Soft Systems Methodology: Its Origins and Use in Librarianship

This Web document is a reformatted version of Chapter Three of Brown-Syed, C.  From CLANN to UNILINC: An Automated Library Consortium from a Soft Systems Perspective (Australia, Peter Checkland). Thesis [Ph.D.]; 1996. University of Toronto (Canada); 0779. ISBN: 0-612-11862-2. DAI, Vol. 57-08A, Page 3309, 00308 Pages.

This chapter was presented with slight modifications under the title, "Soft, Appreciative, and General Systems: Idealism in Action", at the Third Canadian Conference on Foundations and Applications of General Science Theory, "Universal Knowledge Tools and Their Applications", June 3 - 5, 1993 Oakham House, Ryerson Polytechnic University, Toronto, Ontario, Canada. This thesis is available in its entirety from UMI, and is Copyright ©, 1996. This Web version is Copyright © Christopher Brown-Syed, 2000. All rights reserved. This document may be distributed freely for educational purposes, provided that it is distributed in whole, without alteration, and that this copyright statement is attached.

Index:
 

Introduction	General System Theory	Sir Geoffrey Vickers	Soft Systems Methodology
The Seven-Step Model	SSM's Developed Form	Related Works	Bibliography

ORIGINS AND DEVELOPMENT OF SOFT SYSTEMS METHODOLOGY

This chapter outlines the Soft Systems Methodology (SSM) developed by Peter Checkland at the University of Lancaster, and notes connections among SSM, the General Systems Theory of Ludwig von Bertalanffy, and the "Appreciative Systems" envisioned by Sir Geoffrey Vickers.¹ These theories will be viewed as developments within the broad field of "holonic thinking" or "systems thinking", and as attempts to deal with complexity in organizations and ultimately, in the universe. The philosophical assumptions and consequences of adopting this view will be noted, and some of the practical benefits described.

These approaches have various notions and assumptions in common: They can be viewed as reactions to use of classical physics as a paradigm for the social sciences. Von Bertalanffy, Vickers, and Checkland all view human organizations as open, hierarchical systems, similar to those found in the life sciences. They assume the validity of the doctrine of emergence; that as systems grow more complex, properties emerge which cannot be explained in terms of simpler forms. To paraphrase Vickers, people in a crowd may behave like raindrops, but raindrops never behave like people. They are process oriented, in their approaches, and may involve several iterations of investigative processes. Finally, they may be characterized as broad epistemic approaches rather than fixed sets of methods or techniques, and while Checkland in particular has proposed a canonical set of investigative techniques, his approach allows the researcher the flexibility to incorporate corroborative or triangulation techniques drawn from systems analysis, organizational research, ethnomethodology, and other related fields of inquiry.

Soft Systems Methodology (SSM) developed by Peter Checkland and his colleagues at the University of Lancaster, and the Appreciative Systems envisaged by Sir Geoffrey Vickers, are to some degree dependent upon a philosophical stance which can be described as phenomenological and idealistic. As well, in its adoption of isomorphism among the sciences, and its belief in an underlying set of laws common to them, the General Systems Theory of Ludwig von Bertalanffy, to which these later developments are in indebted, seems in essence idealistic in the classical German philosophical sense. Finally, SSM, which may be described as an epistemological framework for the study of organizations as purposive human activity systems, may also be viewed as a species of action research, at least in respect to its fieldwork methods (Mansell, 1993).

The ideas proposed in von Bertalanffy's General Systems Theory have been acknowledged both within the systems analytical literature, and in fields seemingly far removed from computing, such as psychology, and marital and family therapy (Elkaïm, 1990). Consequently, GST is fairly well known on the North American continent, and finds a place in introductory systems analysis texts. Soft Systems Methodology, while arguably of greater influence in the United Kingdom, has been employed in various North American, Australian, and European studies and dissertations. Their topics range from fields such as agronomy, health care management, industrial psychology, and education management, to fisheries development, naval architecture, pastoral counseling and theology. This chapter also explores the application of SSM to workplace problems. In use for over two decades, SSM has been incorporated into the planning regimens of various government, corporate, and academic organizations.

General System Theory

Ludwig von Bertalanffy taught for a time at the University of Alberta in Edmonton, and at the State University of New York (SUNY) at Buffalo. In 1968, von Bertalanffy, the psychologist Jean Piaget, and others formed the "Alpbach Symposium" for the express purpose of producing a "consensus" book under the editorship of British author, Arthur Koestler.

It is Koestler's neologism, "holon", derived from the Greek holos and the "-on" suffix common to subatomic particles, which von Bertalanffy, Geoffrey Vickers, and Peter Checkland adopt as an alternative to the over-worked term "system". The main elements of Bertalanffy's thoughts about the necessity for a new social scientific paradigm are spelled out at length in General System Theory, and summarized succinctly in the Alpbach volume, entitled Beyond Reductionism. Reacting to the notion of reductionism, which he held derived from classical physics, von Bertalanffy proposed a new model of science which he called "perspectivism", based upon the life sciences.

