Charlton BG, Andras P. What is management and what do managers do? A systems theory account. Philosophy of Management. 2003; 3: 1-15.

What is management and what do managers do? A systems theory account
Bruce G Charlton and Peter Andras*

from the Schools of Biology and Computing Science*,
University of Newcastle upon Tyne, UK

Correspondence to:
Bruce G Charlton MD
Reader in Evolutionary Psychiatry
Department of Psychology
University of Newcastle upon Tyne

Tel: 0191 222 6247
Fax: 0191 222 5622

This is an age of management for the developed nations, in which to be some kind of manager is the commonest form of skilled employment, and managerialism is the representative political mode. Clearly, the management function is in some sense definitive of modern societies. Yet, perhaps because of this very dominance, the function of management remains opaque. What is management, and what does it do? The nature of management may be conceptualised from a perspective of Systems Theory as the process by which an organisation generates a global representation of its own processes. In other words, management depends upon modelling an organisation. Modelling allows management to perform its distinctive information-processing activities such as monitoring, evaluation, prediction and control. The purposes to which these activities are directed defines the function of management. The function of management is a product of the interaction between a management system and its environment. This is a consequence of the way that management systems will tend to adapt to survive and grow in whatever specific context in which they are operating - this can lead to very different management functions in different environments. There is, of course, a vast and diverse literature in the discipline of Management Science which refers to the nature and function of management. It is not our intention here to engage with this literature, but to offer a fresh approach to the subject from the perspective of Systems Theory.

Management as a system

Defining systems

Systems Theory is based upon the analytic division of the natural world into environment and systems [1,2]. This division constitutes the major foundational, axiomatic philosophical assumption of Systems Theory. On the one hand there is an infinitely complex ‘environment’, and on the other hand there are self-replicating systems. Systems are engaged in processing information. Systems also model the environment, and can respond adaptively to environmental changes. The planet earth presumably began as pure ‘environment’ with only simple, self-organising physical and chemical systems which arose by the chance contiguity of components and energy [3]. For instance, the water cycle is a system by which water molecules associate and dissociate in processes such as evaporation and condensation. More complex systems can only be built incrementally, so that over time, simple physical and chemical systems must have evolved into complex biological systems such as those observable and biological systems eventually led to the social systems generated by human culture [4]. A system might therefore include entities such as a single cell, a multicellular organism such as a human, or social organisations of varying sizes from a corporation, to the UK National Health Service, to a whole nation [5]. For any given analysis, the 'environment' is everything that is external to such a system under consideration [1]. Systems are actually defined in terms of processes, but sometimes processes coincide closely with physical structures so that a cell's environment might (approximately) consist of everything outside the cell, including other systems such as other cells, the whole organism and other organisms [6]. For a human organisation such as an autonomous hospital, the environment external to the system might include physical aspects such as climate and geography, but also other organisational systems such as politics, the law and the media. Management systems (where they occur) are a form of social organisational system which is engaged in modelling the organisation it manages. For a system of management, everything other than itself is ‘environment’, but the organisation that is being managed constitutes the most immediate environment.

Advantages of complexity

Management systems are a consequence of complexity in human social organisation. Systems deal with environmental complexity by making simplified 'models' of the environment. This entails systems generating complexity within themselves [1]. (Complexity may be defined in terms of the minimum length of a string of information in order comprehensively to summarise a state or process. Roughly, this translates as how many 'bits' are necessary to instruct a system [7,8]. The more complex a system, the more complex its model of the environment can be; and the greater potential the system has for adaptive responses when interacting with the environment [1]. For instance, the visual system of a cat is more complex that that of a slug, and the cat can model more aspects of the visual environment than a slug. Hence the cat has the potential to respond adaptively to a wider range of visual information. Likewise, the more complex a management system, the more comprehensively it can model an organisation, and the greater is management’s potential for triggering adaptive responses. This potential adaptive advantage for complex systems is the underlying reason why biological evolution has generated ever-more-complex organisms over the history of life on earth [2,3,9]. The largest and most dominant organisms in the history of life on earth are late products of evolution and still alive today (the largest animal is the blue whale and the largest plant is the giant redwood; the most dominant animals in terms of biomass are humans in the temperate zones and ants in the tropics and equatorial regions - both complexly social [10]. The potentially adaptive advantages of complexity also explain why cultural evolution has led to increasingly large and internally-specialised economic and social institutions - more complex organisations have the potential to interact with the environment more adaptively. The appearance of management systems, and the elaboration of these systems to high complexity, are examples of the same trend.

