Adobe SVG Viewer plug-in recommended






Michael Church




" is impossible to represent the organising principles of a higher level by the laws governing its isolated particulars."

Michael Polanyil (1)


"Those who are obsessed with practice, but have no science, are like a pilot setting out with no tiller or compass,

 who will never know for certain where he is going."

Leonardo Da Vinci c. 1470 (59)




In a world that is becoming increasingly complex, the need to develop a new paradigm of organisation and management is widely recognised. One approach has been to view organisations as complex adaptive systems (CAS); it has been found, repeatedly, that important behaviours of different kinds of CAS can be described simply - the basis of the development of the science of complexity. So far, no set of principles have been proposed to explain what might make any form of human social organisation, more, or less, adaptive; most theorising has been limited to the weaker

and sometimes misleading explanatory levels of metaphor and analogy. In this paper it is suggested that there is an underlying order in the universe, reflected in the phenomena of discontinuity, levels of order, emergence, autonomy and coherence. By understanding this underlying order, six systemic principles can be identified to explain the organisational basis of the unique ability of living systems to respond and adapt to, learn from, and shape a complex environment. It is then shown how these six principles can be applied literally to understand and shape adaptiveness in any human social organisation; two further principles are also identified, necessary to take account of the singular qualities of human beings. Three of these principles - level specific processes, level specific information, and values - are explored in more detail in this paper.




In 1994, a study (2) by the American Management Association found that a quarter of companies investigated had undergone three or more episodes of downsizing in the past five years, yet less than half of these companies subsequently raised their profits. Governments also appear to be having difficulty in dealing with an increasingly complex global environment, with seemingly simple decisions leading to unintended consequences, and the subsequent poor management of emerging crises - eg BSE in Britain - greatly compounding their scale and impact. In 1996 seventeen separate investigations into alleged child abuse in British social service run homes were initiated, indicating, at a very human level, the organisational failure in our social institutions.


Such a picture highlights the important social and economic implications of developing our understanding of organisation. Yet in spite of a widespread recognition of the need for a new paradigm of organisation (3), an examination of the prevailing organisational  literature points to a lack of scientific progress in this field in recent years.


Let me start by identifying two of the most serious limitations. Firstly, the richness of many of the descriptions of this new paradigm has been matched by a theoretical vagueness. Seminal books by Margaret Wheatley (4) who stimulated interest in the application of chaos theory and the science of complexity to organisation, and Peter Senge (5), have raised awareness of the properties of certain features of biological organisation, for example, 'acting locally', 'control without controlling', and

self-organisation. However, until such ideas about the nature of control and distributed decision-making are operationalised with respect to human social organisation, they cannot be applied with predictable outcomes to human social organisation. In general, attempts to demonstrate the utility of biological theories in understanding human social organisation have failed to move beyond the weaker explanatory level of metaphor and analogy (6).


Secondly, the growth in new ideas about social organisation have taken place in a context where practice has come to dominate theory; organisational restructuring is big business, with a proliferation of gurus, each with their own' brand of organisational practice. World-wide, one of the most influential of these brands has been Michael Hammer's business process re-engineering (BPR). Hammer appears to have drawn some of his ideas from the work of W. Edwards Deming (who developed a theory of organisation, founded on an understanding of process, variation, information, systemic thinking and scientific method). However, rather than trying to build upon or extend Deming's work, BPR has been simply spawned as a set of techniques in which theory is notable by its absence. As Hammer (7) wrote: "...the absolutely essential element in every reengineering project is to be directed at process...practically everything else in reengineering comes down to technique..."  Later, at a conference (8) he explained the high number of re-engineering failures estimated in his original book: "Why had 70% failed? Because they were not doing it right, and if you do not do it right you will fail too, but if you know what you are about then you will succeed."


The absence of theory is problematic in any applied field: practice all-too-easily becomes tampering when it is driven by faith; circular argument and self-fulfilling prophecy replaces prediction and hypothesis testing; and the opportunity for scientific progress is greatly diminished.



The development of the science of complexity has been made possible by the repeated finding that important behaviours of different kinds of complex system can be described simply. In this light, a question of particular relevance to our understanding of social organisation is whether or not a small number of principles can be identified that have explanatory power and provide insight into the phenomenon of adaption? (I will define adaption to mean the ability to respond to, learn from, and shape the environment.)


The approach I have taken in trying to answer this question has been to identify the principles underlying adaption in biological systems (which show an unsurpassed ability to adapt in complex environments), and see if these can be applied literally, rather than metaphorically, to social organisation, showing how the important differences between biological and social systems might be accommodated within any such theory. As it turns out, much of the knowledge required to do this can be found in the existing systems literature (though not particularly accessible, nor presented in the framework outlined in this paper) and empirical research in various branches of biology provides a rich source of supporting information. Before I begin to go into more detail about human social organisation, it is necessary to lay some conceptual groundwork. Firstly, I will indicate that there are some regularities underlying all phenomenon - microscopic to macroscopic - in the universe. Secondly, the relevance of these regularities to the adaptive fitness of

biological systems will be explored. Thirdly, the implications of these with regard to different types of control in complex adaptive systems will be considered.




One of the most interesting prepotent phenomenon in our universe is that of discontinuity. This can be seen as the vast array of different levels of order or complexity; from fundamental particles to galactic clusters. What might the physical basis of these discontinuities be? Most of us are, at least, dimly familiar with the Second Law of Thermodynamics: this affirms that all real world dynamics proceed spontaneously so as to maximise entropy (in effect, our universe - if it is a closed system - is running down to a state of homogeneity). However, since entropy is produced faster by order than disorder - the law of maximum entropy production, Swenson (9) - the universe has an important order producing quality.


Thus, because every jump to a higher order state greatly increases the rate of entropy production, the emergence of discontinuities and ever-higher levels of complexity is an inevitable consequence of the universe moving towards maximum entropy. Let me be more specific about what I mean by discontinuities and higher levels of complexity. Table 1 shows some simplified examples, or snapshots, of naturally occurring discontinuities.


Table 1

Discontinuity, Levels of Order and Emergence: Some Examples






Binding property with other molecules - eg Carbon

Monoxide (CO), a simple molecule, has a strong

affmity for haemoglobin (300 x that of Oxygen).


Randomly folded polypeptide



(ie folded polypeptide in

medium supporting

disulphide exchange)

Enzyme activity.


Kidney glomeruli

molecular filtration (of molecular weight < 68,000)


pH of blood held constant at 7.4




Individual ant processes

30 - 40 simple 'rules of thumb' behaviour

continue hunting for certain foodstuff if the present

foraging load is accepted by nestmates.


meeting the needs of the colony (eg re nutrition,

temperature, defence, etc.)


