Images of Systems and the Systems Images

IFSR Newsletter 1987/1988 No. 2 (16) Winter

Professor Peter Checkland, foto by Dr. Stephen Sokoloff, IFSR Newsletter 1987 No. 2 (16)

Professor Peter Checkland, foto by Dr. Stephen Sokoloff, IFSR Newsletter 1987 No. 2 (16)

Excerpts from the Presidential Address to ISGSR’ Budapest June 1987
Prof. Peter Checkland
Abstract: The systems movement – a failure? To a superficial observer it might seem that the success of the systems idea has indeed been very limited. In his talk, however, Prof. Checkland showed that the real impact of systems thinking has been much more profound than is generally assumed.
A great deal of confusion, Prof. Checkland told us, results from the double meaning of the term “system”. It is used both to designate parts of reality and as an abstract concept for a relationship among things. In order to choose the appropriate method for a given problem, we have to understand what kind of system we are talking about in a particular case. Practicing clear thinking could help to advance the systems cause.
Prof. Checkland began his talk by tracing the origin of systems science. In the 1920’s the Russian Bogdanov developed a general science of organisation, which he called “tectology”. Unfortunately, because of the language barrier, his writings remained unknown in the West. Who were then the pioneers of our systems movement?
Once we have created the notion of “systems thinking we can retrospectively identify as systems thinkers historical figures as different as Spinoza, Marx and St. Thomas Aquinas. But the creation of the notion which enables us to name these historical figures as systems thinkers is recent, occuring only in the late 40’s, and its prime creator was Bertalanffy. Kenneth Boulding, who was present at the lunch table at the Centre for Advanced Study in the Behavioural Sciences at Palo Alto, at which the idea for what became the Society for General Systems Research was hatched, writes (1984) “Ludwig von Bertalanffy is rightly regarded as the founder of general systems”.
What Bertalanffy did was to point out that general ideas which he and fellow organismic biologists had developed concerning organisms could in fact be applied to whole entities (systems) of any kind. Bertalanffy was intrigued by such things as the fact that mathematically the curve of the loss of weight of a starving rat is the same as that showing the growth of the scientific literature. Surely there must be some general systems laws behind the unity of such disparate observations? (Bertalanffy, 1968).
What the Palo Alto lunch party of Bertalanffy, Boulding, Gerard and Rapoport did was to agree that a Society should be organised to investigate the isomorphy of concepts, laws and models in various fields, thus promoting the unity of science. Bertalanffy’s vision was that general system theory would be developed as a high-level meta theory, mathematically expressed, covering phenomena of many kinds.
Boulding (1984) rightly points out that:

The fact that the Society for General Systems Research has survived for thirty years suggests that there was an empty niche waiting for it

but he adds:

The niche, however, has been fairly small.