Disadvantages of complexity

The value of complexity arises from its greater potential for adaptive interaction, because a complex model can contain more information than a simpler model. But complexity in a system is not good in itself, nor is complexity necessarily more efficient. Complexity always carries a cost - the increase in internal 'transaction costs' by which more resources are consumed and more time is spent on internal processing within the system as complexity increases [1,2]. For instance, a complex organism such as the human required the evolution and maintenance of a large and elaborate brain and nervous system to maintain internal co-ordination and to model the environment - such that in a modern human, brain metabolism uses about one fifth of the total energy input [11]. Management arises and becomes more complexly elaborated mainly in the context of ever-more-complex organisations operating in ever-more-complex environments. The process of increasing division of labour and specialisation of function entails increasing co-ordination of organisations, and the presence of other complex organisations in the environment and continuation of rapid social change also contribute to the need to model the interaction between organisations and their environment. Both the organisational and environmental co-ordination are management functions.

Systems are characterised by 'languages'

Systems are composed of processes the currency of which is information. A communication entails emitting, transmitting and receiving signals - what turns a signal into ‘information’ is its interpretation. This implies that information only exists within systems, because signals that are not part of systems are not actually part of communications [1]. Information is expressed in terms of a communications code or ‘language’ [1]. Systems are therefore characterised, and may be differentiated, by their specific languages. Languages have distinctive signals akin to ‘vocabularies’ and processes akin to ‘grammar’. There is a language of within-cell communication based upon cycles of metabolism and the concentration levels of molecules such as calcium . There are also languages of human organisations and social institutions - for example the technical terminology of medicine and the formal (and informal) mechanisms by which clinical information is processed and communicated [12]. Since the environment is infinitely complex, all system languages are inevitably highly selective samples of environmental information, there are also temporal and other constraints upon processing this information, therefore a system’s understanding of the environment is always based on highly simplified models. The vast amount of unsampled and unprocessed environmental data just counts as 'noise'. Noise may have significant consequences for a system, as when an external happening such as a hurricane or a political revolution may lead to the destruction or closure of a hospital. But when environmental events are not modelled by the system, the system does not 'know' about them, not even in a simplified summary. A blind animal may be desiccated by sunlight that it cannot perceive. All knowledge is therefore knowledge-within-a-system [1,6]. This implies that a management system can only 'know' that information which it has sampled, encoded and processed. Other information may of course be 'known' to human individuals who perform managerial functions (office rumour and gossip - for example), but unless such information is incorporated in the processes of the management system it is not 'information' and will either be ignored, or dealt with (unsystematically) outside the managerial system and using another system - for instance when decisions based upon individual morality over-ride the outcomes of managerial systems.