Creation of new open systems in new habitats when

eg steady state reached, (eg dispersal flights, nest build-

ing, mating, reproduction, etc.)



Sensory-motor system

unskilled and skilled movement


concrete thought


x Levels


abstract thinking


x Levels


Lowest level





In all non-living and most biological systems the structural basis of discontinuity is visibly obvious (10) See examples, Table lA, lB, & lC). In each case the transition from one level to the next higher order level results in the emergence of new properties: atoms react to become molecules with new emerging properties such as binding or catalysis; polypeptides in their natural medium undergo an oxidative process and fold into a specific shape from which enzymic properties arise; and homeostasis is an emergent property of the body organ metasystem, of which molecular filtration is but one of many processes at a lower level.


In example Table lD higher levels are not so visibly obvious. At the lowest level individual ants in a colony operate using some 30 to 40 rules of thumb (11) or distributed processes (eg 'continue hunting for a certain foodstuff if the present foraging load is accepted by nestmates'). The queen or queens are part of higher order processes concerned with longer term adaption, increasing the probability of genetic replication, eg by creating new open systems in new habitats: when a steady state is reached in a particular environment, dispersal flights are triggered to seed and develop new colonies through reproduction.


Finally, in Table IE higher levels of order can be seen purely as different levels of Thought (12), both concrete and abstract, arising from the unique structure of the human brain with its parallel and distributed processing.




In evolutionary terms, adaptive fitness can be viewed as the probability of genetic replication (13). The development of higher order levels of complexity can be shown to increase the probability of this in two important ways: 1) Each higher order level adds the greatest possible value, as its emergent property allows the management of relationships and maintenance of a dynamic equilibrium with a qualitatively unique slice of the environment (14); (15).


The notion of each higher order level in dynamic equilibrium with a different qualitative environment reflects a well recognised and fundamental systemic property of biological systems irreducibility (16). As Richard Dawkins (17) puts it:  " every level the units interact with each other following laws appropriate only to that level, laws which are not conveniently reducible to laws at lower levels." The irreducible laws referred to by Dawkins are embodied in different space-time relationships; thus, the local environment of each higher level of complexity has different 'rules of the game', reflected in the qualitatively greater scope of environment, longer process cycle time (18) and different order of information and resources required at each level. The example of the ant super-organism Figure lD demonstrates this. It does not take too much of a leap of imagination to realise that the scope of local environment served, process cycle time, information and resources required for an individual ant to run its processes (eg following a pheromone trail) are very different (19), to that characterising the emergent properties found at higher levels; eg meeting the nutritional needs of the colony, or the process of developing new colonies in other habitats as the existing colony reaches a steady state in its environment. It should be clear that such higher order processes enhance the probability of genetic replication, by extending the activities of the ant colony

temporally and spatially.


2) As a direct corollary of emergence, the maximum autonomy is created for every higher and lower order level of complexity (20) - reflected in what Herbert Simon (21) calls loose vertical and horizontal coupling. Each level in a biological system has the maximum freedom to learn about, manage the relationships in, and maintain a dynamic equilibrium with its unique environment: top down interference is minimised; the misuse of time and resources - eg to double-guess or tamper with the work of lower levels is prevented; and responsiveness to the environment is maximised as there is no chain of command (22).


We can draw on our everyday experience to illustrate this complementary aspect of emergence - autonomy. Consider an individual walking to work across a busy city; at the same time as planning a major culture change initiative using concepts and ideas (abstract thought), her sensorimotor system - a process operating at a lower level of complexity, below consciousness (like an autopilot) - gets her safely to work. The common motorway experience of 'losing miles' is another example of this partial autonomy of process levels. How much abstract thinking would be possible, I wonder, without this partial autonomy of levels; if each of us had to constantly monitor and control our every movement?


As a prelude to the next section on the nature of control in complex adaptive systems I would like to highlight a crucial difference between the levels of complexity, emergence and autonomy found in biological systems and what might be seen in human social organisation. It has already been noted that in biological systems, levels of higher and lower order complexity are usually visibly obvious, although they may not be (as in Example 1E). This has led many theoreticians (24) to make the fallacious leap of logic that the visibly obvious levels - eg management levels, organisational hierarchy - in human social organisation are equivalent to levels of complexity found in biological systems (25). In fact, this is most often not the case. Because social institutions and organisations have usually developed without regard to a proper understanding of process, their structures have become a mishmash of layers or echelons often with two or three echelons covering a level or part of a level of complexity (26). These echelons violate the underlying order of the universe and as a consequence the powerful natural ordering phenomena of emergence and autonomy are lost, reducing the quality and range of relationships with the environment (27).


Since these phenomena are also the basis of the 'order for free' found in nature, it is almost inevitable that the nature of control in much human social organisation is top-down command/control rather than what Senge (28) calls "...control without controlling".




Let us look now at the nature of control in biological systems. Table 2 shows a continuum on which control or order in systems can be characterised. At one end of the continuum the elements in the system have complete autonomy, there is no control or ordering of relationships. At the other end of the continuum, there is no autonomy for individual elements as relationships are fixed or rigid (ie as in dead or mechanical systems).


Table 2


Continuum of Control and Order in Systems


Description of Control





No Control


Independent un-coordinated


Strange Attractors


Random relationships.

Control without Controlling


Highly ordered interdependent relationships in which the costs of achieving order are minimised; ie no premium paid.

Command & Control


Predominantly dependent relationships. Order is controlled from above with a premium paid.

Total Control


Fully dependent,

Fixed and immutable relationships.


Typical human social organisation is usually to be found towards the rigid end of the Continuum (29), with top-down command/control. Such control is not only costly to operate, but significantly limits autonomy, impeding self-organisation, reducing responsiveness and the ability to learn from the environment.


In contrast, living systems show coherence, aptly described as 'order for free' – anyone who has studied a flock of birds, a school of dolphins, or an insect colony, will attest to the almost magical quality of this phenomenon, in which, paradoxically, a high degree of local autonomy is preserved alongside highly ordered coherent whole system behaviour. A move either way on the continuum results in this balance being lost; chaos lies in one direction and increased rigidity the other.


It has been known for some time that even simpler coherent behaviour in nature cannot be reproduced using centralised command/control. Even disregarding the massive associated costs, before long, as the number of entities to be command/controlled increases, it becomes computationally impossible to direct a complex system top down, a point shown empirically using computer simulations (30) of flocking behaviour. In human social organisation this 'difficulty' has been reflected in the commonly seen cycles of centralisationldecentralisation.