General System Theory also rests upon the notion of isomorphism among the laws of various branches of science. Instead of attempting to reduce all scientific and social laws to a few ones derived from physics, GST views the world as an organism or an organization, and human institutions as "open purposive systems". GST incorporates various phenomenological concepts such as teleology, negentropy, purposiveness, goal-directedness. Because it views most systems as "open" rather than "closed", it challenges a particular view which equates "information" with "entropy". In von Bertalanffy's view, order is at least potentially more probable in open systems, whereas closed systems tend toward chaos. He held that viewing worldly phenomena as part of open, growing, organic systems presents a better means of dealing with complexity, and a suggests a new scientific paradigm.
 

"We may state as characteristic of modern science that this scheme of isolable units acting in one-way causality has proved to be insufficient. Hence the appearance in all fields of science of notions like wholeness, holistic, organistic, gestalt, etc." (von Bertalanffy, 1968:45)

Von Bertalanffy argued that while classical science employed basic mathematical notions such as number and species, taxonomies, relationships among elements of a system must also be accounted for. On this view, an examination of the applicability of theories, mathematical equations, or descriptions of events in different fields yields the possibility of a unified science based on isomorphy of laws in disparate sciences, suggests that there is an underlying order in the cosmos. For instance, certain mathematical curves used by Malthus in describing population growth, might apply equally to the theory of compound capital interest. While different societies may perceive and express these laws in various fashions, their a priori applicability suggests such formal laws do in fact exist. Ultimately, this gives rise to the concept of "the world as a great organization".
 

"A unitary conception of the world may be based, not upon the possibly futile and certainly farfetched hope finally to reduce all levels of reality to the level of physics, but rather on the isomorphy of laws in different fields." (von Bertalanffy, 1968:48)

Of closer concern to social science is von Bertalanffy's idea of the roles of individuals in organizations. Referring to Aristotle's dictum, von Bertalanffy elaborates: "Man is not just the political animal. He is before and above all an individual." Calling this individual "the ultimate percept", he uses biological examples to suggest that individuals behave differently in organizations than they do on their own, and organizations are dynamic.

Hence, treating organizations as static is insufficient. The history and structure of phenomena must be viewed as a continuum, and as one not entirely subservient to the arrow of time. Past events can shape current ones, but future hopes and goals can direct them. Citing the medieval philosopher Nicholas of Cusa, von Bertalanffy goes so far as to suggest that "true reality" might best be viewed as a coincidentia oppositorum - a conflict of opposites. Thus, in some measure at least, von Bertalanffy adopts the phenomenological technique of Hegel's dialectical logic, and places himself firmly in the methodological camp of historians such as the Canadian Harold Innis, who argued for a cyclical study of social phenomena.

The Work of Sir Geoffrey Vickers

Sir Geoffrey Vickers is perhaps remembered primarily for his roles in government and industry, rather than his scholarly work. Vickers was born in 1894 and served with distinction in the First World War. He served in a high ranking intelligence role during World War Two, and afterwards directed the British Coal Board (1948-55), and helped set up the National Health system.² Vickers also participated in two-year Toronto based project called the Round Table on the Impact on Human Well-Being of Rapidly Evolving Industrialization.³ Vickers is perhaps most noted for his humanistic approach to organizations, a propensity which has impressed itself upon Checkland and the Lancaster school.

Vickers acknowledges his debt to von Bertalanffy in various passages, but perhaps most clearly in the following:
 

"The difference between the inorganic and the organic world seems to be largely the difference between relatively closed and open energy systems. The whole world of experience begins to look like a hierarchy of systems; and the main task of science to formulate laws by which they interact. These changes seem to me to open the way to a better relationship between physics, biology, psychology, and the social sciences, that is , between the concepts which we use in trying to understand the processes of matter, of life, of mind, and of society." (Vickers, 1956, citing von Bertalanffy, 1952).

In other words, Vickers accepts von Bertalanffy's premise that organisms and organizations should be viewed as open, hierarchical systems, and hence, as ones exempt at least in some limited sense, "disentropic". Vickers continues:
 

"There is no reason to suppose that the laws of organization which account for the atom and the star will be sufficient to account for the cell and the elephant, let alone for the human being and the Ontario Hydro Electric Power Commission. On the contrary, there is every reason to think that these new forms have come into being through the emergence of new capacities for self- organization." [emphasis added]. (Vickers, 1956:5-7, citing von Bertalanffy, 1952).

