to be dissipative structures.

In physical systems, a now well-known illustration of a dissipative process is the Belousov-Zhabatinski (BZ) autocatalytic chemical reaction. The experiment involves a number of chemical reagents that react with each other in a system kept open with pumps. When chemical throughput is slow there appears to be an even distribution of chemicals as indicated by a lack of dominance of any one color from dyes that are injected to indicate excess ions of a particular chemical. The system could be described as ‘close to equilibrium’. However, when the flow rate increases and the system moves far-from-equilibrium, the reaction changes such that it manifests a blue color, indicating that a particular ion has increased its concentration. It stays blue for a short time then changes to red, indicating that another ion has come to predominate. The system then oscillates from blue to red with regularity so precise that Prigogine calls the reaction a ‘chemical clock’. Barton explains that this regularity:

… occurs because the chemical processes that result in the red state coming into existence become linked to the processes resulting in the blue state. When this happens, the two states codetermine one another in a cyclical, nonlinear fashion (Barton 1994, p.7).

The BZ reaction and other similar observations of emergent self-organisation such as the BJnard convection cell (Sandar and Abrams 1999, p.75-6) reveal in experimental contexts forms of self-organising behaviour in matter. These sorts of self-organising behaviour were previously thought only to operate at more complex levels in living systems. Complexity theory has developed to the point where theorists have speculated on the possibility that there might be common properties and dynamic processes operating across all types of complex natural systems. The complexity theorist Stuart Kauffman argues for the concept of a background natural order with which all other ordering processes interact. Kauffman suggests:

…contrary to our deepest intuitions, massively disordered systems can spontaneously ‘crystallize’ a very high degree of order. Much of the order we see in organisms may be the direct result not of natural selection but of the natural order selection was privileged to act on (Kauffman 1993, p.173).

The problem for students of complexity, despite the idea that all complex systems might be subject to common physical laws, is that such systems can change in ways that are unpredictable when in a state distant from equilibrium. There is potential for either greater complexity or greater disorder; unless one has complete knowledge of all initial and boundary conditions, uncertainty will dominate. In simple systems where boundary conditions can be accurately specified, we know that dissipative structures will become more complex as they adapt and change over time. This insight offers some hope that complexity can be clearly understood. Davies also suggests that we can develop ‘idealised complex or irregular systems’ (Davies 1989, p.23) that will assist in approximating the features of real systems. The concept of an ‘attractor‘ within complex systems provides further opportunity to offer detailed explanation of the order that arises out of dynamically changing systems.

3.2.3. Attractors and The Edge of Chaos

Complexity theorists have developed the concept of a point called ‘the edge of chaos’ where the forces of order and disorder come into competition. As a system reaches its most chaotic state, it also has the ability to self-organise into a new form of order. Such creative self-organisation is possible because of the richness of the interactions that occur with maximum potential for the exchange and processing of information. It is within this region of tension between order and disorder that complexity emerges. There are forces that can pull the system one way or the other and theorists have termed these forces ‘attractors’. Attractors are ‘states to which the system eventually settles’ (Lewin 1993, p.20). Brian Goodwin uses the example of a marble entering a bowl and moving in the bowl until it comes to rest. He describes the point of stability as:

… an attractor, and the bowl defines the basin of attraction for all the trajectories that converge on the stable point. For certain classes of dynamical system an energy function can be defined and then the stable state corresponds to a minimum of energy which is like the bottom of the bowl where the marble comes to rest. (Goodwin 1994; 157)

At least four mathematically defined types of attractors are used to explain pattern emergence in dynamic physical systems (Barton 1994, p.7.), ( Goodwin (1995 p.169-73) and Kauffman (1993) apply the concept of ‘dynamical attractors’ to explain pattern emergence in natural systems (i.e., organs, organisms, species and ecosystems). What attractors do is act to pull a complex dynamic system out of instability or chaos into order or vice versa. In general terms, as Goodwin describes it:

For complex non-linear dynamic systems with rich networks of interacting elements, there is an attractor that lies between a region of chaotic behaviour and one that is “frozen” in the ordered regime, with little spontaneous activity. Then any such system, be it a developing organism, a brain, an insect colony, or an ecosystem will tend to settle dynamically at the edge of chaos. If it moves into the chaotic regime it will come out again on its own accord; and if it strays too far into the ordered regime it will tend to “melt” back into dynamic fluidity where there is rich but labile order, one that is inherently unstable and open to change (Goodwin 1995, p.169).