Knowledge and systems

As described above, information is a product of systems and all information occurs in the context of a system. Therefore, the defects of systems are only apparent in terms of discrepancies within systems, appearing as contradictions or imbalances within the system [1,6]. Environments are infinitely complex and disorganised - they cannot be observed directly in their completeness. The simplified models of environments that occur within systems contain the only information concerning environments. This means that the non-modelled aspects of the environment are inferred as a possible explanation for deficiencies in the performance of systems [13]. Management systems therefore 'know' their performance only indirectly, by comparing its system predictions with actual measured outcomes. If outcomes match predictions (within an accepted margin of tolerance) then the management system ‘works’ and the model is assumed to be 'true'. But when information sources conflict, or predictions and outcomes are significantly discrepant, then management knows that its modelling is in some way deficient. However management systems as such detect only the existence of a problem; they detect an inconsistency in their systems, but they do not provide information on the reason for that inconsistency. Understanding the nature of the problem involves setting-up a new system making observations from outside the management system - a system for ‘management-of-management’, in effect. Such sub-systems may be found in some large complex management systems - for instance the supervisory function of auditing management performance and reporting to the owners, or the function of corporate prediction and planning. Management-of-management systems usually model not just the management system, but also the organisation system and the external environment.

Management as a system of global self-reference

We are now in a position to define management in the terms outlined - as a system of global self-reference. However it is necessary to unpack the meaning of these terms. All systems function by means of a specialised language. Although (by definition) the quantity of internal information processing within a system exceeds the amount of information exchanged with the environment [6], nonetheless it is a function of systems to process environmental information in order to predict, respond-to and manipulate the environment. But some systems also model themselves. When a system models itself, then this self-referential-model may be termed a 'second-order' cybernetic system [1]. As a minimal definition therefore, management is a form of self-representation by an organisation, and management systems model organisation systems [14]. But this definition is not sufficiently precise. Partial forms of self-reference are almost universal in human organisations because most organisations require a substantial degree of inter-communication between members of the organisation in order to function. This entails one part of a system referring to another. For instance, nurses in a hospital must communicate in order to look after patients optimally - by itself, this kind of self-reference does not constitute management. We suggest that management is constituted by global self-reference by an organisation: in which a management system operates on the assumption that it models all the necessary aspects of organisational activity, such that the model may be used for monitoring, prediction and planning of the organisation as a whole. If global self-reference is taken as definitive of management, then it is clear that not all organisations will have a management system. Simpler forms of organisation typically neither need, nor benefit from, management systems. Other organisations - while having second-order cybernetic systems - do not have systems of global self-reference. For example, in the past many universities had administrative systems that sampled and modelled a variety of non-core information about the organisation (eg. financial aspects, infrastructure, personnel) but this system did not 'model' or manage all significant aspects of university function (for example, research publications or teaching activities of academic staff were not necessarily monitored). Such organisations could best be described as partially-managed. More recently, UK universities have developed global systems of self-reference, and now constitute managed organisations. The self-reference is global because the management system operates under the assumption that they are modelling all necessary aspects of organisational activity. (Whether or not those outside the management system would agree that all necessary aspects are indeed being modelled is immaterial - the point is that the management system operates on this basis.)

What is a management system?

Not all management occurs in the form of an organisational system - after all, management may be a single human individual. Nor is the system of management deducible from the payroll of people employed as ‘managers’. Indeed, the first task of any analysis based upon Systems Theory is to define the system under consideration, and to establish its operational boundaries. Systems are based upon the binary division of the world into system and environment. Management systems are no exception - the system samples information from its environment, including the organisation it manages. Only this sampled information is subject to processing. For instance, management may sample salaries and working patterns of its employees - this represents information that the system exchanges with its environment. This communication relating to salaries and working is then encoded into the relatively more precise and concise form of the management language - perhaps as numerical summaries. Internal-processing then commences - for example this information may be used to generate statistical analyses to examine the relationship between pay and performance, these statistics may be compared with those of competitors and national norms, and models generated to predict the effect of possible changes to rewards and conditions. This information may also be set in a context of other predicted financial constraints such as expected turnover, profits and market share. Finally, decisions will be made on the basis of this information, and these will be communicated to the organisation. The results of these communications will then be monitored. The purpose of the above illustration is to emphasise that the more complex a management system, the greater will be the amount of internal processing compared with the amount of information exchanged with the environment. The management system is defined as the pathway and processing of the sampled information. It can be seen that the management system may include human brains and bodies, but also the outputs of machines (such as clocks, camera and typewriters); and that processing may include non-human cognitive activity - such as computers which perform statistical analyses. Management is therefore not synonymous with human managers - and the amount of management in an organisation is not the same as the number of ‘managers’ being employed. Management systems are abstract information-processing systems, and their delineation requires empirical study of the specific system under investigation.

