Georg Ivanovas From Autism to Humanism - systems theory in medicine

4. Systemic Basics

previous -- home -- content -- next

4.7 Systems theory

Systemic concepts are slowly on advance in basic research (European Science Foundation 2007). Also Science, Nature and other important scientific magazine address the subject of ‘systemic biology’ ever more. Scientists understand that linear concepts will not solve their problems as they are confronted with an exponential growth of data (Pennisi 2003). Variables are so numerous that it became somehow difficult to reproduce experiments in the ‘same way’ (Szalay/Gray 2006) – a crucial precondition for a reductionist research. Reductionism in its old sense seems no longer to be possible. “Biology today is at a crossroads. The molecular paradigm, which so successfully guided the discipline throughout most of the 20th century, is no longer a reliable guide. Its vision of biology now realized, the molecular paradigm has run its course. Biology, therefore, has a choice to make, between the comfortable path of continuing to follow molecular biology's lead or the more invigorating one of seeking a new and inspiring vision of the living world, one that addresses the major problems in biology that 20th century biology, molecular biology, could not handle and, so, avoided. The former course, though highly productive, is certain to turn biology into an engineering discipline. The latter holds the promise of making biology an even more fundamental science, one that, along with physics, probes and defines the nature of reality. This is a choice between a biology that solely does society's bidding and a biology that is society's teacher“ (Woese 2004).


Ever more scientists try to adopt the concepts of systems science. But this turns out to be rather difficult, especially in the sciences of the living. Systemic biology became “a mutated soup of artificial life, computatinal biology and compuational chemistry with a biz of mathematics, physics and computer science thrown in. Because it is so broad and has few recognized boundaries and plenty of funding, it is attractive to anyone who has ever thought about life and has some relevant technical expertise” (Werner 2007).


General Systems Theory (GST) has been called a ‘dynamic field’ (Jackson et al 2000a). This is a nice way to say that GST is chaotically interdisciplinary. Systems conferences are characterized by an extreme methodological pluralism (Jackson et al 2000a: 28). Although “Systems Science try to find homomorphism among different phenomena or objects, using a more abstract language that would enable the description of homologous phenomena and the unification of science and other disciplines” (Paritsis, 2000a: 55), GST has difficulties to define its paradigm, if there is any. A major problem is even to define what a system is. There are a lot of definitions, but no one seems really to satisfy strict critics (Guberman 2002).

One definition of systems talks of the scientific exploration of ‘wholes’ (Bertalanffy 1968: xx), or of a science of ‘wholeness’” (Bertalanffy: 37). However, this word has been well misused in medicine. It is connected to the unbearable concept of a ‘holistic medicine’ where nobody knows what this might mean. ‘Holistic therapists’ claim to treat the whole person. In my reading this is always true, because always a whole person is treated. But some find it more holistic to talk about family problems than about the gall bladder. On the other side there is no holism as every approach concentrates on a certain field of observation. E. g. computer-tomography of the whole person, coming into fashion in America (Illes et al 2004), is – in a certain way – holistic, but this is probably not meant with wholeness.

Sometimes it is said that “the whole is more than the sum of parts” (Bertalanffy: 55). But may be it is more correct to say that the whole is something different from its parts (von Foerster 1976).


There have been definitions like “complexes of elements standing in interaction” (Bertalanffy: 33), or “systems of elements in mutual interaction” (Bertalanffy: 45), or “a set of members, of their properties, of their relations and of the emerging properties of the system” (Paritsis 2000b: 178). But such expressions are quite vague. Of course there have been more refined attempts to define systems:

(Daellenbach cited by Jackson et al 2000a: 17)

It seems that a system is what an author likes to be a system. In a discussion von Foerster was asked whether a table could be seen as a system. He answered: “Yes, but it is a boring one” (unpublished interview with Monika Bröcker). I prefer to call GST a think tank in the investigation of complex phenomena.


Medicine is in the advantageous position to have a quite clearly defined system: the human (chap. 4.8).


GST was a result of European thinking. It was especially predominant in the German speaking part (Austria and Germany), whereas the Anglo-Saxon world was more concerned with cybernetics. The main representatives of these two different movements were Bertalanffy for GST and Ashby for cybernetics (Bertalanffy: 94). Cybernetics is more analytical, concentrates on control, whereas GST is more concerned with processes and changes. However, cybernetics and GST mingled and today nobody would claim to follow either of them.