Different attractors can influence a complex system in different ways causing new forms of order/disorder to emerge. Major system shocks or ‘perturbations’ might also act to push or pull the system toward a new attractor and the establishment of a new pattern.

Per Bak has expanded on the idea of ‘the edge of chaos’ to describe what he calls ‘self-organised criticality’ (Bak and Chen 1991) where a complex system ‘shows waves of change and upheaval on all scales as if the size of the changes follows a power law’ (in Waldrop 1992, p.308). A ‘power law’ can be described as the likelihood of a major change in a dynamic system such as a pile of sand that has new grains being constantly added to it. As summarised by Waldrop, such a system can exhibit all kinds of behaviours:

But the steadily drizzling sand triggers cascades of all sizes- a fact that manifests itself mathematically as the avalanches’ “power law” behavior: the average frequency of a given size of avalanche is inversely proportional to some power of its size (Waldrop 1992, p. 305).

Thus self-organised criticality can be understood as a ‘general type of attractor’ (Goodwin 1994, p. 175) that assists in the understanding of systems that exhibit patterned behaviour. Cascades of change in all types of dynamic systems might follow power laws and such laws are in principle discoverable.

Goodwin has applied this type of thinking to explain the evolution of the eye, a structure that has evolved independently in about forty different evolutionary lines. Following patterns laid out in embryonic development and other forms of cellular growth, the recognisable form of an eye is the result of pattern emergence within a sea of possibilities. As Goodwin explains it, ‘ … there’s a large attractor in morphogenic space that results in a functional visual system (in Lewin 1993, p. 40).

In a sense, the eye is the result of an attractor for light and sight. Rather than a situation of infinite flexibility and possibilities, certain ‘laws of form’ might operate in all kinds of contexts where complexity exists and emergence is possible.

Some attractors might exert very powerful forces on a system and maintain a pattern of order over very long periods of time. This is clearly the case for the attractor for the eye and other anatomical features that are ubiquitous in nature. Others might have only short-term influence and a perturbation will shift the system towards some new attractor or leave the system in a state of chaos for some time. The rapid evolution of micro-organisms will come under the influence of a dynamic field or ‘landscape’ containing attractors that rise and ebb in importance.          Attractors and Social Systems

Human social systems have evolved within the context of self-organised ecological and physical systems. So it should come as no surprise that the idea of an attractor generating order can be applied to the complex dynamic systems that are human societies. Social attractors act as catalysts in the production of regularities or patterns that can be discovered and studied in society. Such attractors of social order can range from charismatic individuals to ideological systems. Rather than a situation of ‘anything goes’, it is possible that the number of states to which [any] dynamical system settles will be determined by a finite number of attractors.

At different scales of human social systems different attractors define the range of possible configurations of order. Some cities, for example, under the influence of a wide range of influences might be conceived of as attractors in the context of the global economic order, pulling in enormous amounts of energy, information and technology and growing as a consequence while others degenerate and lose resources. Within cities, different attractors such as work and sleep can produce very different patterns of order such as the congestion of traffic during peak hours producing a uniform distribution of vehicles and the random distribution of traffic late at night and very early morning (see Sandar and Abrams 1999, p. 132-3).

At smaller scales of human interaction, quite different sets of attractors might be found that limit the number of possible patterns of interaction. Something as simple as the quality of coffee at a particular coffee shop might explain the popularity of a particular establishment in a sea of competing businesses. The promise of complexity theory is that it invites those who are attempting to understand the dynamics of complex systems to look not only for obvious attractors but also broad ordering principles within which attractors and the things that perturbate systems and attractors are constrained.          Summary: The Features of Complexity

From this brief summary of complex adaptive systems, we can distil a number of interrelated characteristics of complex systems.