Vickers incorporated into his thought von Bertalanffy's concept of emergence. Men in crowds, he says, may behave like raindrops, but "raindrops never behave like men." (Vickers, 1956: 9). This notion of an hierarchy of systems, each more ordered and complex, each displaying properties which cannot be reduced to those of the previous levels, logically depends upon the possibility of disentropic behaviour among living systems - upon their abilities to create conditions of stability, in fact, of organization. As well, Vickers emphasizes that open living systems must be viewed as dynamic and functional:

"We have been accustomed to think of things as existing, apart from what they do. It seems that this is a bad habit."[emphasis added]. (Vickers, 1956: 7).

So far, we have examined notions central to two authors whose works are acknowledged antecedents of SSM. These authors contribute various assumptions and notions:

a general dissatisfaction with reductionism,
the belief that organisms and organizations should viewed as open systems rather than closed ones,

a tendency to look to the life sciences rather than the physical sciences for paradigms, to organic growth rather than mechanism,

the premise that the world can best be viewed as a number of hierarchical systems of increasing complexity,

the assumption that there are certain "emergent" properties detectable at levels of increasing complexity,

an underlying humanism and a firm belief in the importance of individuals.

Their conceptions can be viewed as reactions to the use of classical physics as a paradigm for the social sciences. They are process oriented, rather than static. They view organisms and organizations as open, hierarchical systems. They assume the validity of the doctrine of emergence; that as systems grow more complex, properties emerge which cannot be explained in terms of simpler forms. It should be noted that our object here is neither to challenge the identification of entropy as a property of closed systems per se, nor to defend in toto the notion that reductionistic explanations stand in opposition to notions of purposive behaviour, goal-directedness, or teleology. Vickers and von Bertalanffy were reacting to their own conceptions of the physics of their days, and perhaps unduly to classical models inherited from previous centuries. Recent developments in cosmology, quantum physics, and related disciplines may have called their judgements into question. What is of concern to us is the fact that von Bertalanffy rejected the notion that a unified theory of science could be developed by reducing all science to a few laws derived from one of its branches.

As an alternative to reductionism, von Bertalanffy and Vickers suggested that a unified science might be based upon a certain "isomorphism" of laws which they believed obtained within all of the various discreet sciences. The models of one scientific field seemed to apply with similar success in others. Vickers believed that this line of thought would lead to agreement among the sciences and humanities. Both believed that the discovery of quantum particles which resembled energy more than matter and whose characteristics were only describable in abstract mathematical terms, had rendered the classical paradigm obsolete within physics itself. If science really ended up describing immaterial entities, the positivist or empiricist foundations of science seemed to them suspect.

Von Bertalanffy and Vickers argued that organizations and societies had been thought of in the thermodynamic terms descriptive of closed systems (particularly the law of entropy). However, since they were manifestly capable of disentropic trends like growth and increasing organization and complexity, biological organisms and human organizations might better be described as open systems, hierarchical in structure, and complex in nature. Perhaps most controversially, biological and social entities were seen as being governed by goal-directedness, purposiveness, or teleological behaviour.

Soft Systems Methodology

Several of the concepts developed by von Bertalanffy and Vickers have been incorporated into the broad notion of "systems thinking" adopted by Peter Checkland. SSM depends upon the notion that organizations and their various subsystems can best be described as "holons", and as manifestations of purposive, goal-directed behaviour

Peter Checkland acknowledges his affinity with the work on the philosophy of social science done by theorists such as the philosopher Wilhelm Dilthey and the sociologist Peter Winch, but most of all the influence of Vickers and von Bertalanffy. Both Checkland and Vickers hold the world to be a complex interaction or "flux" of ideas and events - what Vickers called a "two stranded rope". Both are concerned with the feelings, motives, and well-being of individuals within organizations, and conscious of the researcher's interpretations. Soft Systems Methodology in particular recognizes the Weltanschauungen or "world-views" of researcher and subject as both being integral to the investigation. As suggested by its adoption of the German term, Weltanschauung, this approach is certainly phenomenological, and perhaps philosophically idealistic rather than empiricist.

However, as Rennie (1989), and others have noted, its dialectic is mainly indebted to that of Dilthey, whose philosophical approach was more positivist and empirical than were those of early phenomenologists like Hegel, or later ones such as Husserl. For instance, Dilthey believed the distinction between natural and social sciences to be largely artificial, claiming that "knowledge of natural sciences blends with that of human sciences." (Dilthey, 1989;86). According to Dilthey,
 

"[A] man's intellectual life is part of the psychophysical unity of life as this human existence and life manifests it, separable from that unity only by abstraction. The system comprising these living individuals is the reality which constitutes the object of the historico-social sciences." -- ^{Dilthey, Wilhelm (1989:84)}

All three of the theorists we have been discussing were concerned with purposive, goal-directed or teleological behaviour. Extending the notion of holons to the study of nature at large raises the possibility that nature itself may behave teleologically. Arguably, this notion is implied in von Bertalanffy's notion of isomorphism. If all the sciences discover structurally similar laws, where did these laws and structures themselves come from? Such metaphysical questions remain open. The concern of the remainder of this chapter is to outline the development SSM as an epistemology, and to review its application to practical systems problems.