What is the function of management?

General statements relating to the function of management are not sufficient, and attempting to deduce management function from aspirational exhortations such as mission statements is likely to be even less enlightening. Due to the cognitive specialisation within modern management systems, the function of a management system is not necessarily understood by any of the participants in the system - just as an assembly-line worker need not necessarily know the identity of the product he is helping to manufacture. The function of any specific management system must therefore be discovered by empirical investigation. From the perspective of Systems Theory, management has a primary and necessary management function, and further secondary and optional functions. The primary function of a management system, as for any system, is its own replication (or autopoiesis [6]). Without this attribute management would neither be observable, nor would it be a system - because all systems by definition process information in order to reproduce themselves. This is similar to the principle of ‘homeostasis’ in physiology. For instance, whatever the secondary function of a nerve cell in terms of its role in vision, memory or whatever - the cell’s first function is to stay alive: to maintain its own metabolic processing over time, and in the face of environmental challenges. So, whatever else it does, management must maintain its own activities. But there is a very general ‘common sense’ idea about the function of management, along the lines that management exists to improve the effectiveness and/or efficiency of organisations - to help the organisations do whatever it is that they do. And Systems Theory adds that management consists of systems for generating a global self-representation of organisational activity. But - so far - this does not describe the function of any specific management system. After all, a management system might use its self-representing model of organisational activity to perform a variety of tasks. For instance, the secondary functions might be to maximise profit, provision, growth or market share. However, management might use its organisational model purely in its short-term self-interest - for instance to enhance the pay and conditions of managers. Management might even use their monitoring apparatus to spy on employees and to inform the secret police of their transgressions. The relevant question concerns what specific functions a management system is actually performing in a specific instance (and in a specific time frame). In a nutshell, the function of management is a consequence of its previous interactions with the management system’s environment - it is the result of accumulated adaptations in response to selection pressures that act upon the self-replicating management system.

Selection and functionality

Understanding the functionality of management, or any other complex system, is best approached by considering the factors that lead to the evolution of complex systems. As described above, complex systems cannot arise in a single step by pure chance [3]. Complex systems are built-up incrementally and elaborated from simpler systems. (However, once a complex system has been formed in this step-wise fashion, it can be applied to different purposes, can be used to process different inputs. For instance, complex universities originally evolved to provide an education in theology - and this already-existing complex system was then quite simply adapted to educating students in other subjects such as law and medicine.) Complex management systems are therefore a consequence of selection processes, and the functionality of the system will depend on its history of selection [15]. For example, a management system may be a product of the selection of organisations by interaction with their economic market environments. If (in such circumstances) it is found that all car factories have management systems, then we might infer that those organisations with management systems had economically out-performed organisations that lacked management systems. In systems terms, the nature of the management system affects the nature of the organisation, and the organisation interacts with an environment in which economics is the major selection pressure. The major determinant of managerial replication is then the economic viability of the organisation. To be more exact in terms of selection processes [15], the system-environment interaction is at the level of an economic market selecting among organisations by favouring some and suppressing others. Organisations are systems of various types, and some system variants will be more favoured than others. Some of the variation in organisations will be caused by differences in management systems (or the difference between having a management system and not having one). So, selection of organisations will indirectly affect the replication of management systems. This indirect interaction between management replication and the organisation’s environment can then shape the nature and functionality of management. On the other hand, the main environmental selection pressure on organisations may be political rather than economic. This often applies in the ‘public sector’ and nationalised industries. The main interaction of public sector organisations is with government and the public administration bureaucracy, and this shapes their functionality. Organisations that disobey or antagonise the state apparatus will decline or disappear over time, while those that remain are those that have been relatively better at satisfying the demands of the state. In a long-lasting centralised command economic system such as the former-USSR, the survival and growth of organisations was often more dependent upon political factors than economic efficiency. The function of management in this context was related to political goals - for instance maintenance of good relations with The Party, full-employment, and political discipline. The point is that both a market-place and command economy management system may be equally functional in the sense that both may be equally-well adapted to their environments. However, they may have adapted to radically different environments. It is the history of environmental interaction that accounts for different function of management systems. The Soviet Union management of a car factory was presumably economically inefficient by comparison with a Japanese manufacturer - but likewise a Japanese-style organisation with Japanese-style management could not have survived in the political environment of the USSR. This emphasises that the functionality of systems, including management systems, is something that has arisen as a consequence of historical, and perhaps contingent, factors. Whether a system is currently adaptive is something that cannot be determined, but at best may be inferred. The same applies to biological evolution - the fact that Homo sapeins has survived and reproduced for two hundred thousand years tell us that the species was adapted to the environment over that time span - but this does not imply that we will last another hundred thousand years if the environment changes significantly in the future.