In biological systems, centralised control - as a detailed specification of what is required, held centrally, passed downwards as a continuous flow of commands to other parts of the system which are dependent upon these commands for their behaviour - is minimised by creating clear boundaries (31), within which the optimum self-organisation can take place. In an earlier paper I used the term regulated autonomy to try to give a feel for this form of control.


Three key criteria that underpin regulated autonomy are introduced next using the ant superorganism as an illustration:


Minimum critical specification of processes (32) : ie specify the simplest possible processes - ie fewest number, simplest design within and between processes – to enable the particular purpose to be achieved. For example, Bert Holldobler and Edward O. Wilson (33), have found that individual ants operate using only some 30-40 "...rules of thumb, elementary decisions based on local stimuli that contain relatively small amounts of information ", such as "...continue hunting for certain foodstuff if the present foraging load is accepted by nestmates; follow a trail if sufficient pheromone is present; feed the queen more if final instar larvae are present; and attend the larvae and other immature stages if regular nurse workers are absent." I need not remind my readers of the contrast with human social organisation, where the overcomplication (34) of processes is endemic.


Foster inter-dependant relationships within the system - specify processes to produce the minimum necessary dependence in relationships and the maximum independence (35) - ie creating inter-dependent relationships. For example, by not specifying how to follow a pheromone trail, the ant process 'follow a trail if sufficient pheromone is present', creates a boundary within which the maximum possible independence is produced. Again, this criterion is often violated in human social organisation, eg when over-dependent relationships are created by adding echelons of management, and through overcomplicated, poorly specified or understood processes, with rigid procedures that limit independent action, severely constraining self-organisation.


Distribution of processes - rather than hold processes centrally, distribute these processes (and the means to run them) locally (not just to individual ants, but to that level of complexity). As noted earlier, in ants this distribution takes the form of internally wired rules (the means to run these processes - such as information and resources - are also distributed). Picking one aspect of distribution - information – it is my view that the massive problem of information overload identified by Reuters (36) is largely explainable by the failure to redesign and properly distribute information processes when restructuring organisations.


The two important advantages of this form of regulation can be summarised as: . minimising the economic costs of achieving and maintaining coherent purpose (included in this are costs arising from changing relationships, such as leaving or rejoining the ant colony), minimising constraints on autonomy; hence responsiveness to the environment is preserved.


Let me address one final common misunderstanding of this form of regulation and the implications for self-organisation as applied to human social organis,ation. Although the importance of autonomy has been recognised in the literature, it has been interpreted in purely concrete terms (37) - ie the self-managed team; which has a visibly obvious physical boundary (like the worker ants in my example). However, unlike worker ants, a human self-managed team will often not be responsible for running transformation processes covering a qualitatively distinct environment (38). Rather such a team will usually still be part of a overlayered structure of echelons and although creating a clear boundary will help to facilitate self-organisation, any gains will be undermined if that team is not responsible for managing a qualitatively distinct environment. A visual image of this human self-organisation as applied to ants would be that of worker ants, having a smaller team of worker ants above them providing direction - the proverbial monkey on the back, ie managerial tampering - rather than adding value by doing something distinctive.


To summarise, in biological systems the primary means of achieving coherence are the levels of higher and lower order complexity; with processes, information, and resources, minimally specified and having the property of inter-dependence, distributed to allow highly autonomous operations within each level of the system.




Table 3 summarises six principles or necessary conditions for defming the distinctive organisation that leads to what I have already shown to be the desirable phenomena of emergence, autonomy and coherence. It should be noted that the separation of these principles into those shaping autonomy and those shaping coherence is merely to indicate the relative contribution of principles in each area; in reality each principle is necessary to form a system with the properties of autonomy and coherence. Although these principles can be applied literally to human social organisation they are, in themselves, insufficient; they fail to take account of two key differences between biological systems in general and human beings in particular.


Table 3


Principles for Enabling Adaptiveness in Living

Systems and Human Social Organisations


Living Systems

Human Organisations

Have levels of complexity with

emergence, autonomy and distributed control reflected by:

Can be designed to have levels of complexity, with emergence, autonomy and distributed control reflected by:


Level specific transformation processes;

Level specific information processes;

Level specific resource control.


Level specific work/business processes;

Level specific information processes;

Level specific resource control;

Level specific capabilities.


Relationship authority;

Relationship style;

Co-ordionating mechanismns


Relationship authority;

Relationship style;

Co-ordionating mechanismns;

















Firstly, humans are the only species whose cognitive powers have developed to enable action significantly beyond the limited sphere of the immediate physical world; this is evidenced by the extent of human achievement in inventing, creating, discovering, and organising. An important aspect of this cognition is what has been called self-reference - the ability to stand outside or above one's self (39), reflect upon, and change what one does. Whether or not you or I are able to self-refer to produce effective decisions in the face of a particular real-world problem depends on our level of capability (ie level of mental abstraction).


Secondly, in contrast to physical systems - which are bound by laws - people may choose (or chose not to) follow rules. We are all actively engaged in creating meaning in our worlds, and the meaning-made by each of us represent the intemalised rules guiding our behaviour. The principle of values is required to understand this - as we shall see later this is a very different formulation from the usual top-down prescriptive list of ill-remembered corporate values.


Because of time limitations it is not possible to address in detail each of eight principles proposed in this Paper. However, a brief summary explanation of these principles is given in Appendix I, with introductory references for those who wish to follow up specific areas. In my own practice I have found these eight principles to be a powerful systemic framework for reviewing and developing a deeper understanding of any organisation.



I am aware that many reading this Paper (presented at the British Deming Association annual conference 1997) have a considerably deeper understanding of Deming's work than I do. With this in mind, I had hoped that some of the connections with his theory would be self-evident. In case my assumption has proved false, and to focus our minds on the three principles I have choose to illustrate in more detail (as applied to human social organisation), I have selected four of Deming's axioms (40) and in Table 4 and highlight some common themes with these principles.


Table 4


Four of Deming’s Axioms



Axiom 1 : Control of a business is established by leadership and co-operation.

Axiom 2 : Improvements are due to increasing division of work, information and creativity.

Axiom 5 : Inconsistencies and contradictions which become apparent on analysis

of the system may be used to detect and isolate the built-in flaws of the system.

Axiom 6 : Role of Management is to create a secure environment eg free from fear.





Axioms 1 & 6 identify the importance of having the right values shown

by management as leaders, and the need to move away from continuous

top-down monitoring, command and control. The principle of values

outlined in this Paper are a means of operationalising key aspects of

leadership: ie it shows how values might be distributed through system

design, hence enabling top management to scrutinise and reflect upon its

role in leadership and its responsibility for managing social processes.