However, it is crucial to note that in SSM terminology, there are as many possible interpretations of perceived reality as there are perceivers, while methods of interpretation may be more or less systemic or holistic or conforming to phenomenological logic, and to selected views of perceived reality. Checkland says:
 

"SSM is a systemic process of enquiry which also happens to make use of systems models. Thus, it subsumes the hard approach, which is a special case of it, one arising when there is local agreement on some system to be engineered." (Checkland, 1990; 25).

SSM acknowledges and even seeks to exploit the processes of modeling which analysts' minds go through when examining a problem, while at the same time seeking to avoid confusing perceived realities and mental expectations. Adoption of SSM does not preclude the use of more widespread system analytical methods. Checkland has claimed that cybernetics and mainstream systems analysis should be considered special cases of systems thought, useful where there is already agreement about what is to be done.

The Seven Step Model

We will begin by describing the seven-step investigative process identified in Checkland's first book-length exposition of the methodology, Systems Thinking, Systems Practice, (Checkland, 1981). We will then describe some of the subsequent revisions of and additions to the critical apparatus of SSM made by Checkland and other practitioners over the past two decades, with reference to various SSM theses and studies.

Soft Systems Methodology was first envisaged as a seven-step process of inquiry. The preliminary steps of this "classical" SSM consist of "identifying a problem situation considered problematic", and "expressing the situation in plain language". These steps are rooted in the empirical. The intermediate steps, called collectively "systems thinking about the real world", consist of "developing root definitions" of associated purposeful activity systems, which give rise to "conceptual models of potential systems" (i.e. holons). The process concludes with a "comparison stage", whereby those models are held up against the real world situation. (See Figure 1). Out of this comparison may emerge "changes systemically desirable and culturally feasible". In the best case, these suggestions result in actions to be taken to improve the problem or situation. (Checkland, 1990; 27).

Both in its original form, and in its current one, SSM centres upon transforming processes - upon the conversion of some input to some output. To illustrate this notion in the context of a library network, one might consider as "transformations", system activities such as "converting catalogue records in the Australian National Library to catalogue records accessible over the network's local online public access catalogue." A computer systems manager in such a network might propose another possible transformation, such as "converting services potentially available to users to ones actually accessible during certain hours." From the viewpoint of a particular library within the consortium, another transformation activity might consist of "converting potentially shared bibliographic resources to resources actually exchanged among members." As each category of participants - designers, managers, users, or other beneficiaries - may have different views of the organization's main goals or activities, many other transforming processes will be identified, and many other viewpoints elicited during an SSM investigation. The major aim of SSM fieldwork is to elicit these various notions, which participants may work with subconsciously or unreflectively on a daily basis, and which constitute the real business of the organization as perceived by the various owners of, actors in, and beneficiaries of, or even victims of, its ongoing processes.

Once these activities have been discerned and expressed, they can serve as the basis for formally expressed root definitions from which conceptual models of systems can be developed. By examining the most relevant of these, it is possible to develop an overall notion of a system in terms of some purposeful human activity. This process is accomplished by examining systems in terms of their owners, actors, beneficiaries, transformation processes, environments, and the world-views or Weltanschauungen of those involved. The investigation may help determine which activities are essential to the organization, and which are superfluous. As well, since many perceptions are elicited, potential conflicts may be uncovered, or potential areas of commonality revealed.

At base, "transformations" consist of the conversion of some input to some output by means of some transforming process. Not all of these transformations will have been envisaged by a system's designers or an organization's managers. Quite often, a study may be indebted to non-traditional viewpoints and formulations which evolve during the analysis process. Thus, while "transforming books on library shelves to ones out in the community" might be envisaged as a main activity of libraries, Checkland (1990; 34) can talk about the conversion of "books" to "dog-eared books" as another outcome of their activities. A similar, unforseen outcome is mentioned by Lynda Davies, a Lancaster researcher whose SSM project of the British Army's computing services was terminated by management. At one point, Davies investigation suggested that the Army's training regimen could be described as "a system designed to perpetuate a chronic shortage of staff." (Davies & Ledington, 1990).

SSM seeks to describe an organization's key transformation activities, to identify the major groups of participants in the process, and to elicit the world-views or belief-systems under which they operate. These elements are expressed in terms of a basic schema or analytical framework. That framework has remained at the core of the methodology, despite subsequent revisions to the original "seven-step" model, and despite the incorporation of field techniques borrowed from other types of inquiry, such as ethnography.