Functionality is relative.

There is no way of knowing the absolute functionality of a system. Functionality of systems can, however, be compared - although the comparison is restricted to a specific sample of information over a specified time span [1]. Nonetheless, on this basis, one system may emerge as superior to another. For instance, it might be correct to state that throughout the 1980s the USA had more economically efficient car factory management than the USSR, and that US car factory management was itself less economically efficient than the Japanese. Such a comparison would be restricted to particular defined variables (for instance, the number of managers employed for each car produced), and the comparison would only apply to a particular time frame (eg. the calendar years of 1980-89 inclusive). In other words, measurements of functionality are comparative and contextual. But it cannot formally be determined whether Japanese management was optimally efficient, nor how far from optimally efficient it was (nor even whether Japanese management is more efficient today or tomorrow). What of ‘non-functional’ management - a management system that actually does nothing substantive except consume resources? Since the first function of management is to maintain itself in its relationship with the organisation it manages, this implies that in some contexts the dominant interaction may be that between the management and the organisation that is managed. When organisations are not under significant external threat, there may arise a situation in which management evolves to optimise its own growth even at the expense of the organisation. In such circumstances, the managerial practices that best survive and expand may be those that best ensure the power of management in relation to their own organisations, rather than to their effectiveness in promoting efficiency in the organisation relative to external competitors. In other words, the phenomenon of 'parasitic' management may be a consequence of a particular set of interactions and selection pressures in relation to the management system. This would tend to evolve systems of management which are able to grow at the cost of organisational performance. However, even ‘parasitic’ management systems are not truly ‘dysfunctional’ - indeed, they may function very effectively with respect to their dominant interaction. Rather, the functionality of management systems should be defined in relation to their environment, and the force and nature of selection pressures.

The evolution of management

The management function occurs mainly within a context of complex human institutions in which people perform ever more specialised tasks. In such circumstances a complex globally-self-representing management system may be highly adaptive. But such organisations are a relatively recent feature of human social history. For most of its biological history the human world consisted of small scale groups whose economic activity was hunting and gathering. In the hunter-gatherer economy there was some specialisation of labour on the basis of age and sex, but every economically-active person interacted directly with the environment and there was no specialised system of management [16]. The nearest approach to a manager was perhaps a 'shaman', who used altered states of consciousness such as dreams or trances to access relatively complex knowledge that was useful to the band [17]. However, once agriculture had been invented there began a long term trend favouring more economically efficient societies, societies that extracted more resources from the environment per person [9]. This began the trend towards ever greater 'division of labour'. But division of labour can more abstractly be conceptualised as cognitive specialisation and co-ordination. Indeed, specialisation and co-ordination is the underlying principle of biological evolution by natural selection. From simple beginnings, the trend of life on earth has been towards the evolution of larger and more ‘social’ organisms involving the co-operation of many specialised cells, organs and organisms [4,10]. The combination of specialisation and co-ordination is a clue to the rise of managerialism. Adam Smith's famous example of the pin factory demonstrated how specialisation of labour could increase economic efficiency - division of the task of pin-making into a sequence of specialised tasks enabled a substantial increase in productivity - enabling more pins to be produced per unit of labour, or less labour for the same number of pins [18]. Such forms of human organisation as the factory are able to overcome the limitations of a single human organism by generating systems within which many human organisms combine to achieve more than the sum of individuals.