  • Local interaction can produce global order and global order can affect local behavior.
  • The interaction of local and global levels of complex systems determines their properties. Such interaction might be subject to ordering influences that are internal to the system or may be universal features of all types of complex adaptive systems
  • Interactive causal relationships exist within and between entities and are at their richest at the edge of chaos, the point between order and disorder.
  • Complex systems can self-organise and evolve towards states of increased complexity.
  • Complex adaptive systems can form patterns and follow predictable paths of development. The identification of attractors or states to which a system finally settles, is one clue as to why certain patterns (order) not others are created.
  • The properties of complex adaptive systems cannot be reduced to their constituent parts.
  • There is order in what appears to be chaotic; order arises from fluctuations or perturbations within a system.

3.3.  Complexity and human health

Most health problems exist within the full spectrum of levels of complexity from the sub-molecular to the planetary. Integrating common conceptual frameworks will be useful in gaining some leverage over these domains. However, when outcomes follow the interaction among many variables, as when major changes to systems precipitate from small perturbations and global order affects local behavior, even detailed knowledge of elements of the system may ‘not lead to useful knowledge about the behavior of the system as a whole’ (Barton, 1994, p.7).

A useful point of entry into the study of complexity in human health might be the identification of health related attractors that influence health outcomes at the various scales of relevance from the micro-biological to the transnational and global. Health attractors are health or disease states towards which the (health) system eventually settles and they range from genetic and somatic pulls to social and political pushes that shape the patterns of health. These pulls and pushes can either create order or disorder in complex systems within which health is embedded. One way of looking at these opposing forces is to conceptualise health as existing in dialectical tension between the most basic attractors of life (dissipative structures) and death (entropy).

At micro-biological levels the major factors that influence the micro-organisms that affect human health are evolutionary and ecological. The HIV retrovirus evolves under the selection pressure exerted by the human host’s immune system. Many strains of enteric bacteria are now evolving under the influence of the routine use of antibiotics in animal livestock management. This practice means that selection pressures in bacteria now include the expanded environments of medical (humans) and veterinary (animals) use of antibiotics. In such expanded environments it should come as no surprise that resistant strains of bacteria are now emerging and are major threats to public health (Saradamna, Higginbotham and Nichter 2000).

Attractors at this level include normal natural selection pressures (fitness) between host and organism and new attractors such as the industrialisation of animal husbandry and the use of antibiotic drugs to maximise animal health, productivity and hence, profitability. A beef cattle feedlot might be conceptualised as a new type of attractor creating new types of interactions between elements such as farmers, cattle, antibiotics, protein supplements and consumers that were formerly related in different ways. The feedlot defines a new space of possibilities, called a fitness landscape (see Kauffman in Lewin 1993 p. 57) for the interaction of all these variables, one quite different for the space of possibilities for free range beef.

Other new attractors in the fitness landscape for pathogenic micro-organisms are created by new environments produced by human technologies that are designed to maximise human comfort. These new artificial fitness landscapes include air conditioning cooling towers and mass population water supply systems.

At the level of the human body, one attractor that has been the object of considerable study is the attractor for effective beating of the heart. It seems that when the heart experiences a heart attack or cardiac arrest, this is not caused simply by the heart ceasing to beat in rhythm. As Firth explains:

In the physiological sphere, one would suppose the human heart to have an oscillatory attractor. In fact there are indications that the healthy heartbeat is actually slightly irregular, indeed chaotic. Be that as it may, it must have a fixed point attractor (cardiac arrest) into which it can be “kicked” from the normal state by an electrical or other shock. With luck or skill, or both, it can sometimes be kicked back again: resuscitation. (Firth 1991, p.1567)

Goodwin (1994, p.60) suggests that ventricular fibrillation (uncoordinated and ineffective contractions) in the heart muscle can be caused by the presence of ‘infarcts’ or small patches of damaged tissue. These infarcts act as an attractor for the waves of heart contraction, breaking the normal rhythm and creating a new pattern (ventricular fibrillation) that is unable to sustain pumping functions compatible with maintenance of life. The action of defibrillation machines is to shock or perturbate the muscle of the heart to leave the infarct attractor and to resume normal rhythm (around its chaotic attractor) with normal pumping activity.

