5.2 The language of systems

Whichever definition you prefer, the term system is closely, indeed logically, associated with two other terms environment and boundary. The definition and essential meaning of these terms is straightforward. The environment of a system comprises those elements, activities, people, ideas and so on that are not part of the system but which may nevertheless be important in understanding it. System is the foreground; environment is the background, the relevant context of the system. As for the term boundary, that is basically where the system ends and the environment begins. We can therefore add a fifth part to our definition:

5 Putting a boundary around this organised assembly of components distinguishes it from its context or environment.

None of these ideas, in itself, should present any difficulty. However, their use in thinking about situations is both trickier and more rewarding than you might expect.

Messes can be distinguished from difficulties by their characteristic of being unbounded in important respects. Of course, if a problem is literally and completely unbounded it extends to include ‘Life, the Universe and Everything’. In practice things are usually not that bad. Nevertheless, there is a genuine and important dilemma: on the one hand one wants to avoid too limited and local an analysis; on the other hand, one really cannot rethink and change everything at once. The area of interest extends in numerous directions. So in tackling messy situations there is a recurring dilemma: how much one bites off. Enough to deal with the hunger pangs, but not more than those concerned can chew. But how much is that and can such a mouthful actually be separated from what is not eaten?

The language of systems does not solve this problem, but it does provide a way of addressing it. The task is essentially one of finding a workable provisional boundary for the system containing the problem, or at least a significant part of the problem. But in distinguishing between system and environment one accepts that the problem is not self-contained, that it can only be partially disentangled from its broader context. The simplest way to introduce the fruitful use of the term ‘system’ is to give an example (see Box 2).

The passage in Box 2 illustrates a number of points of considerable importance. First, the system described is not solely comprised of tangible elements – departments, people, workshops, and so on. The assumptions and norms that shape behaviour are all important elements. One cannot really understand how the ‘system’ works without understanding how things look to those inside it. Such established norms and assumptions account for many of the persistent relationships and behaviour patterns that occur.

Secondly, the complex described is not contained within any recognised institutional boundary. Although the people, agencies, activities, policies and so on, have not been chosen arbitrarily, the overall configuration is slightly unusual and probably not what you would have expected if you had thought about it beforehand. It includes a number of agencies and institutions that are clearly related, and the people they serve, but it also includes parts of other institutions as well as people who are not part of any relevant institution. It is clear that the complex distinguished ‘makes sense’, but it is not a system that is publicly recognised. In one sense, it obviously exists ‘out there’; or rather, the various elements in the ‘system’ obviously exist. But the focusing on this particular configuration of elements is the writer's own view, and, in this sense, the ‘system’ does not exist: it is simply a way of conceptualising the various elements and their relationships, which is useful for the writer's purpose.

This leads to an important distinction between two ways in which the term system is used:

1. Recognised systems that it is convenient to think of as existing ‘out there’. Such systems are widely acknowledged either because they are deliberately created (a stock control system, a computer system, for example); or because they are fairly discrete, naturally occurring phenomena that have long since been delineated and analysed by scientists (the nervous system and the solar system for example); or just because they are popularly referred to as systems in a vague though useful way (the legal system, the economic system, are examples). In general, such systems are based on widely shared perceptions.

2. Explanatory systems, such as in the phrase ‘it's the system’, whose status as entities is much more problematic. Indeed, if a system of this sort exists anywhere, it is in the mind of the individual who conceives it. It is simply a particular way of thinking about selected aspects of the world and their interrelationships which is useful in relation to the individual's concerns. Systems of this second sort embody particular points of view and are useful to the extent that they offer some insight into what is puzzling or troublesome.

You may have noticed that in making the distinction between the two ways in which the term system is used, I carefully avoided saying that systems of the first sort really do exist out there. I said that it is ‘convenient to think of’ them as existing out there, and this was in order to side-step a philosophical issue. The point is that it is rather doubtful whether the recognised systems really exist either: arguably, they too are just particular conceptualisations of phenomena, albeit ones that are widely shared. If advances in physics or physiology lead scientists to think about the movement of the planets or the functioning of organisms in rather different ways, our ideas of the solar system and nervous systems might change significantly, even though the phenomena themselves remained the same. Nor is a computer system quite the discrete entity that it first appears: does it include the software, the people who operate it, the data it manipulates, the uses to which it is put? In this sense all systems are in the eye of the beholder. On the other hand, if this argument is sound it may also be rather trivial, because it applies to everything, and not just systems. When we talk about the world we are not talking about the world as some absolute or ideal entity, but about what we mean by ‘the world’. We can never talk about ‘reality’ except within the confines and conventions of a language. Reality itself is what the term reality is conventionally used to refer to. So in this view the claim that ‘systems do not really exist’ is a bit trivial; the same is true of everything, in an extremely weak sense. Yet we know perfectly well that the terms we use map well on to those aspects of the world to which they refer – otherwise life would be very much more confusing and difficult than it is! So, following this line of argument, if the ‘system’ is a useful conceptualisation then it, too, can be said to exist ‘out there’.

In fact, it may help to think of explanatory systems and recognised systems as opposite poles on a continuum: in between are the systems perceived by groups of people with shared concerns, and those views of systems that are personal modifications and elaborations of more widely recognised systems. There is some shifting about between the poles too: the person who first coined the term ‘old-boy network’ was delineating an explanatory system, but it caught on, and has become a recognised one.

The third general point to make from Box 2 is that systems are nested within other, wider systems. Saying that ‘this’ is the environment, ‘this’ is the system, and ‘these’ are the sub-systems, of which the system is constituted, reflects a choice of the level at which you will work. Russian dolls, which fit snugly one inside another, provide a useful analogy. No single one of them is ‘the doll’; each one fits inside a larger one. Instead of trying to identify ‘the system’ it is more helpful to think of a hierarchy of systems which fit inside each other from which you have to select the system-level at which you will work by exploring the most relevant ones.

The use of the Russian doll analogy is an example of a set of techniques that can be used to explore complex situations, others being the use of metaphors, diagrams, and models. We can build up our view of the system being considered by wheeling in particular models of various systems and using them to highlight the presence or absence of particular interrelationships and patterns of behaviour within the system. It is as if we display the raw complexity of the system on an overhead projector slide and then superimpose different sorts of systems on it as overlays, to draw attention to different aspects of the way the system works and the way the system can be perceived by other people who are interested in it.

This is important because if thinking in terms of systems is to be of any use it must involve more than mentally grouping a number of elements together and calling them a system. The whole point is that these elements are interrelated, so it is important to be able to grasp the ways in which they characteristically combine and interact. An understanding of these interrelationships, of how certain elements ‘hang together’, is likely to provide a basis for deciding what to include in the system in the first place.