Specialisation in management

Specialisation is a way of overcoming human constraints. (Technological development adds to and interacts with this process [9]). When specialised functioning and systematic organisation are combined, then human limitations may be overcome. But the process does not merely apply to the special case of physical labour, but to cognitive processes in general. Specialisation and co-ordination can overcome the information-processing constraints of the human mind. In the simplest case the management 'system' is merely one human mind that is able to specialise in the limited management function - but further complexity is possible by applying ‘division of labour’ to management. On this basis, it will be seen that management systems arise because it is more efficient for some people to dedicate their necessarily limited mental capacity to tasks related to dealing with aspects of the environment (e.g. the pin-manufacturing 'operatives'), while other people dedicate their equivalently limited mental capacities to tasks related to co-ordinating these people dealing with the environment (ie. managers). In order to make systems more predictable and controllable, and to reduce dependence upon specific human individuals, there is an incentive for the cognitive complexity involved in human information-processing to be reduced, and for this cognitive complexity to be transferred to the organisation [19,20]. Instead of highly skilled 'experts' operating in a simple organisation, as in traditional societies; the trend of ‘modernisation’ is towards having less-skilled (therefore cheaper and more reliable) people operating in more highly complex organisations. The efficiency advantages gained by de-skilling routine operations then enable ever-more-specialised experts to apply their cognitive capacities to ever-smaller samples of information - thereby generating technical innovations[9,14]. Such benefits of specialisation in increasing both efficiency and innovation constitute major potential advantages driving the increase in size of human organisations. The principal cost of this strategy is that complex forms of organisation are necessary to co-ordinate the activity of all these specialists - and complex forms of organisation may require complex systems of management.

The power of cognitive specialisation

One aspect of modern social organisation is that the cognitive capabilities of organisational systems are vastly greater than the cognitive capabilities of any individual human [19,20]. This applies even to hunter-gatherer groups, in which prolonged and focused group discussions among relatively unspecialised individuals are used to make high level management decisions such as where to hunt or when to move camp [16]. In the most successful modern societies, this gap between the capabilities of the organisational systems and its human members is continually growing. In this sense, all human societies are information-processing systems with a power much greater than the sum of their individual human parts [9]. Modern systems of management exemplify this principle. Management systems co-ordinate numerous specialist experts (in law, economics, accountancy, statistics etc) in complex systems that divide and integrate information flows into subsystems such as finance, personnel, sales, buying and so on. No individual could - even in principle - master the whole spectrum of information processing. Indeed, the modularity of such organisation means that all the information in the system is never gathered in one place; instead each hierarchical level of integration samples the minimum necessary information in order to perform its function. The complexity of the management system of an organisation may often become so great to require a further system of central executive management. This central executive will sample, integrate and model information from these strands of management to enable understanding, prediction and planning of the management structure itself.