At the level of human social systems we can identify attractors that exert their influence on health status. Such attractors range from social and cultural pressures on diet and lifestyle to the influence multinational corporations in the medico-industrial complex exert on the availability of drugs worldwide. Byrne (1998, p.108-21) argues that inequalities in health within the populations of cities can be understood in relation to attractor forms based on socio-economic criteria such as inequality and the life chances (e.g., poverty) that are structurally connected to them. Byrne argues for this position with respect to tuberculosis (TB):

We can say here that an unequal world city will have a TB problem, but that it is possible for us to recast the city as more equal and in that attractor state there will not be a TB problem (Byrne 1998, p.119).

At the level of small scale human interaction, factors such as human values and their expression in policy and management will act as endpoint determinants or attractors. Miller et al. (1998) have used attractors in this sense to understand the dynamics of a social example of a complex adaptive system, a group of doctors operating a primary care clinical practice. They suggest that:

Practice attractors can include a particular income goal, a specific understanding of patient care success, meeting patient and community expectations, or a particular practice vision. Attractors can also be understood as the motivators and values of the practice (Miller et al. 1998, p.371).

Dean (1997) argues that ‘human nature’ itself might be subject to deep underlying structures that can be understood by complexity theory. Social and health problems such as drug and alcohol dependence (intoxication) might be explained by factors such as ‘security’ and ‘status’ that are ‘forms of primary selective categories’ or attractors for the evolution of behaviour that might result in alcohol abuse. Although negative in their impact on human health, such adaptive behaviour can nevertheless ‘fit the circumstances of a person’s existence’ (Dean 1997, p.153) and prove to be beneficial in a social or collective context. Similar conclusions are given in our case study of heart disease among men in the Australian coalfields (see Chapter 5).

Derrickson-Kossmann and Drinkard (1997) use the concept of an attractor to understand the dynamics of a psychological condition — dissociative identity disorder (DID). They note how therapists use cognitive interventions to perturbate the attractor for DID into an adaptive state more likely to increase internal communication and coherence.

The factors that influence health, taken as a whole, form nested sets of attractors (after Tarlov (1996) in Byrne 1998, p. 108). These nested attractors operate in ways that we suggested in Chapter 2 using the health hierarchy diagrams. Byrne suggests that that there are three main domains where health attractors operate:

The individual attractors are lifestyles – the product of the interaction of constraint and volition. The social attractors are the grand social forms which pattern the possibilities of lifestyles. Even these may be embedded within a wider Gaian biosphere level of possibilities … which is under serious perturbative assault from human industrial production and resource consumption. (Byrne 1998, p.116)

We can attempt to map the array of attractors for factors influencing human health across the full range of scales in the following way.

Table 3.1. The Full Scale of Human Health Attractors

Scale Attractors
Microscopic and other types of pathogens Selection pressures, fitness, genetic norms, favourable anthropogenic environmental changes (cooling towers, feedlots, (‘wet’ markets), mass intercontinental transport) vectors.
Organs Genetic norms, heart infarcts and fibrillation, tumours, health and death.
Body Size, Shape and Fitness Cultural norms for body shape (perfectibility), fast food and diet (McAttractors), sedentary lifestyle (computers), surgery.
Individual, and Small Group  Behaviour Security, status, goals and values, expectations, levels and types of communication, professions, sub-cultures, health gurus (Pritikin?)
Societal and Institutional Public and private health care and insurance systems, public policy, advertising, globalisation, mass media, transnational organisations (TNCs, UN, WTO)
Planetary Systems Gaian feedback loops (global warming, El Nino-Southern Oscillation) with, eg., shifts in vector populations  (mosquitos)

With an awareness of the main attractors that influence human health, careful attention can also be focused on specific attractors that might operate in unique settings. Miller et al. (1998) identify specific attractors that shape the form of the system that is a family practice primary care centre. A particular centre has a vision of what its modus operandi or internal model will be.  The core functions of the practice are created within the context of broad philosophies and policies of physician style, income generation, patient care, prevention services delivery and organisational operations (Miller et al. 1998, p. 371). The various elements within the primary practice attempt to achieve these core functions by engaging in strategic actions to achieve the desired goal or end state. The end states or points are described as attractors and include an identified income level for its staff, patient care success rates and the meeting of community expectations. After identifying these small-scale social system attractors, the next step in their analysis is to identify ways to influence the attractors so as to achieve desirable outcomes. They suggest three ways to change attractors:

  1. joiningenhancing existing attractors using the known internal models;
  2. transformingchanging an attractor or creating a new one;
  3. learning – increasing awareness of attractors and internal models. (Miller et al. 1998, p.373)

There may be many more ways of influencing the range of attractors that operate at different scales. Some large-scale attractors might not be capable of being influenced by players operating in the small scale (i.e., local level) system. The primary practice is nested within larger scale systems such as the medico-industrial complex, the health insurance industry, the national health policy and the biophysical environment. At these levels, different attractors exert their power and different ways of influencing them must be attempted. At the level of micro-organisms and the larger biophysical environment within which they evolve, pathogens such as the influenza virus can affect general practice in ways that simply cannot be influenced by agents in small scale social groups.

3.3.1. Complexity and Health: Case Studies and Examples

Discoveries about the complexity of health issues come as no surprise to those prepared to think through the total picture in a systematic way. For instance, some disturbances that humans are enacting on long stable cultural and ecological systems are manifested as epidemics of disease. An example of a small change to a system leading to a large effect is provided by Desowitz’s case study of ‘tractor-induced’ Japanese encephalitis in Northern Thailand.          From Buffalo to Tractor  Attractors

The traditional practice in Northern Thailand was to plough the rice paddies with water buffaloes and to keep pigs for food and market. Since Culex tritaeniorhynchus [the mosquito carrying the encephalitis virus] prefers steak to pork, the water buffaloes acted as a ‘blotter’, limiting viral transmission. Then, heeding the call of progress, the farmers of the region replaced their buffaloes with tractors. As the buffalo population declined, the mosquito(s) turned their attention to the pig and to man. Many pigs now became infected; the virus multiplied in the pigs; more and more mosquitos became infected; and, in turn, so did more and more humans. Hundreds died, and many of these victims were children (Desowitz 1981, p.21-2).

Desowitz’s observations show the characteristic in complex systems of interactive causality, in this case among people, the mosquito, the virus, domestic animals, and introduced technology. As the landscape was changed by eliminating the buffalo attractor for the virus, the new tractor attractor created the circumstances where the system changed to favour humans as an end point for viral transmission.The next example provides further evidence of how perturbations to eco-systems can end up having unexpected health outcomes.          War and Weather Attractors

During the Korean war a mysterious disease emerged which affected thousands of United Nations troops, killing many. It was later discovered that the cause of the disease, known as Korean Haemorrhagic Fever, was a virus transmitted to humans via rodent urine and excrement (Hantaan or Seoul virus) as noted in Chapter One. The virus had remained dormant perhaps for centuries until heavy machinery and digging by soldiers disturbed the environment and exposed them to the pathogen through contact with soil and dust. Perturbation to a long undisturbed natural system created the conditions for transmission of a new viral disease. Forty years later, the Seoul virus, possibly transmitted by adventurous rats aboard ships, is now endemic in major cities in the US and is being detected in patients on the East Coast (McAuliffe 1990, Garrett 1994)

Coincidentally, in 1992-1993 a mystery virus struck at least eighteen people living on or near a Navajo reservation on the New Mexico-Arizona border, fourteen of whom died (MMWR 1993). The pathogen was discovered to be a member of a previously unknown type of hantavirus, called sin nombre, and is a worrying example of what are termed ’emerging pathogens’ (Le Guenno 1995, p.30-1). What appear to be novel viruses may have existed for millions of years but have only come to light with contemporary environmental disturbances.

Le Guenno (1995, p.32) asserts that the primary cause of most haemorrhagic fever outbreaks is ‘ecological disruption’ which brings humans into contact with animal vectors (e.g., Korean War). However, perturbations to ecosystems also result from natural disturbances. The emergence of sin nombre in the USA was the result of unusually heavy rain and snow in 1993 in the mountains and deserts of New Mexico, Nevada and Colorado.  The principal host of sin nombre in this area is the deer mouse which lives on pine kernels. The high humidity resulted in an abundant crop which created conditions suitable for an explosion in the deer-mouse population which in turn coincided with the epidemic of sin nombre.          Bat Attractors