One problem is that the general systems idea pays for generality with lack of content. Given the variety in the world, all you can say about practically everything is virtually nothing. Another problem is the practical professional one: people working professionally within a subject are not easily convinced that outsiders armed with generalised topic-free models can make specific contributions to their field.
For these and other reasons there is less to show for the general systems enterprise in the late 1980’s than the pioneers anticipated 30 years ago.
This is less true, however, when we consider the use of systems thinking within other subject areas, rather than across disciplines in the way Bertalanffy envisaged. Within disparate areas of knowledge the systems image has been shown to be powerful, and systems thinking has been established as a useful tool. This can be shown by conducting a thought experiment. Imagine going into a university library and asking to be shown the systems books. You will be taken to shelves containing books on systems theory, control theory, cybernetics. But clearly the real answer to your query is that the systems books are all over the library. In the politics section David Easton’s Systems Analysis of Political Life and Karl Deutsch’s The Nerves of Government (which presents a cybernetic information processing model of government) are systems books. On the geography shelves Chorley and Kennedy’s Physical Geography: a system approach rethinks physical geography as the study of systems of four types. The section on social work will no doubt contain Pincus and Minahan’s text which treats social work systematically. The books on industrial relations will contain Dunlop’s volume conceptualising industrial relations as a system which establishes and administers rules. In this way, in most fields, systems thinking has slipped in – whether as an invited or an uninvited guest – and has made a contribution. Systems thinking has established itself as a feasible approach in most fields, and in so doing has established systems as a meta-discipline, one which can be used to discourse about the subject matter of any discipline.
Unfortunately, the process of mapping the abstract concept “system” onto aspects of perceived reality has been too successful for the good of the systems enterprise. The success has led to a ubiquitous error which pervades the systems movement and has held up the development of systems thinking in the last twenty years.
What has happened is that the success of “system” in mapping aspects of the world is such that it is used in everyday language in an unreflecting way as if it were a label word for an assumed ontological entity, like “cat” or “table”. We casually speak of “the education system”, “the legal system”, “the health care system”, “the industrial relations system”, as if all these were, unproblematically, systems. Actually you could justify the use of the words; not, “the education system”, but some such phrase as “those structures, processes, behaviours and values concerned with the provision of education which I choose to regard as a system”. It is much easier to say “the education system”. But it is still an error to use “the same word for an abstract epistemological device and for assumed ontologicai entities in perceived reality.
The error is endemic, and we have all committed it many times, from the pioneers of systems thinking onwards. Although Bertalanffy chose the word “system” as the name of the abstract epistemological device (and that was probably the crucial error) the first two words in his book General System Theory immediately introduce confusion between epistemology and ontology in the heading at the very start of Chapter 1, which is: “Systems Everywhere”. And J in the article by Boulding quoted earlier (1984, pg 4) the acceptable phrase “the world as a system” (my emphasis)
becomes in the next paragraph “the world is a total system” (my emphasis). That step from “as” to “is” is so easy to make casually, so hard to justify intellectually!
The solution to the problem, of course, would be to abandon the word “system” to everyday unreflecting use, and adopt a new word for the consciously-used abstract concept. The systems literature itself offers “org” (Gerard, 1964) “holon” (Koestler, 1967) or “integron” (Jacob, 1974), none of which has been taken up. Alas, no one loves a neologism!
Now why is this confusion between images of systems and the systems image important? It is important because it constrains and limits systems thinking, without this being noticed. The constraints are not too damaging in relation to the use of systems thinking to conceptualise objects in the natural world – such as frogs and foxgloves – nor in relation to the design of man-made objects such as cars and cow sheds. But the confusion is very damaging in relation to attempts to use systems ideas in trying to understand the phenomena of the social world, since it restricts the application of a systems approach there to a crude form of functionalism.
As a concrete’ example of the kind of problems that can arise when we confuse the two kinds of systems, Prof. Checkland mentions the debate that has been going on in Britain since the 1970’s concerning the relevance of systems thinking to operational research.
In that debate it was normally assumed that adopting a systems approach entailed the assumption that organisations were organism-like systems with departments and sections as sub-systems or sub-sub-systems. In taking part in that debate I found the protagonists consistently quoting the same source for their view of the social theory implicit in a systems approach (Checkland, 1983, Pg 673). They all quoted Silverman’s The Theory of Organisations (1970) in support of their view that systems thinking is merely a version of functionalism. Silverman’s influential but now dated book assumes that a systems approach entails taking organisations to be systems.
Now, there is nothing at all wrong with making this assumption consciously, as a chosen strategy in a research process. The error creeps in if it is assumed that it is the only systems-based strategy which could be adopted; and that error comes straight from the confusion which is the result of using “system” both as an everyday lable word and as a conscious epistemological device.
In fact, systems research in the 1970’s and 80’s has shown that there are fundamentally two strategies for using systems thinking in exploring human affairs (Checkland, 1981). The first is that discussed above, namely to make the assumption that R (perceived reality) is systemic, while M (the methodology in which concepts including systems are used to explore perceived reality, R) can be systematic. This characterises what I have termed “hard” systems thinking such as is found in systems engineering and RAND systems analysis. The second strategy, characteristic of “soft” systems thinking, and adopted in characteristic of “soft” systems thinking, and adopted in soft systems methodology, assumes that R is problematical while M can be systemic. In other words, “the System” in question is not part of the world but is the process of inquiry: that is the system which is “engineered”. Systemicity is shifted from the world to the process of inquiry into the world.
It is obvious that “hard” and “soft” systems thinking are complementary, and that the availability of both enriches the repertoire of approaches to problem solving. But the very possibility of the latter was long obscured by the characteristic error of the systems movement, namely, making the casual, unexamined assumption that “system” is a label for parts of the everyday world.
Prof. Checkland warns against a too narrow focus on general systems theory. The primary task of the movement is to find out “whether or not, and if so how, systems concepts can help our understanding of the world”.
The learning process can occur in many different subject areas, and will gain cogency from being grounded in the specific issues and problems within the various disciplines. If testable general systems theory is to emerge, its source will be the projects within different areas which have in common their use of systems ideas.
Prof. Checkland concludes that general theory should be regarded “as a welcome epiphenomenon of the learning process in multiple fields, rather than the prime goal of systems work”.

| Category: IFSR NEWS