Systems Theory and management in context

Perhaps contrary to the impression given by this paper, Systems Theory is a large and complex philosophical and empirical field, with many variants of Systems Theory emphasizing different aspects. For example, there are variants of Information Theory, deriving from the pioneering work of Shannon; there is the related area of Cybernetics, deriving from Weiner; the Second-Order Cybernetics of Von Foester; and more recently some aspects of the work on Chaos and Complexity Theory. Luhmann’s work builds on the sociology of Talcott Parsons. In philosophical terms, Luhmann’s Systems Theory does not conform to any of the standard philosophical positions such as empiricism or idealism, although many partial resonances may be noted with the work of a number of philosophers such as Bergson, and post-modernists such as Lyotard. Systems Theory does, however, seem to have a close relationship to the pragmatism of William James, John Dewey and Richard Rorty – for example in terms of the evolutionary selection of systems. Systems Theory is often discussed in a highly abstract, logical and mathematical terminology, and the most important intellectual roots of the subject lie in computer science. But, as with other sciences, the evaluations of Systems Theory are always contingent and open to revision in the light of further evidence or more complex theories. Luhmann would agree with pragmatists that there can be not absolute epistemological evaluations, but only comparative evaluations in practice; and would also agree that there can be no knowledge of absolute timeless truths. Although the stylistic differences between American pragmatism and German Systems Theory are extreme, it may turn out that Luhmann’s work will be regarded in the future as providing an underlying metaphysical basis for many of the assertions and recommendations of pragmatism (ref Metaphysical Club and Mod Imp). Systems Theory is not a matter of ‘anything goes’ relativism, as sometimes seems to be the case with post-modernists. While both Luhmann and Lyotard deny the validity of a single master ‘meta-narrative’, and therefore agree on the impossibility of a modern society being governed by a single rationality, Luhmann would assert that rationality and cumulative capability continue to operate within social systems – such as science, economics or law. On the one hand, Systems Theory denies the possibility of absolute and eternal knowledge – we can never know for sure that something is a system, or what a system does and all knowledge is potentially subject to revision in the light of new information or more complex information-processing capability. But on the other hand, there are rational and quantitative grounds for comparing and thereby evaluating propositions. While definitions of communication are operational, and measurements of communications subject to sampling errors – nonetheless communications can be detected and measured and systems thereby identified and compared. There are resemblances between Systems Theory and the mainstream management science literature. For example, there are explicitly ‘applied’ approaches such as the ‘soft systems’ methodology associated with Checkland and the Sheffield group, and there are many different types of loosely ‘systems-’ or ‘information-theory’ based approaches that are often used by management scientists or consultants. However, the relationship between these practical ‘troubleshooting’ methods and Luhmann’s rigorous philosophical analysis usually constitutes little more than an overlapping vocabulary. Luhmann-derived Systems Theory is a fundamental abstract approach to understanding, a preliminary to analysis rather than a ‘hands-on’ management ideology or tool.

Systems Theory in practice

An example should make this clearer. Systems are constituted by communications of information. It is the differential in quantity between communications within the system, and communications between the system and the external environment, that defines a system. The bigger this differential between intra-system and inter-system communications, the more complex is the system. A complex management system will communicate within its won boundaries much more than it will with the environment. The first step in research is to define and measure communications. The second step is to measure communications and look for differentials in quantity – for differences in the density of networks of communications. When an observer measures communications in relation to a system, then the system boundaries would be defined as lying between ‘clusters’ of communication. If communications were mapped on paper, then the systems would be represented as concentrated networks of reciprocal communication. The spaces between systems would be regions of low-density communications. All communication requires systems; therefore the first step in analysis is to partition communications into two categories of within-system communications and between-system communications. Systems are defined as those clusters having significantly denser communications within the defined boundaries than between system cluster boundaries. Because communication density is significantly greater in systems, this implies that the observed differential in the density of communications is great enough that the observed difference is statistically unlikely to have been generated by chance. This also implies that the more statistically significant the observed difference, the more likely that there really is a system. Because systems are defined as those units which communicate more with themselves than they do with their environment, a managerial system would therefore demonstrate a greater density of communications between managers within a defined managerial system, than between that defined managerial system and its environment. And if there was no such difference in the density of communications, then there would be no managerial system (or, alternatively, it could be said that there were as many managerial systems as there were managers - each managerial system existing in the mind of a manager). As management becomes more complex in its sub-divisions and specializations, this increases the density of internal communications between members of the managerial system much faster than the amount of information coming-into and going out-from the managerial system. A possible use of this analytic framework would be to analyse organizations to compare the ‘official’ structure (such as might appear in the flow diagrams of Quality Assurance documentation with the structure of organizational sub-systems as revealed by an analysis of communications. The first step is to make an operational definition of communication. In other words, to compare the way that information is purported to flow within the organization, with the way that information actually does flow. For example, communications might operationally be defined as official memos, or the e-mail communications, or the participants in meetings. Having defined communications for the purposes of the study, the location and frequency of these communications may be plotted to reveal the sub-systems within the organization. If the formal organizational structure does not correspond with the communication sub-systems then this may indicate that the formal structure needs to be revised. For instance, if such a study revealed that there was more communication between academic departments in a university than there was within departments, this may imply that the existing organizational arrangements are inefficient.