The emergence of new diseases and unexpected outbreaks of ‘old’ diseases are becoming increasingly common in an era of increasing air travel, the development of tropical mega‑cities and the encroachment of humans into previously untouched natural environments (McAuliffe1990). The outbreak of Ebola in Africa has been linked to the emergence of the virus from an as yet unknown reservoir that had previously been left undisturbed by the actions of humans (see Le Guenno 1995, p.34). Deadly viruses that have fruit bats or flying foxes as their host have recently been identified. Nipah killed over 100 people in Malaysia in February 1999 and caused the death and forced culling of over one million pigs, while Hendra claimed the lives of both people and horses in Australia in the mid-1990s. Humans are coming into more regular contact with these animals as their habitat contracts under the pressure of agriculture and forestry.  Carers, or people who look after sick or injured bats, are most likely to be the ones most likely to come into contact with the new viruses, especially if they are bitten by their ‘patients’. That these viruses are capable of crossing species barriers makes their threat to humans and livestock very serious.          Factory Farms and Feedlots as Attractors

A social factor with a notable impact on human health is factory farming.  Many of our food sources are now the result of what is widely known as ‘agribusiness’.  Agribusiness treats food production much like any other form of production in a capitalist society where costs are minimised and maximum profit is sought. In the attempt to maximise output, producers have used a variety of ‘techniques’ to enhance what could be produced by conventional farming methods.  Genetic engineering, the use of antibiotics to control infection in intensive animal husbandry, and chemicals and drugs to control parasites and pests, are all well-established practices in agribusiness. So too are growth hormones and animal protein supplements used to minimise food costs in raising vegetarian animals for human consumption.

The practice of protein supplementation has been implicated in the health problem known as Mad Cow Disease (MCD) in cattle and Creutzfeldt-Jacob Disease (CJD) in humans. In the mid 1980s British veterinarians identified a new disease in cattle which was named Bovine Spongiform Encephalopathy (BSE). The symptoms of this disease of the brain include nervous and unpredictable behaviour and extreme lack of coordination. The brain develops sponge-like holes as the disease progresses towards total degeneration of the victim’s nervous system. In the last few weeks of the disease cattle are so badly affected by the destruction of their brain tissue that they appear ‘mad’, hence the popular name of the condition.

What was a serious disease in cattle, with the potential to cost the British beef and dairy industry many millions of dollars in lost production, became an issue of international concern when it was proposed in the early 1990s that BSE was capable of transmission to humans through the food chain. In other words, it is now thought that BSE and CJD are one and the same. The reasoning behind this proposed causal link was based on evidence from diverse sources.

Current research suggests the best explanation of the origins of CJD is the ‘prion hypothesis’. According to Prusiner,

…the infective agents that cause some of the degenerative diseases of the central nervous systems in animals, and, more rarely, in humans might consist of protein and nothing else (Prusiner 1995, p.30).

Prusiner has called these abnormal infective proteins, ‘prions’, and has claimed that they are capable of self-replication within the brain tissue of mammals.  Prions are very robust and cannot be destroyed by normal cooking heat nor by a wide battery of chemical agents. The causal link from animals to humans is still considered tentative, however, it involves the transmission of prions from one species to another.

One of the earliest known suspected prion diseases is ‘scrapie’ in sheep and goats.  Scrapie causes its victims to lose coordination and to itch; hence the appropriateness of the name as infected animals scrape their wool/hair to relieve the itch. It was common practice in the UK in the 1970s and 80s to use parts of sheep not used for human consumption as animal feed in intensive beef/veal/milk production. The bone, offal and flesh were ground up to produce a food supplement that represented the cheapest way to maximise protein input to commercial herds.  It is thought that scrapie infected sheep could have passed their infective prion to cattle which in turn manifested itself as BSE. Humans who then consumed infected beef and dairy products could contract CJD, the human equivalent of BSE.

The epidemiology of BSE suggested that this link was plausible. Since animal-based food supplements were banned for ruminants (i.e., grazing animals like sheep and cattle) in 1988 there has been a steady decline in the number of cases of BSE in the UK. According to the Institute of Science and Technology (UK):

…the prohibition of the ruminant material from the animal feed, and the successive year-by-year reductions in confirmed new cases over recent years seem to demonstrate that the animal feed was a key factor.