The basic schema used to categorize a situation in SSM can be expressed in terms of a mnemonic - "CATWOE", which Checkland (1990;35) explains as follows:
 

C customers the beneficiaries or the victims of T

A actors those who would do T

T transformation process the conversion of input to output

W Weltanschauungthe world view which makes T meaningful in context

O owners those who could stop T

E environment elements outside the system which it takes as a given

"The core of CATWOE", explains Checkland, "is the pairing of transformation process T and the W, the Weltanschauung or world picture which makes it meaningful. For any purposeful activity, there will always be a number of different transformations, by means of which it can be expressed, these deriving from different interpretations of its purpose." (Checkland, 1990; 35).

In classical SSM, (Checkland, 1981), as well as in the "developed form" of the methodology outlined recently (Checkland & Scholes, 1990), this CATWOE schema is used to construct "root definitions" of organizations, which identify all the principal participants and describe both the outcomes of activities and the belief-systems under which participants operate. To continue with the example drawn from librarianship, we might construct a root definition of an automated library consortium:
 
 

"A consortium owned and operated system to transform potentially accessible information at individual libraries into information being used system wide in research and teaching, in a resource-sharing environment...."

Once an organization's major transformational activities have been expressed in concrete terms, the analyst's task is to consider, from a reflective stance, what an abstract or "possible" system to perform such transformations might look like. Checkland stresses that in SSM, this reflective activity is aimed at producing an ideal, pure, or abstract model - not a mirror of the system as it actually exists. In the classical seven-step model, this reflection was described as "below the line" thinking, as taking place in the mind of the analyst, rather than in the real world of the problem situation. The temptation to describe what actually exists rather than what owners and actors seek to do, must be resisted during the model-building phase. There is a sharp division between the "real world" of the problem situation and the "systems thinking world" in which models of potentially useful systems arise.

Such systems may then be used to bring about changes to ameliorate difficult situations, or to enhance an organization's effectiveness. Such changes should be both "organizationally feasible" and "desirable". That is, they should be made when one is fully cognizant of the complexity of the situation, the organizational culture, the potential or actual conflicts among members, and so forth.

In the spirit of Vickers, SSM rejects economic rationalism as the sole criterion for management decision making. Instead, it suggests that decisions should be measured against five (originally three) criteria, known as the "5Es":
 

EfficacyDoes the means [i.e. the proposed technical process] work?

EfficiencyAre minimal resources being used to produce the desired outcome?

Effectiveness Is the desired outcome being produced?

Ethicality Is the action fitting, moral, etc?

EleganceIs the result aesthetically pleasing? (after Checkland & Scholes, 1990)

The 5Es can be applied both to situations under investigation and to the actions proposed for their amelioration. As well, they may be viewed as the set of ethical norms against which the analyst's intervention can itself be measured. It is perhaps in the adoption of these criteria that SSM's debt to the thought of Vickers' humanism is most evident, for they reflect that ethos of concern for human beings - workers and management alike - which so characterized his thought.

SSM's "Developed Form"

Since the publication of Systems Thinking, Systems Practice, the number and variety of SSM studies has continued to grow. Librarians, systems consultants, and academics from South Africa to Australia have employed the methodology. In particular, it has enjoyed considerable success in Australia - notably, in Western Australia, where it was introduced by Bob Galliers, (now head of Warwick University's business school). Under Galliers' direction, a Perth-based project sought to incorporate techniques drawn from the study of decision support systems into the SSM process. Today, under the direction of Lynn Allen, the state's chief librarian, SSM is viewed as a basic part of the managerial apparatus for the Library and Information Service of Western Australia (LISWA). (Galliers, ####, Allen, 1992).

Lynn Allen, State Librarian of Western Australia and head of the Library and Information Service of Western Australia (LISWA), explains in a 1989 paper delivered at the Paris conference of IFLA that year:
 

"In the development of plans of any type... I prefer it for its conceptual thinking and its appropriateness as a generalised problem solving methodology. You can add other techniques to it depending upon your preferences ... Edward De Bono's lateral thinking, problem based planning, data modeling and so on. These techniques can all be used at some point within Checkland's methodology." (Allen, 1989;3).

Describing LISWA as a service provider to 221 public libraries throughout the state, Allen described the utility of the methodology in the ongoing review of LISWA's operations. Among its strengths, Allen notes SSM's ability to force recognition of environmental factors, to force conceptual or `blue-sky' thinking, to split the ideal neatly from the realities of a situation, to concentrate upon those things for which you have sufficient information to conduct a gap analysis, and because its very flexibility helps managers to avoid `recipe-book' approaches to corporate planning. (Allen, 1989;6).