Axiom 2 identifies the importance of the distribution of information and

processes as a means of enabling continuous process improvement

(moving away from top-down control). The principle of level specific

information processes provides a methodology for showing how information

can be distributed in the correct form to the appropriate system level.

Axiom 5 reflects the fact that no system can be perfect, eg as much as

90% of productive power can be wasted in over-complication. One

important area that the principle of level specific processes can throw

light upon is over-complication arising from the failure of management

echelons to add value by not making a unique contribution.































The three principles I have chosen to illustrate further are level specific work Processes; level specific information processes, and values. It should become apparent from these examples that all the principles interact strongly with each other and unfold in a non-linear fashion (as one might expect from our understanding of other complex adaptive systems (41).


Level Specific Work Processes

Table 5 provides a brief summary of the seven process levels that can be used to describe work of any complexity - whether it be cleaning a floor or the process Gandhi used to secure India's independence from colonial rule. The different qualitative nature of each level is indicated by the different temporal and spatial characteristics (42) of each higher order level. When organisations are viewed through the lens of level specific processes, they will often be found to have an overcomplicated nesting of processes (43), a command/control hierarchy in which extra echelons make no distinctive contribution to the work of lower levels; they simply undermine economic quality by managerial rework or tampering.


Table 5


The Seven Generic Process Levels in Work




Brief Description of the

Transformation Process

Spatial Boundary

of the Process

Process Cycle Time

Added Value for Customers





Bring into being a product or service with minimum waste using available means.

Immediate physical world to locality.

Up to 3-months


Meeting needs directly or indorectly through the specification of the means required to create product or service.

Locality to sub-national geographic area.

3-months to 1-year


An open system that meets the needs of an existing group of customers to provide products and services that are continuously being improved.

Sub-national geographic area to single nation state.

1 to 2-years

Added Value for the Future


 new value chains)


Transforming signals of change in a value system of main stakeholders into new products and services that can meet stakeholders stated or unstated needs.

National state to region of the world.

2 to 5-years


Creating an intent which enables the internal (organisation) and external (industry structure) change necessary for full connection with the evolving socio-economic environment.

Region, or regions, of the world.

5 to 10-years

Cultural Transformation

(Creating different

world views)


Redefining political, socio-economic, and technological relationships to manage turbulence in the environment for groups of businesses, regional groupings, national states and transnational organisations.

Regional to global.

10 to 20-years


To bring into being current or nascent contexts for future generations or institutions, shaping different global futures.


20 to 50-years

















In contrast to some of the new forms of 'horizontal' (45) organisation, which have much greater clarity and less overcomplication (two good examples are that of the strategic network form of organisation - eg Benetton - and the middle-up down management (47) of some Japanese companies), some of the most overcomplicated nesting of processes are to be found in organisations that are or have been state owned. For example, following privatisation, British Gas had a structure of some 15 echelons, with bundled processes obscuring the actual economic value created by particular process streams. Two subsequent reorganisations and the recent demerger of part of its business 'Centrica' (to manage the problem of monopoly regulation of the its Gas Business in Britain) has led to an organisation in which the main processes are unbundled, with greater clarity and a much closer fit to the six levels of complexity required by the business.


The author used the principle of level specific processes within one of BG's businesses - TransCo - enabling the shaping of a new organisational structure that was process aligned and has the potential to add the maximum economic value, prior to a detailed process map of the organisation being available. With the manager and project staff of a core customer facing process (Deliver Gas), it was possible to define this process as requiring only three process levels (1, 2 & 3 in Table 5) in contrast to the previous overcomplicated, many-level structure. Some two years on, this three level structure has been implemented, providing an organisational context that is facilitating the introduction of continuous improvement with the information processes and supporting culture change required.


To summarise, the principle of level specific processes is a powerful tool for managers to be able to model and gain a deeper understanding of the shape of organisation that will support a move to process working. It also provides a means of auditing for overcomplication of managerial processes at all managerial levels (48).


Level Secific Information Processes

Three types of interrelated information process have been identified (49):

. control - for steering and monitoring processes,

. strategic - for making meaningful choices,

. audit - for sampling and deeper understanding.


Viewing these as distributed through the different process levels challenges the traditional top/down - strategic, tactical, operational - view of organisations; a view that implicitly devalues the contribution of the majority of employees in organisations. In the distributed model a continuous stream of decisions at every level is a necessary condition for maximising organisational adaptiveness: with front-line workers and managers at every level enabled to organise their own work and make meaningful decisions regarding the work processes for which they are responsible, removing the need (and justification) for a steady top down flow of commands.


Rather than valuing the work of one level above another, the model presented here suggests that each level of work required by the organisation (depending on its purpose and the nature of its business environment) are equally important but different; in the time frame over which value is added (see process cycle times, Table 5), the nature of the environment served and hence the nature of information processes required. One way these differences can be summarised is to view levels 1, 2 & 3 as making up a domain that adds value for customers, with a process cycle time of up to two years, whilst levels 4 & 5 add value for the future with a process cycle time of 3 to 10 years.


It is my view that many Japanese companies have been particularly successful in understanding, clarifying, and embedding these two different domains of organisational adaptiveness. Indeed, understanding the differences between these domains, in the processes and information necessary, is the key to managing within a complex, unpredictable, differentiated but global environment.


The level 3 process outlined in Figure 1 is an open system that adds value for customers, ie meeting the needs of a known group of customers (the focus of much total quality management). The assumption underlying this process is that of a predictable environment with various trends unfolding in a regular manner. Economic quality and adaption at level 3 is founded upon managing and continuously improving existing relationships in the value chain.


However, the business environment has become increasingly likely to show discontinuities; eg when markets mature and decline, with technological innovation, as personal and social values suddenly shift, when a wave of selling is triggered around the world's financial markets. In order to deal with the qualitatively different complexity inherent in a discontinuous environment, the processes found at level 4 (and 5) are necessary. At process level 4, economic quality depends on bringing into being new relationships with stakeholders; creating value chains that do not yet exist. The information processes necessary to successfully run level 4 work processes reflect the broader more complex environment of this process level; for example, the control information process (see Figure 1) must be able not just to measure, but also to influence and shape the values and reactions of stakeholders.


One of the important consequences of these two different domains not being clearly embodied in organisational process, information and structure, is the loss of the loose coupling (or autonomy) between the 3/4 axis, directly compromising the ability to adapt to discontinuity. When these domains are too tightly entrained through the 3/4 axis (eg by the presence of extra echelons) the new organisational responses that must be developed by levels 4 and 5 to deal with the discontinuities occurring, will be constrained and more likely to disrupt existing operations that add value for customers (3, 2 & 1). Loose coupling enables discontinuities impacting on levels 4 (and 5) to be managed, for example, by the creation of new level 3 open systems and/or modifying the outputs or relationships between existing open systems (50).