The evidence for the direct link to humans is less compelling since it is possible that CJD has a very long incubation period; perhaps up to 30 years. However, as Lacy (1996) and others have pointed out, there is at least one historical precedent of spongiform encephalopathy in humans that should have provided vital clues as to what was happening with CJD. Lacy points out that ‘Kuru’, a fatal brain disease of the Fore people in Papua New Guinea has a remarkable similarity with CJD (see Chapter One).  Kuru was a major health problem for the Fore people until the cause of the disease was discovered to be the eating of human flesh and in particular, human brains.

Historically, CJD has a very low incidence. According to Prusiner, about one in a million suffer from it and it usually strikes at around 60 years of age (Prusiner 1995, p.31). The numbers of cases of CJD in the UK is relatively small (80 dead or dying by September 2000). A number of victims were dairy farmers who have had direct contact with BSE in their herds.  A disturbing trend is the relatively young age (under thirty) of the some of the victims given the long incubation history of prions.

A further cause for concern is the discovery that cows can pass BSE to their calves so the practice of destroying infected adult animals will not be sufficient to ensure that BSE is eradicated from the food chain.  Moreover, because prions are so robust and cannot be destroyed by conventional means (including radiation) the fact that they can be transferred from mother to calf means that the infectious prions must be able to pass from the central nervous system to the blood. Infectious prions in the blood means that all beef products will be potentially infectious to humans and this includes products such as gelatine which are used in an enormous variety of other food products. The Nobel Prize for medicine was awarded to Stanley Prusiner in 1997 and this recognition of his work suggests that there is now greater scientific acceptance that ‘prions must be added to the list of well known infectious agents including bacteria, viruses, fungi and parasites’ (Cooke 1998, p.26).

3.4.            CONCLUSION

The possibility that humans have created, through their agricultural practices (feeding animal protein to vegetarians for economic reasons), a health epidemic that will reach far into the future is a perfect illustration of how a seemingly small change in a complex system can produce a very large effect. Understanding CJD has required insights from anthropological history, epidemiology, cellular biology, agricultural science and neurology. Such complexity of modern health problems has led to the suggestion that:

In the end, it is likely that the classification of diseases into infectious, environmental, psychosomatic, auto‑immune, genetic and degenerative will prove to be applicable only to a sample of cases where one factor overwhelms all others. The more accurate viewpoint will encompass full complexity of this network of factors that leads to recognisable disease (Levins et al. 1994, p.55).

Chapters in Part 2 demonstrate a number of the principles of complexity theory that we have covered in this chapter. In Chapter 5, for example, we draw on complexity theory to analysis an ‘epidemic’ of heart disease among people in a coalmining community in New South Wales, Australia. Similar illustrations of complexity processes are highlighted in the ensuing chapters on the dynamics of the global pharmaceutical industry, infection control in hospitals, and HIV/AIDS in Africa. Our aim is to stimulate researchers to extend transdisciplinary thinking to other complex health problems where the dynamic systems initiating or exacerbating evolving disease patterns remain to be illuminated.  However, before exploring these health problem case studies, we turn in Chapter 4 to our basic framework for conceptualising health and taking action to improve it—transdisciplinary thinking.


Bak, P and Chen, K. Self-organised criticality. Scientific American 1991; 264:26-33.

Barton, S.  Chaos, self-organisation and psychology.  American Psychologist, 1994:49;5-14

Byrne, D. Complexity Theory and the Social Sciences: An Introduction, London, Routledge, 1998.

Chu, C.M., and Simpson, R.  Ecological Public Health: From Vision to Practice. Nathan: Institute of Applied Environmental Research, 1994.

Cooke, J. Cannibals, Cows and the CJD Catastrophe. Sydney, Random House, 1998.

Davies, P. The Cosmic Blueprint. London: Unwin, 1989.

Dean, A. Chaos and Intoxication: Complexity and adaptation in the structure of human nature. London, Routledge, 1997

Derrickson-Kossmann, D. and Drinkard, L. Dissociative Disorders in Chaos and Complexity, in Masterpasquia, F and Perna, P.A., (eds) The Psychological Meaning of Chaos, Washington DC, American Psychological Association, 1997.

Desowitz, R.S.  New Guinea Tapeworms and Jewish Grandmothers: Tales of Parasites and People. New York: W.W. Norton & Company, 1981.