LISWA's corporate mission statement was developed from root definitions produced during SSM analyses. However, Checkland (1991) cautions that SSM is not merely a scheme for creating mission statements, and Allen agrees. Describing SSM as a technique "continuously available to managers", Allen has encouraged the methodology's ongoing use as part of organizational decision making within LISWA. To assist managers in this effort, LISWA has devised program and subprogram definition forms, which concentrate upon objectives and outcomes, and take into account performance indicators and ongoing evaluation criteria. (Allen, 1989; Appendix). These programmes and subprogrammes correspond to the human activity "holons" and operational or control "subholons" prescribed by SSM.

By 1988, the Australian SSM practitioners, Watson and Smith, were able to identify seven major variants of the methodology, and to describe their use in eighteen separate studies which had employed the approach in that country. (Watson & Smith, 1988; 4-7). By the end of the decade, Checkland and Scholes felt it necessary to set out minimum criteria for orthodoxy, "not so that miscreants can be struck off some imaginary list of SSM users [...but...] to enable coherent critical debate about the experience of SSM to take place." (Checkland & Scholes, 1990 : 286). They also acknowledged (their own and other people's) concerns about the ridigity of the original seven-step method.

In this more recent monograph, Checkland and Scholes stressed that SSM could be thought of as conforming to two "modes". Mode 1 pertains to a single, discreet, or one-time "intervention" into the real world, while Mode 2 refers to the use of SSM as an ongoing epistemological technique useful to scholarly, managerial, or ongoing design and planning activities. An example of SSM's use in "Mode 2" can be drawn from the Western Australian library service, whose director suggests that SSM be "available at all times" to planners and managers. (Allen, 1989).

In fine, stressing the primarily epistemological nature of SSM, Checkland and Scholes conclude that Mode 1 is merely a special case of Mode 2. While noting that there will be as many variants of SSM as there are researchers, they present five criteria of orthodoxy, emphasize two principal modes of application, and develop a canonical glossary of applicable terms.

In summary, SSM can be viewed as an epistemological stance, with its own special vocabulary. Any "full" application of SSM should involves structured thinking about real- world situations, which should acknowledge that although the real world may be complex or appear chaotic, its investigation may be systemic. It should distinguish clearly between real world situations and systems models of them, and should describe systems using "holon" models. Each practical application of SSM should contribute to the theoretical apparatus.

Checkland & Scholes assert that "developed" SSM is a structured way of thinking about a real-world situation, whether ongoing or through an individual, special study. It is to be viewed as an "explicit epistemology" with a "corresponding terminology" according to which, [at least potentially], any valid SSM application can be described. To claim to be using SSM in its "full" form, a study must acknowledge that the real world is not necessarily systemic, distinguish carefully between unreflective participation in the "unfolding flux of events and ideas" and conscious reflection upon it using systems thinking. Practitioners should construct "holons" incorporating the basic ideas of "emergent properties, layered structure, processes of communication and control," and use those holons to explore changes "deemed desirable and feasible". (Checkland & Scholes, 1990).

On this view, each SSM study must include thought about applying the methodology to the situation in hand. Finally, each application will necessarily yield lessons about the methodology itself, as well as about the situation at hand. This is inevitable, since "SSM is methodology [the logos of method], not [merely a specific] technique". (Checkland & Scholes, 1990 : 286-7). Together with the ever expanding terminology, Checkland and Scholes suggest, these few rules are sufficient to define SSM well enough to facilitate its "coherent discussion". In its developed form, SSM need not adhere strictly to the seven-step process.

Although SSM itself is to be viewed as a methodology or as a branch of thought, the authors acknowledge that its practice has yielded a number of useful techniques. For instance, in my own recent research, the activity of developing "rich pictures" was found particularly effective, both in eliciting knowledge and in developing root definitions and conceptual models. Expressing the various elements of a situation using diagrams or drawings, even or perhaps especially "messy" ones, allows the opportunity to visualize overlapping aspects of a situation, to identify inconsistencies and relationships, to revise emerging ideas "on the fly".

Drawings made in the field, however, present a slight drawback for readers - on the whole, the more information they are made to contain, the messier they become. Some of the really productive drawings are likely unsuitable for publication, either because they are unsightly, or because they contain compromising information and speculation about the situation and its participants.

In its developed form, SSM may involve using various ancillary techniques, schemata, and sets of criteria against which to evaluate systems. These include a set of three analytical steps, and a way of splitting an investigation into two streams of inquiry. The three forms of analysis are:
 

Analysis 1; an examination of the intervention interaction in terms of the roles of "client", "problem solver", and "problem owner", and an identification of plausible roles, selected by the researcher, from which the problem can be viewed.

Analysis 2; an examination of the social or cultural characteristics of the problem situation by means of interacting roles, norms, and values.

Analysis 3; an examination of the power-related or political aspects of the problem situation. (after Checkland & Scholes, 1990).