Of course, disruption of the loose coupling between levels is not the only way organisational adaption to discontinuity can be undermined. The poor understanding and design of information processes at process levels 4 & 5 seems to be widespread in organisations and social institutions - some examples of common failures are listed in Table 6.


Table 6


Some Common Failures in the

Information Processes at Levels 4 and 5



Relationships with important stakeholders are ignored or poorly managed.


Inter-relationships / interaction between different technologies not properly considered.


Conflicting technical expert advice / findings are not fully explored.


Tacit (Level 1) knowledge is not accessed when developing new products / services.


Failures to scan for or detect early weak irregular signals of major discontinuities (butterfly effect).


Process outcomes that guide information process design (eg vision, scenarios, purpose) are based on the assumption of a predictable incremental future.


Social and / or technical relationships within the wider system not properly understood.


Values signalled by the information processes are negative or contradictory.




In spite of the long recognised importance of human organisation as socio-technical Systems (51), the widespread failure to effectively manage social processes remains enshrined in HR and Trade Union functions; what should be a direct set of relationships between employees and management is dealt with indirectly, through third parties.


If there is to be a sea change in the nature of relationships in work, the leaders of our organisations and social institutions will need to draw into their practice a deeper understanding (52) of the basis of human social co-operation: a key aspect of this is a formulation of values that is quite different to that found in traditional value based management.


Macdonald (53) has identified six core values and their opposites that are necessary for social co-operation:

Honest   v   Dishonest

Loving   v   Unloving

Fair   v   Unfair

Courageous   v   Cowardly  

Dignifying   v   Undignfying

Trustworthy   v   Untrustworthy


In contrast to the typical list of organisational values, these core values have two of the inherent characteristics crucial to the development of coherence. Firstly, these core values are pre-otent and distributed (54) to every individual regardless of culture (unlike the often ill-remembered organisational values). Secondly, as part of every individual's tacit knowledge, they facilitate inter-dependence by providing a point of reference from which each of us can hold an internal dialogue (55) about our own and other behaviours. Contrast this with traditional lists of organisational values which, in spite of their apparently positive aura, are prescriptive; implicitly putting employees into a dependent relationship and increasing the risk that these values may be used to stifle debate, dialogue, and the questioning of assumptions, thereby reducing organisational adaption.


Values and System Design

Let us return now to Figure 1, the generic model for information processes at level 4. Even a brief reflection on recent discontinuities - the BSE and E coli crises; cash for parliamentary questions; top managers' pay in Britain - gives enough ground for suggesting that the reactions of stakeholders are frequently misjudged and poorly shaped; indicative of a serious shortcoming of level 4 control information processes. At the heart of this is the failure of top managers and leaders of our institutions to understand information and values, and the implications for system design that flow

from their interrelationship.


At level 4, information cannot simply be treated as technical and tangible, rather the emphasis must be upon its intangible social quality. Knowing whether or not systems are doing what they say they are (level 4 audit process, Figure 1) therefore demands not just effective technical design (how easy are they to use, how efficient are they etc.) but social design so that appropriate values are signalled. Macdonald suggests that the prime attribute that results in systems being positively or negatively rated on this core values continuum is the degree to which these systems differentiate or equalise. As our society has become more egalitarian it has brought into sharp relief systems that are highly differentiated.


Reducing the degree of differentiation of a system can lead to a dramatic improvement in process capability. In a recent year-and-a-half long study in London it was found that ensuring police officers routinely provided a rights information leaflet when initiating a stop and search procedure (a high percentage of ethnic minorities are subjected to this procedure) led to a decrease of more than 50% in the number of arrests. Here, the provision of tangible/technical information altered the nature of the social relationship, dramatically reducing false arrests, ie waste (process capability in level 1 processes is evidenced by the amount of waste).


This example illustrates two further points. Firstly, it shows how powerfully appropriate values can be signalled (and distributed) through apparently minor changes in organisational systems and processes. Secondly, it indicates that leaders must not just consider their own view of their organisation's systems, rather they must understand how other stakeholders perceive56 any system or process. In the example above, one might guess that those from ethnic minorities have been subject to a very different set of experiences leading them to perceive the stop and search procedure as highly differentiating. A rights information leaflet is an organisational signal indicating equalisation to both the police and members of ethnic minorities, and indeed is likely to act as control infonnation modifying any implicit differentiation in the way the procedure is carried out by the police constable.



As the world has become increasingly complex, there has been a steady stream of signals indicating an all too frequent failure to tackle complicated problems effectively. Shaping sustainable economic development and reconciling differing economic interests (eg between poor and rich countries); managing the social implications of scientific development and new technology; moving towards a just society in which rights are balanced by responsibilities, and poverty, crime and violence are reduced; creating proper health-care rather than illness-treating systems; maintaining even one of earth's many gifts to us - bio-diversity; will all require more sophisticated solutions than those we have so far been able to deliver.  Indeed, since our world is on the edge of another major discontinuity, as the ability to manipulate our DNA pulls humanity into what Stephen Hawkin (57) calls a new phase of "...self-designed evolution..." how successfully our social institutions can be reshaped will have profound consequences for future generations. No amount of personal, interpersonal and team skills, important though these are (58), will be sufficient, if the organisational practice of our leaders is not founded on science. We would do well to remember the words of Leonardo Da Vinci (59) who noted c. 1470 that: "Those who are obsessed with practice, but have no science; are like a pilot setting out with no tiller or compass, who will never know for certain where he is going."


This paper has argued for a science of organisation based on an deeper, system based, understanding of our natural world. Eight principles for shaping the adaptiveness of organisations and social institutions have been identified - it is, I believe, a theoretical framework that can help us address some of the pressing challenges of organisation that our world faces now and into the next millennium.







Eight Organisational Principles: Summary and Brief Explanation


1) Level Specific Work Processes

- Processes: are they properly specified, understood, mapped, aligned, in control ? (60) Is there a technological infrastructure that supports process working?

- Process levels: are they clearly defined reflecting processes distributed to the correct level, does each level add full value in relation to the local environment served ? (6l)

2) Level Specific Information Processes

For each work process level, three types of information process (62) are required: Strategic (meaningful decisions); Audit (sampling and deeper understanding); Control (monitoring and steering).

- Are the information processes in place to enable staff at each level to make effective decisions, steer and continuously improve processes?

- In particular, is properly specified control information available (right person, place, level, and frequency) to allow work processes to be steered (or is top/down control used as a substitute)?