Dyke, C.  Cities as dissipative structures. In Entropy, Information, and Evolution: New Perspectives on Physical and Biological Evolution, ed. B.H. Weber et al. Cambridge, Massachusetts: The MIT Press, 1988.

Firth, W.J. Chaos – predicting the unpredictable. British Medical Journal, Vol. 303, 1991, pp. 1565 – 1568.

Goodwin, B. How the leopard changed its spots: The evolution of complexity. London: Phoenix Giants, 1995.

Goerner, S.J.  Chaos and the Evolving Ecological Universe.  Camberwell: Gordon and Breach, 1994.

Greene, J.C.  Qualitative Program Evaluation: Practice and Promise. In: Handbook of Qualitative Research, ed. N.K. Denzin and Y.S. Lincoln. London: Sage Publications, 1994.

Guastello, S.J. Chaos, Catastrophe, and Human Affairs: Applications of Nonlinear Dynamics to Work, Organizations, and Social Evolution. Mahwah, New Jersey: Lawrence Erlbaum Associates, Publishers, 1995.

Hayles, K.N. Chaos and Order: Complex Dynamics in Literature and Science. Chicago, The University of Chicago Press, 1991.

Hegel, G.W.F. The Philosophy of Right. trans. T.M. Knox, London: Oxford University Press, 1976.

Institute of Food Science and Technology (UK).  Bovine Spongiform Encephalopathy (BSE): Part 1/2, IFST: Current Hot Topics, 7 June 1999

Institute of Food Science and Technology (UK).  Bovine Spongiform Encephalopathy: Part 1/3. IFST: Current Hot Topics, 18 July 1996.

Jantsch, E.  Technological Planning and Social Futures. London: Cassell, 1972.

Kauffman, S.A.  The Origins of Order: Self-Organisation and Selection in Evolution. New York: Oxford University Press, 1993.

Lacy, R. How Now Mad Cow? Viva guides, 1996.

Le Guenno, B.  Emerging Viruses. Scientific American, 1995 October:30-37.

Levins, R.T. et al. The emergence of new diseases.  American Scientist 1994:82;52-60.

Levy, S.B. Antibiotic Availability and Use. Journal of Clinical Epidemiology, 1991:44  supp.II;83s-87s.

Lewin, R. Complexity: Life on the Edge of Chaos.  London: Phoenix, 1993.

McAuliffe, K. The Killing Fields: Latter-day Plagues.  Omni, May 1990.

Miller, L. et al. Understanding Change in Primary Care Using Complexity Theory, in The Journal of Family Practice, Vol. 46. No. 5 (May) 1998, 369-376.

Mortality and Morbidity Weekly Report.  Update: Hantavirus Disease – Southwestern United States, Atlanta, Ga: Center for Disease Control, 1993 July:30;570.

Ormerod, P. Butterfly Economics, Faber and Faber, 1998.

Pais, A.  Niels Bohr’s times, In: Physics, Philosophy and Polity. Oxford: Clarendon Press, 1991.

Prigogine, I., and Stengers, I. Order Out of Chaos: Man’s New Dialogue with Nature. New York: Bantam Books, 1984.

Prusiner, S.B. The Prion Diseases. Scientific American, 1995 January:30;37.

Reice, S.R.  Nonequilibrium determinants of biological community structure. American Scientist, 1994:82;424:435.

Sandar, Z. and Abrams, I. Introducing Chaos. Cambridge, Icon Books, 1999.

Schneider, E.D. and Kay, J.J. Order From Disorder: the thermodynamics of complexity in biology, in Murphy, M.P. and O’Neill (eds) What is Life? The Next Fifty Years. Cambridge, Cambridge University Press, 1995.

Waldrop, M.M. Complexity: The Emerging Science at the Edge of Order and Chaos. London, Penguin, 1994.

Weber, B.H., Depew, D., and Smith, J., eds.  Entropy, Information, and Evolution: New Perspectives on Physical and Biological Evolution. Cambridge, Massachusetts: The MIT Press, 1988

[1]             We make no new claims about Complexity Theory, our intention is to summarize the main developments that we believe are applicable to a transdisciplinary view of human health.