In conclusion, Systems Theory is an abstract philosophical framework that nonetheless entails a highly empirical and investigative form of management science. Systems Theory suggests an evolutionary explanation for the dominance of management in contemporary Western societies. It also clarifies the role of complex organisations in modern society; and predicts that the complexity of organisations, and therefore the role of management, will probably continue to increase – at least for so long as the efficiency-enhancing potential of complexity can continue to outweigh its inevitably increased transaction costs.


1. Luhmann N (1995). Social Systems. (Translated J Bednarz Jr, D Baecker). Stanford University Press: Stanford, CA, USA.

2. Charlton B, Andras P. (2003). The modernization imperative. Imprint Academic: Tavistock, UK.

3. Cairns-Smith AG. (1982). Genetic takeover and the mineral origins of life. Cambridge: New York,.

4. Maynard Smith J, Szathmary E. (1995).The major transitions in evolution Freeman, Oxford.

5. Pokol, 1991 Complex Society. Co-ordination office for higher education: Budapest

6. Maturana and Varela (1972) Autopoiesis and cognition. D Reidel: Dordrecht, Holland.

7. Rissanen J. (1978). Modelling by shortest data description. Automatica 14: 465-71.

8. Vapnik VN. (1995). The nature of statistical learning theory. Springer Verlag: London.

9. Wright R. (2000) Nonzero: the logic of human destiny. Pantheon: New York.

10. Ridley M. (1996). The origins of virtue. London: Viking.

11. Dunbar R. (1996). Grooming, gossip and the evolution of language. London: Faber.

12. Andras P, Charlton BG. (2002) Democratic deficit and communication hyper-inflation in health care systems. Journal of Evaluation in Clinical Practice 8: 291-297.

13. Luhmann N (1992). Observations on Modernity. (Translated W Whobrey). Stanford University Press: Stanford, CA, USA.

14. Charlton BG, Andras P. (2002) Auditing as a tool of public policy: the misuse of quality assurance techniques in the UK university expansion. European Political Science. 2: 24-35.

15. Hull DL (2001). Science and selection. Cambridge University Press: Cambridge, UK.

16. Lee R, Daly R (Editors) (1999) Cambridge Encyclopedia of Hunters and Gatherers. Cambridge University Press: Cambridge, UK.

17. Brody H. (2001) The Other Side of Eden: hunters, farmers and the shaping of the world. Faber and Faber: London

18. Smith A. (1910) The wealth of nations (originally published 1776-7). Dent: London.

19. Gellner E. (1983). Nations and nationalism. Blackwell: Oxford.

20. Gellner E. (1988) Plough, sword and book: the structure of human history. London: Collins.



also by Bruce Charlton
Quality Assurance Auditing
The Malaise Theory of Depression
Public Health and Personal Freedom
Psychiatry and the Human Condition
Pharmacology and Personal Fulfillment
Awareness, Consciousness and Language
Injustice, Inequality and Evolutionary Psychology
Peak Experiences, Creativity and the Colonel Flastratus Phenomenon