3) Level Specific Resources

- Resource Availability: are the resources available to allow every level to achieve its goals (eg people, their skills and competencies?)

- Resource Control: is there the correct scope of resource control for each level; ie reflected in managerial accountability, budgetary control and process time scales?

4) Level Specific Capability

Traditional approaches to understanding what cognitive attributes – eg intelligence and critical reasoning - enable contribution to higher management jobs or process levels have limited explanatory power, accounting for less than 20% of the variance. A non-reductionistic theory based on the notion of capability as the exercise of discretion which suggests that individuals have level specific capability, growing at different rates for different individuals. Level specific capability should be regarded as a necessary condition for the

individual to make a flow of effective decisions at the matching process level complexity:

- Does your organisation have the right pool of capability (including potential) so that a correct balance between the needs for organisational stability and change or development can be achieved?

- Are their effective organisational systems (64) for keeping people in-flow and fostering personal responsibility for career development?

5) Relationship Authority

A minimum authority in relation to processes, based on an understanding of variation, must be specified in order to sustain any system's integrity. This must include vertical relationships (eg distinction between out-of-control points and process capability) and horizontal relationships (as process capability and input/output specification). In human social organisation this minimum authority in relation to processes must be translated into, minimum authority in different roles (65).

- Are peoples' formal accountabilities matched with the correctly specified and understood authority laterally and vertically with regard both to processes (re variation) and people?

- Do the other organisational systems (eg information processes) enable or undermine these authority relationships?

6) Relationship Style

As an organisation becomes flatter, the maintenance of top/down control removes much of the organisational benefit (of increased responsiveness) and can lead to loss of control (eg with large spans of control). The loose coupling necessary for coherence is in fact a loose/tight relationship (loose in respect of the high level of autonomy given, tight in the sense of a high level of clarity about outcomes). In human social organisation this must be reflected in the nature of relationship between managers and their direct reports.

- Do managers have the right skills to support process working with larger spans of control? .

- Do direct reports have the skills to shape and improve their managers style?

- Are these skills reflected and recognised in selection procedure processes/criteria?

7) Co-ordination Mechanisms

Creating an effective organisational system depends on understanding the key inter-dependencies between processes and therefore the minimum necessary co-ordination mechanisms66 eg include systems, processes, procedures, information processes, service level specification etc.

8) Core Values. System Design and Leadership (67)

All of us tend to evaluate the behaviour of others and organisational systems and processes (HR etc.) on a small number of core values (necessary for social cooperation) - trust, honesty, courage, love, dignity & fairness. How positively or negatively organisational systems are evaluated depends on the degree to which these systems differentiate or equalise (differentiated systems tend to be

rated negatively). A primary leadership role of top management is the design of systems to reflect positive core values.

- How do you predict the current systems in your organisation would be rated on this values continuum by staff? And how do they actual rate it? Is there a gap?






(1). Michael Polanyi, The Tacit Dimension, Routledge & Kegan Paul, Great Britain, 1966.

(2). Cited in: When Slimming is not enough. The Economist. September 3rd, 1994 pp 67-68

(3). For example see: Nitin Nohria and James T. Berkley, An Action Perspective: The Crux of the New Management. California Management Review, Summer. 122.. 4, pp 70 - 92. John A. Byrne, The Horizontal Corporation, Business Week, Dec. 20th, 1993. Homa Bahrami, The Emerging Flexible Organisation: Perspectives from Silicon Valley. California Management Review, 4, 33-52, 1992.

(4). Margaret J. Wheatley. Leadership and the New Science- Learning ahout Organisations from an Orderly Universe. (Berret-Koehler: USA, 1992)

(5). Peter Senge, The Fifth Discipline. The Art & Practice of the Learning Organisation. Doubleday, USA, 1990.

(6). For example, see John Mingers (1995 detailed exploration of autopoiesis - Maturana and Varela's theory of the living - in which he suggests that "...using autopoiesis metaphorically is reasonably unproblematic...", p 151. This is in contrast to attempts to make a more literal application to social organisation. John Mingers, Self-Producing Systems. Implications and Applications of Autopoiesis, Plenum Press, New York, 1995" Humberto R. Maturana & Francisco Varela, The Tree of Knowledge: The Biological Roots of Understanding. Shambhala, Boston, 1992.

(7). Michael Hammer and James Champy, Re-engineering the Corporation A Manifesto for Business Revolution. (Nicholas Brearley: London 1993).

(8). Quote from transcript of Michael Hammer's presentation to British Gas, TransCo, at Birmingham Metropole Hotel, November 1995.

(9). R. Swenson. Emergent Attractors and the Law of Maximum Entropy Production: Foundations to a Theory of General Evolution. Systems Research, 2, 1, 1989.

(10). As a set/superset pair, or subsystem/system/metasystem nesting. For a mathematical treatment of different levels of order and some interesting biological examples see: Clifford Grobstein, 'Hierarchical Order and Neogenesis' Chapter 2 in Howard H. Pattee (ed.) Hierarchical Theory: The Challenge of Complex Systems. George Braziller, New York, 1973.

(11). Bert Holldobler and Edward O. Wilson. The Ants. Harvard University Press, Cambridge, 1990.

(12). There is now a considerable body of evidence to support this view in the area of infant development: see Annette Karmiloff Smith, Beyond Modularity: A Developmental Perspective on Cognitive Science. (MIT Press, Cambridge, 1992) who uses the term representational redescription to describe this process. Although the empirical evidence is less comprehensive, a comparable process

(which may be the same fundamental process) is apparent in younger and older adults described by Elliott Jaques as different levels of capability which grow at different rates for different individuals.

See E. Jaques, (1976) A General Theoy of Bureaucracy, Heinemann, London (1976).

(13). Richard Dawkins, The Selfish Gene, Oxford University Press, Oxford, 1976.

(14). The distinction between environment and organism at any level of complexity is of course an artificial one and is used to try to aid clarity. Christopher Alexander calls the natural organism and its physical environment a "...biological ensemble" p16, emphasising the whole as a system of relationships. See Christopher Alexander, Notes on the Synthesis of Form, Harvard University Press,

Cambridge, Mass., 1964. Maturana and Varela (1987) op. cit. make a similar point as part of their concept of autopoiesis - they describe what they call structural coupling in which the system

 structurally co-determines a particular set of relationships with the environment.

(15). This can be seen as recursions (the unfolding of complexity) in the Viable Systems Model, see Stafford Beer, The Brain of the Firm, Wiley, Chichester, 1982.