The main focus of both, systems theory and cybernetics, is the investigation of organization. Characteristics as “growth, differentiation, order, dominance, control, competition etc.” cannot be described with analytic methods (Bertalanffy: 47). GST describes the ‘logic of the process’. It is concerned with the operation, neither with the operator nor with the operand (chap. 4.2),

Descriptions and control of organization have been the domain of physiology from the beginning. But this has been restricted mainly to organs (the heart) and/or simple functions as respiration. In a more complex sense the concept of the milieu intérieur by Bernard in the 19th century, Cannon’s (1871-1945) homeostasis in the first half of the 20th century and later Seyle’s stress theory were systemic approaches. But it was Bertalanffy (1901-1972) who developed a more general concept with an according algebra, investigating systems in equilibrium. Prigogine (1917 - 2002) further evolved these ideas in investigating systems far from the equilibrium.


Although systems scientists try to describe the function of systems by algorithms, this becomes more and more difficult as the complexity of the defined system rises. For example, there are quite a lot of mathematical models for the development at the stock markets. But a random model based on chaotic choices shows the best results (Ball 2003). The reason is either that market traders are mindless, or that mathematical models are not able to describe human behaviour adequately. After the events in the first decade of the second millennium the most probable answer is: both is true. In any case, to describe complex systems as the human or the stock markets verbal models are necessary. Pure mathematical models are not enough.


Strictness and mathematical accuracy is normally found in systems without environment, that is, in system which are two valued and true. But these systems have no meaning (chap. 3.5). Such systems are closed to matter, energy and information.

But, „the organism is not a closed, but an open system. We term a system “closed” if no material enters or leaves it; it is called “open” if there is import and export of material. There is, therefore, a fundamental contrast between chemical equilibria and the metabolizing organisms. The organism is not a static system closed to the outside and always containing the identical components; it is an open system in a (quasi-)steady state, maintained constant in its mass relations in a continuous change of component material and energies, in which material continually enters from, and leaves into, the outside environment” (Bertalanffy: 121).

One possibility to cope with the complexity of the living is the probabilistic, the statistical approach. But as seen before, this does not describe a semantic pattern (chap. 2.1.d). It is no serious solution for complexity.


Complexity was a central issue of cybernetics from its beginning. Pioneering was Ashby’s ‘law of requisite variety’ (Ashby, 1965: 206-213). It says that in order to control a system and to make it responsible to environmental fluctuations, the controller must command as much variety as the system itself exhibits.

That is, „if a machine has 20 ways of breaking down, we have to respond to 20 different ways to control the machine…If I am a competent photographer, and possess a decent camera, I should possess sufficient variety in terms of distance and exposure to always get my subjects clear. But what if we are faced with systems exhibiting apparently massive variety. How can we cope with this? The answer is that we must either reduce the variety of the system we are confronting (variety reduction) or increase our own variety (variety amplification). The variety of the system confronted must be reduced and/or increased, and this must be done in a way appropriate to the particular system being considered and its goals” (Jackson et al 2000b: 33)

One option is variety reduction. In psychosis for example, variety is reduced by a strict pattern of diagnosis and therapy. That is, the variety of the therapist is reduced. The patient’s variety is reduced by space restriction and drug therapy. Some argue that this might contribute to chronification (Podvoll: 61-68, 125-127, 150-152).

Although medicine tries to reduce variety, systemic models support that the controller of a system has to increase his own variety in order to reduce the variety of the controlled system. This led to the famous statement: “Only variety can destroy variety” (Ashby, 1965: 207), or von Foerster ‘ethical imperative’: “Act always so as to increase the number of choices” (von Foerster/Bröcker 2002: 15-16). For medicine this would imply to develop a methodological pluralism which is the main request of this thesis.

GST provides the appropriate methodology to handle variety either in the individual, in a sub-system or in a whole population. Unfortunately more complex systems are rarely investigated with such a proper methodology. Paritsis puts it that way: “The first is that there are many models that are using different methods of description about different or the same processes and systems. The second is that there are few models that are dealing with global properties of intelligent or behavioral systems. The third is that are even less models for presenting global or general properties for systems that behave, are intelligent and interact with their environment” (Paritsis 2000a: 55)


Some basic systemic principles are:

Struggle between parts (Bertalanffy: 66): A system consists of parts in interaction. The principles of a local activity bringing forth patterns and forms of higher organisation is the domain of emergence-research (chap. 4.10) and can be modelled with cellular automata.

Centralisation: In living systems the outcome is characterized by an ongoing centralisation (Maturana/Varela 1998). Although the individual parts or compartments might work nearly machine like, they are submitted to an overall regulation. This overall regulation is relatively independent from the initial conditions of its parts (Bertalanffy: 68-69).