As well, two separate streams or courses of analysis have been suggested, the first, a Cultural Stream of Enquiry which explores the social roles, norms, relationships, and patterns of communication within an organization. The second, a Logical Stream of Enquiry would deal with organization's functional aspects. A cultural stream inquiry seeks to identify the roles and norms, rewards and punishments, and similar social aspects of an organizational setting. A logical stream of inquiry would concentrate upon technical methods and processes, and would more closely resemble classical systems analysis.

This dissertation attempts to strike a balance between culture and logic by corroborating field interviews with evidence obtained directly from the computing system's internals. In conducting the investigation, every attempt was made to maintain clear distinctions between the real world situation and the systems-thinking domain spaces. The classical CATWOE schema and explicit Root Definitions were used.

However, this dissertation stops short of recommending, much less implementing desirable and feasible changes in the workplace. It terminates in the reflective, model creating phase - the sphere of "systems thinking about the real world". This is because the research was not undertaken in response to some existing problem for which a specific solution was desired. To the extent that SSM's conceptual models are specifically intended to represent no existing system, but on the contrary, to delineate the requirements for a generalized system to perform the desired function, the methodology presents an potentially valuable framework for academic inquiry rather than as "action research" per se.

Since SSM is an exploratory methodology which depends for its ultimate shape and direction upon its own evolving process, it neither possible nor desirable to identify hypotheses for testing before the study begins. In fact, Checkland stresses that science proceeds in two ways: through hypothesis testing and through experiment or reasoned problem solving. This contention is supported by some current discussions of the nature of scientific thinking. For instance, Paul Thagard discusses science as a problem solving activity in Computational Philosophy of Science (1988), while MIT's Marvin Minsky tends to speak of learning itself as a problem-solving activity.

Use of SSM in Other Dissertations and Studies

Since the publication of Checkland's monograph, Systems Thinking, Systems Practice, SSM has been applied widely as a methodology for scholarly research. The publications which have resulted from SSM research undertakings include doctoral dissertations, master's theses, monographs, consultants' reports, and journal articles. As well, SSM has been adopted as an ongoing management tool by organizations such as the Library and Information Service of Western Australia.

As one might expect, numerous dissertations employing SSM have originated within the University of Lancaster. Robert Galliers will serve as an example of a scholar whose doctoral research involved SSM, and who has gone on to develop and promulgate the methodology. Now of Warwick University, Galliers helped introduce the methodology to Western Australia, where he explored its use in conjunction with Decision Support Systems. Lynda Davies and B. Ledington, Lancaster graduates who teach at Griffith University and the University of Queensland, have co-authored a concise and practical SSM textbook.

Graduate theses and dissertations which employ or discuss SSM have included: (a) historical, philosophical, or evaluative examinations of the methodology itself, (e.g. Forbes, 1989; Casar, 1989), and (b) applications of the methodology to problem situations within organizations, (e.g. Ledington, 1989 Fraser, 1993).

While University of Lancaster scholars have been prominent in the field, much SSM research originates elsewhere in Britain and the European Community, and in countries as geographically disparate as Australia (Watson, 1989), Canada (Bourassa, 1988; Rennie, 1989; Cubero, 1991), Finland (Linfors, 1992), South Africa, and the United States (Cook, 1987; Summers, 1988; Parker, 1990; King, 1992; Koble, 1992). The ranges of disciplines to which the methodology has been applied is likewise diverse. It includes various subdisciplines of psychology (Bush, 1985; McBeath, 1986; Powers, 1987; Bourassa, 1988; Frantz, 1989), problems in the management of various industries (Scholes, 1986; Filion, 1988), agriculture and agronomy programmes (Summers, 1988; Mills-Packo, 1989), geography and fisheries development (Rennie, 1989), national health services (Holloway, 1990; King 1992; Marcias Chapula, 1992), government information services (Cubero, 1992), community development (Cook, 1987), and engineering problems such as the design of flight systems for military pilots (Koble, 1990).

Bourassa's 1988 University of Montreal dissertation applies SSM in a study of children with communication difficulties. (Bourassa, 1988), while Frantz (1989) applies SSM in a study of anxiety management in organizations. Mills-Packo used SSM in a 1989 dissertation on agrotechnology in Hawaii (Mills-Packo, 1989), and Summers applied the methodology in a study of agriculture in Thailand (Summers, 1988). Cook, (1987), used SSM in a city planning context. Powers (1987), Filion (1988), and Tedder (1986) used SSM in management research dissertations. Miles (1987), and Koble (1990), discuss the use of SSM in systems analysis and software development. Bush (1985), employs SSM in a study of a professional support institution for religious pastors.