(16). The is a considerable systems literature arguing for the irreducibility of system levels. See L. von Bertalanffy, General Systems Theorey - Foundations. Development. Applications, George Braziller,

New York, 1968. Michael Polanyi provides an instructive illustration of irreducibility in physical systems by posing the question can a machine be fully specified in terms of the principles of physics or chemistry? The answer is no - as a machine has a higher level order represented in its operational principles and component interrelationships as a machine (serving a purpose), p 328-331. Michael

Polanyi, Personal Knowled&e, Routledge & Kegan Paul, London, 1958.

(17). Richard Dawkins, The Extended Phenotype, Oxford University Press, Oxford, 1983.

(18). The change in space-time dimensions at higher levels of order or complexity can be viewed in a number of ways, for example: . in molecules as lower frequency and decreased bond strength,

Planck's Law (from the perspective of an atom amolecule appears to be standing still), in dissipative structures ie Benard cells, as increasing mean-free-path times and relaxation times. Process cycle

time in biological and work organisations might best be defined as the time taken to get feedback from the environment if a process is operating just outside its stable control limits.

(19). In the fire ant, individual ants (the lowest order process level) work to the rule 'continue hunting for a certain foodstuff if present foraging load accepted by nestmates'. The information process

is simple unmodified measure of acceptance of foodstuff or not (the exact control limit has not been identified). At the next higher order level these individual processes are aggregated into a process

of 'meeting the needs for nutrition of the whole colony'. At this level, the information process is based on demand, ie amount of a particular type of food stuff taken by nestmates over time. As demand

falls for a particular food stuff, the specification required to meet the nutritional needs of the colony automatically changes, as reflected in the shift in the aggregated individual processes (as individual

ants seek alternative foodstuffs eg oils verses carbohydrates). See Bert Holldobler and Edward O. Wilson (1990)

 (20). The importance to adaption of a high degree of autonomy between different subsystems has been noted by a number of authors. Ross Ashby, Design for a Brain, New York, John Wiley, 1952, has provided a mathematical rationale for this.

(21). Herbert Simon has noted the importance of loose vertical and horizontal coupling in the evolution of biological systems. Herbert A. Simon, The Organisation of Complex Systems, Chapter 1, in

 Howard H. Pattee (ed), Hierarchy Theory : The Challenge of Complex Systems, George Braziller, New York, 1973.

(22). More accurately the chain of command is specified to be the minimum necessary for systemic integrity - ie top down 'commands' are still necessary for: setting the broad limits (or purpose) within which the lower order level is free to operate: when emergencies have to be dealt with; and for overseeing the development of new repertoires of 'behaviour' at the next lower order level. This is reflected

in the necessary condition of minimum authority relationships.

(23). For example, Baron et al writes "...there is no exe,cutive machinery responsible for guiding the leg movement. The properties of the movement (and the properties of the perception for movement)

arise naturally form the properties of the leg's movements." pl14. Reuben M. Baron, Polemnia G. Amazeen, & Peter J. Beck, Local and Global Dynamics of Social Relations, Chapter 6, in Robin Vallacher

and Andrzej Nowak (eds.) Dynamical Systems in Social Systems, Academic Press, London, 1994. It is thought that the cerebellum facilitates sensorimotor representation and processing and that this

is at an unconscious level in cerebellar terms. E. J. Parkins, Cerebellum and Cognition: The "Unconscious Mind's Eye"? British Psychological Society Annual Conference, 1995. See also Sten Grillner,

Neural Networks for Vertebrate Locomotion, Scientific American, January, 1996.

(24). Even James Millar, who has produced a profound systemic theory of living systems, falls into this trap. See James Millar, Living Systems, McGraw-Hill, New York, 1978. This distinction is not

clear in other general systems theories eg Laszlo's natural systems hierarchy; and Boulding's hierarchy of system complexity. Renneth Boulding, The World as a Total System, Sage, London, 1985.

Ervin Laszlo, Introduction to Systems Philosophy. Harper and Row, New York, 1972.

(25). This confusion is reflected in the every-day use of terms such as emergence. For example, Ralph Stacey draws on the Kauffman's research (concerning self-organisation in evolution and Boolean networks), to make comparisons with 'emergent order' in organisations. One might question the usefulness of such an analogy which tells us little about the boundary conditions necessary for order to emerge, gives a spurious scientific feel to what might be seen as trivially obvious (ie that some kind of order always emerges when people get together), and uses the term emergence in a way that is not comparable to emergence in biological systems (as described in this paper). Ralph D. Stacey, The Science of Complexity: An Alternative Perspective for Strategic Change Processes. Strategic

Management Journal, 16. 1995, pp 477 - 495. Stuart A. Kauffman, The Origins of Order. Self-Organisation and Selection in Evolution. (Oxford University Press: New York, 1993).

(26). This difference between levels of complexity (in biological systems) and echelons (in human organisation) is reflected in the different implicit meanings of the term hierarchy found in the literature.

As John Holland (1992), one of the fathers of the science of complexity points out, "...a hierarchical building block structure utterly transforms a system's ability to learn, evolve and adapt", p169.

 Most managers find this statement surprising, to say the least. This is because they are using the term hierarchy to describe their experience of the stultifying

organisational culture that arises out of too many echelons with command control (in which levels of complexity, emergence, and control without controlling are greatly compromised). John Holland

cited in M. Mitchell Waldrop, Complexity The Emerging Science at the Edge of Order and Chaos, Penguin Books, England, 1992.

(27). For example, with extra echelons, what should be level specific information processes will inevitably become overcomplicated, impairing the organisation's ability to co-determine the environment.

(28). Peter Senge 1990 op cit, pp 293.

(29). The author has noted that public enterprises (eg utilities) seeking to become commercial run the risk of moving from being rigid to too loose – eg accountability can become the casualty of empowerment; and the safety and reputation of the business, which is founded in its core technology, can be put at risk by narrow economic analysis and the devaluation of technical expertise.

(30). See M. Mitchell Waldrop Complexity The Emerging Science at the Edge of Order and Chaos. Penguin Books, England, 1994.

(31). Michael Polanyi (1966) op cit, describes this as ".. the principle of marginal control" in which "...successive working principles control the boundary left indeterminate on the next lower level." p 41-42.

(32). Gareth Morgan, Ima&es of Or&anisations. Sage, London, 1986.

(33). Bert Holldobler and Edward O. Wilson, (1990) op. cit.

(34). F. Timothy Fuller, 'Eliminating Complexity from Work: Improving Productivity by Enhancing Quality,' National Productivity Review, Autumn,.:4,.:4, 327-344, 1985.