Circadian rhythms are produced on a cellular level by clock genes (Gelder et al 2003). They work like trivial machines (chap. 4.5). They are integrated on the level of individual organs such as the heart, the liver or the kidney which have their own circadian rhythm (Yamazaki et al 2000, Storch et al 2002). The organ level is integrated in a time centre in the brain, the hypothalamic suprachiasmatic nucleus which determines physiology and behaviour (Yamaguchi et al 2003). This centre also processes inputs from other parts of the central nervous system, through timing of sleep and wakefulness (Deboer et al 2003), or through light exposure (Iglesia et al 2004). In the forebrain there is another centre producing a circadian rhythm dependent on food intake (Dudley et al 2003). All these different rhythms form the ‘biological clock’ or better ‘circadian temporal program’ (Green/Menaker 2003). There seems to be no overall centre to decide. Inner time is an emergent phenomenon of distributed control, as so many biological phenomena. The inner clock arises through a struggle between individual trivial clocks which are centralized. In such centralized systems the prediction of the behaviour of the system is difficult or impossible when only the parts are seen. Small energetical changes on one level might be amplified “causing a considerable change in the total system” (Bertalanffy: 71). But local changes might also have no effect onto the overall behaviour.

Achieving goals/Self-interest (Paritsis 2000a: 56): The prediction of the behaviour of a system is hardly possible by the observation of its parts. It is, however, to a certain extend possible when its propose, its goal is understood. This approach is called teleological. Teleological thinking has long been seen as anthropomorphic. But it cannot be avoided when complex phenomena have to be described. Von Foerster demonstrates this with a simple example: It is nearly impossible to find suitable models and equations to describe all actions of binding a shoelace. But with a teleological phrase (“this is done in order to bind a shoe”) the process is reduced and makes sense (von Foerster/Bröcker 2002: 31). Similar is the situation in biological research. “Gene regulation, intracellular signalling pathways, metabolic networks, developmental programs—the current information deluge is revealing these systems to be so complex that molecular biologists are forced to wrestle with an overtly teleological question: What purpose does all this complexity serve?” (Lander 2004).

Systems theory says that systems try to achieve a goal using its resources in order to do so. This sounds somehow metaphysical, at first. Nevertheless, teleological concepts are more appropriate to predict the behaviour of complex systems than models where biological processes are just seen as accidental. The developments of the last decades in engineering and social science would have been impossible without teleological thinking, starting in the sixties with the Macy conferences (Pisa 2003). We have today the somehow paradox situation that teleological concepts prevail in computer science and engineering, but are rather neglected in the science of the living where they are more than obvious.

Teleological principles in medicine are mainly used in ‘evolutionary medicine’ (chap 6.8) to understand in how far diseases and symptoms are useful for humans and thus represent an evolutionary advantage (Lewin 1993). They are also found in some vitalistic concepts of CAM. Teleology does, for example, not ask how s symptom is produced by the body but to which purpose, as in Darwinian or evolutionary medicine (chap 6.8). Such a change of perspective induces necessarily a different perception of medical processes.

Equifinality: Equifinality combines the finding of centralisation with the idea of a teleological behaviour. It says that the behaviour of a system cannot be predicted by its parts and that a final state can be approached by different means. “In any closed system, the final state is unequivocally determined by the initial conditions: e.g., the motion in a planetary system where the positions of the planets at a time t are unequivocally determined by their positions at a time to. Or in a chemical equilibrium, the final concentrations of the reactants naturally depend on the initial concentrations. If either the initial conditions or the process is altered, the final change will also be changed. This is not so in open systems. Here, the same final state may be reached from different initial conditions and in different ways. This is what is called equifinality, and it has a significant meaning for the phenomena of biological regulation” (Bertalanffy: 40).

This is quite similar to the polycontextural network of Günther (chap. 3.5), where different points in the net can be reached in different ways. Besides that, equifinality also claims that the system has a certain end, a goal, a τέλος to be attained. “The behavior of a system is intelligent to the extend that it maximizes the chances for self preservation of that system in a particular environment” (Paritsis 2000a: 72).

A simple example from the social field is the question: Why do children play? An analysis shows that playing can promote many tasks like language acquisition, movement development, holding the body in shape, trying out different patterns from riddle solving to mating and so on. But playing is not necessary to achieve any of these aims. All that could be done in a different way and is done in a different way in times of crisis (Henig 2008). But playing is a rather enjoyable way to achieve these ends and fulfils different tasks simultaneously. This is, at least, the teleological assumption.

In neurophysiology it is known that the same results can be achieved in different ways (Sohn et al 2004). For example, after the destruction of parts of the brain other centres are – to a certain extend – able to take over the duties of the destroyed tissue. Equifinal strategies can also be seen with placebos inducing health in using the same or other signalling pathways as a real drug (chap. 2.4.d). But people might get also rid of a lot of diseases just by exercising regularly (Blech 2007b). That is, the aim of restoring health can be attained in many different ways. Many approaches to health are possible: good nutrition, the right use of rhythms, or psychotherapy, just to name some. Equifinality says that everything might serve as a tool to achieve the aimed goal if there is enough reagibility and not too many pathways blocked.

previous -- home -- content -- next