(35). The principle of free markets is one macro application of this criterion. However, free markets by themselves cannot fulfil this criterion (as can be seen in practice), and free market advocates have

often not recognised the need for a minimum necessary dependence (to achieve inter-dependent relationships) – I believe this can be specified in terms of certain values which must then be expressed in

‘the system'.

(36). Dying for Information; Reuters, October 1996.

(37). This can be seen in more traditional discussions of empowerment See Jeff S. Margulies, and Brian H. Kleiner, 'New Designs of Work Groups: Applications of Empowerment', Empowerment in Organisations 3, 2, 12-18, 1995. Also, this concrete interpretation can be seen in the use of biological analogy eg the notion of compartmentation, see Gerald Fairtlough, Biological Models and Business Success, Business Strategy Review,.6, 3, 1995.

(38). This is reflected in the lack of sound theory explaining how the notion of self-organised or managed teams can be applied at levels beyond front-line teams. For example, see Edward E. Lawler III,

The Ultimate Advantage: Creating the High-Involvement Organisation, Jossey-Bass, San Francisco, 1992.

(39). For every perspective there exists a higher order or metalevel perspective and so on ad infinitum.

(40). Axions are discussed in Chapter 8 in Kenneth T. Delavigne and J. Daniel Robertson, Demin&'s PrQfoung Chan&es: When will the Sleepin& Giant Awaken?, Prentice Hall, USA, 1994.

(41). Murray Gell-Mann, The Quark and the Jaguar: Adventures in the Simple and the Complex, Abacus, London, 1994.

(42). The different temporal and spatial characteristics of each level mean that specific organisational 'issues' - such as quality, safety, regulatory relationships - are expressed in a unique form at each

level (for example, quality at process level I is represented as waste; and at process level V as reputation). A deeper understanding of many organisational issues can be gained through use of levels of complexity.

(43). It might be noted that one result of overcomplication as too many echelons, is that process levels required by the organisation often seriously impaired or even missing.

(44). It is no use blaming managers for tampering when the system parameters – eg extra echelons - make it almost inevitable.

(45). Of course, the term 'horizontal corporation' coined by John A. Byrne (1993) op. cit. is a misnomer - once one gets beyond about a size of 50 people, there is no such thing as horizontal organisation

only an organisation with' fewer echelons and different kinds of lateral relationships (eg not based on ownership).

(46). J. Carlos Jarillo, Strategic Networks: Creating the Borderless Organisation, Butterworth-Heinemann, Oxford, 1993.

(47). Ikujiro Nonaka and Hirotaka Takeuchi, The Knowledge-Creating Company, Oxford University Press, Oxford, 1995.

(48). Top management is usually the last (if ever!) part of the organisation to be questioned about how much economic value it adds to an enterprise.

(49). Hoebeke, L., Making Work Systems Better: A Practitioner's Reflections, John Wiley, Chichester 1994. An alternative model of level specific information is given by Elliott Jaques. However, this fails

to fully reflect the importance of feedback information for steering processes, neglects the importance of more intangible, social information (such as perception and values). In general, Jaques'

presentation of his theory is still couched in the language of command/control. See Elliott Jaques, Requisite Or&anisation: The CEO's Guide to Creative Structure and Leadership. Cason Hall, USA 1989.

(50). A good example of developing a new level III open system is that of the Home Bakery, a new product that created a large new customer market for Matsushita when some of its other markets

were starting to saturate. The description by Ikujiro Nonaka and Hirotaka Takeuchi (1995) op. cit., indicates considerable clarity about the difference between process levels 3 & 4 (and also 5).

(51). For example, as can be seen in Deming's profound knowledge.

(52). For example, although game theory (Axelrod, 1984) provides a powerful understanding of a particular aspect of social cooperation (ie winnose situations) it is of little help in explicating the many shades of social cooperative behaviour that are not about winning or losing. R. Axelrod, The Evolution of Cooperation, Basic Books, New York, 1984.

(53). Ian Macdonald et aI, Leadership: A New Direction, British Army Review, December, 1989.

(54). That is not to say each individual can be easily make these cores values explicit on command; rather, they appear to be tacit knowledge to use Polanyi's terminology. See Michael Polanyi, (1966).

(55). Internal dialogue as a means of referring to some 'higher authority' has been shown to be an important factor in reducing the willingness of unwitting subjects to carry-out dehumanising

 procedures (eg electric shock torture in Milgram's classic experiment) on other people; ie it reduces inappropriate dependent behaviour in social conformity.

(56). One might argue that a good indicator of failing leadership is the repeated protestation that the situation is not properly understood by particular stakeholders, ie they are said to have the wrong perception! (commonly used by politicians, in recent months, to describe the public).

(57). Sunday Times, News Section, 30th March, 1997.

(58). For example, those identified by Peter Senge, 1990, op. cit.

(59). Bramley Serge, Leonardo the Artist and the Man. Penguin, London, 1994.

(60). W. Edwards Deming, The New Economics for Business. Government, Education, MIT Press, USA, 1993. For an excellent set of tools for understanding and developing ownership of processes see David Howard's FlowMap System and Only-Connect; Management-New Style, Hill House, Chislehurst, Kent BR7 5NB, England.

(61). These seven levels of complexity were originally identified by Elliott Jaques in A General Theory of Bureaucracy. (Heinemann: London, 1976). For comprehensive description as process levels see: Hoebeke, L. Making Work Systems Better A Practitioner's Reflections, John Wiley, Chichester, 1994. See also; Gillian Stamp & Colin Stamp, Wellbeing at Work: Aligning Purposes, People, Strategies and Structures. The International Journal of Career Management, 3. 1993.

(62). Luc Hoebeke 1994 op cited.

(63). E. Jaques, (1976) A General Theory of Bureaucracy, Heinemann, London (1976).

(64). Gillian Stamp, The Individual, the Organisation, and the Path to Mutual Appreciation, Personal Management, July, 1989.

(65). A detailed description of authority in relation to roles is provided by: Elliott

Jaques, Requisite Organisation: The CEO’s Guide to Creative Structure and Leadership, Cason Hall, USA, 1989.

(66). Stafford Beer has demonstrated that co-ordination is necessary to prevent oscillation. Stafford Beer, Heart of the Enterprise, John Wiley, Chichester, 1979.

(67). Ian Macdonald, Understanding Organisational Change, in Festschrift for E. Jaques. Cason Hall, USA, 1992.




The Author

Michael Church worked for 12 years as a clinical psychologist in the NHS and since the late 1980's has been an independent consultant based in Leicester. He is an Associate of BIOSS (Brunel Institute of Organisation and Social Studies, BruneI University) and has a close working relationship with David Howard of Management-New-Style.   Michael can be contacted on (+44) 07010702 134.