By 1994, Dissertation Abstracts International, listed about 30 graduate theses which involved SSM, and many which explored aspects of von Bertalanffy's General System Theory, acknowledged Sir Geoffrey Vickers' Appreciative Systems, or which mentioned "systems thinking". Several of these projects involved the use of SSM in conjunction with ethnographic techniques, conventional systems analysis, operations research, and similar widely-known research strategies. SSM has been used in conjunction with other methodologies or procedures, such as those associated with decision support systems, the Viable Systems models of Stafford Beer⁴, prototyping, cybernetics, and ethnographic or Operations Research methods.

In a 1993 dissertation, Ian G. Fraser employed SSM in a study of resistance to change in the Australian Department of Marine and Harbors. The study sought to "categorise reasons for change resistance, and... to hypothesise on procedural pathways which might be followed to avoid similar occurrences within other organizations." The study identified difficulties in communication, conflict of values among stakeholders, and other management issues which were deleterious to operations, including "[V]alue conflict and ..... a lack of credibility of management staff in the eyes of blue collar workers." (Fraser, 1993).

A 1992 thesis from Finland, employs SSM in a study of handicraft education. (Lindfors, 1992). This study employed participation/observation and case study methodology, formal and informal interviews, and questionnaire data, analyzed according to the SSM schema.

Several projects have involved studies of health care provision in various countries. A University of Hawaii researcher, Stephen W. King used SSM in an investigation of district health care in Thailand (King, 1992). King notes dissonance between the worldviews of researcher and subjects, and language barriers as drawbacks to the approach. A 1990 study of the performance evaluation mechanism in the National Health Service in the United Kingdom, (Holloway, 1990), contrasted Checkland's methods with 'hard' systems approaches and focused upon the outcomes of health care as criteria for evaluation, as well as four other management topics. Cesar Macias Chapula's 1992 dissertation investigated the impact of scientific and technical information upon 36 physicians who played regional coordination roles within the National Health. The author contends:
 

"Soft Systems Methodology was useful both (1) to tackle information problems at the structure level of health care; and (2) to enrich the different concepts of human activity systems that participate in the delivery of health care at the structure, process, and outcome levels." (Macias Chapula, 1992).

Other dissertations have focused upon the methodology itself or upon its philosophical underpinnings. Within the University of Lancaster, several studies have explored the use of SSM as a corroborative or adjunct technique to be used in combination with other methodologies, or have described the philosophical, historical, or operational aspects of the development of the approach, (e.g. Forbes, 1989; Scholes, 1986). Among the book-length studies of the development and methods of SSM are works by Checkland & Scholes (1991), Davies & Ledington (1990) and Galliers (1992).

A 1988 Lancaster dissertation investigated the use of SSM in conjunction with Operations Research (OR) methods, (Lin, 1988), in an aim to improve the interaction between researchers and subjects. Suggesting that each methodology complemented the other, Lin's thesis proposed a new approach termed the 'Effectiveness Control Approach', linking OR methodology and SSM.

In another Lancaster dissertation, Casar (1989), explores the development of the "systems thinking" and "appreciative systems" concepts of Sir Geoffrey Vickers, and their incorporation into Lancaster SSM, by applying a systems model in an industrial setting.
 

"As Checkland has claimed, all our experience shows that Soft Systems Methodology shares with Vickers an epistemology which subverts that of the 'hard' systems thinking paradigm. The complementarity of both approaches shape a trend within the Soft Paradigm, one which now has a powerful mean for accounting for social processes." (Casar, 1989)

Regardless of the immediate import of SSM studies for the organizations under survey, applications of the methodology invariably lead to introspection and suggestions for its further development. Hamish Rennie's 1989 Memorial University study of the North Labrador fisheries both put forward practical recommendations concerning the industry in question, and discussed the philosophy and application of SSM, contending that the methodology provided a "richer expression [of the problem]... than would have been expected...". He cautions:
 

"The philosophic analysis suggests that the methodology would be more soundly based on Husserlian phenomenology than on the mixture of positivist and existentialist philosophies Checkland espouses." (Rennie, 1989).

Some studies began with practical situations at hand, but resulted in major methodological suggestions. James Scholes' 1986 Lancaster dissertation set out to study the British telecommunications sector, but its major effect was to contribute to an understanding of SSM itself:
 

"The studies constituted a programme of research in which the author was a prime actor rather than an external analyst. The intention was to derive lessons about SSM from its application--rather than test an hypothesis. The main lessons were not of the kind envisaged (which might, for example, have commented on the internal phases of SSM). Instead, the research led to new ways of thinking about SSM." (Scholes, 1986).

Sholes contends that, rather than merely offering practical and particular solutions, (as hard systems analysis purports to do), SSM "offers a meta-level model of the process of making a selection, predicating and so forth."

Despite the intimate connection between researcher and subject noted by Scholes, results of SSM analyses may not always be supportive of the status quo. In a 1991 Toronto dissertation, Flor Cubero Venegas sets out to investigate the information needs of policy makers in Costa Rica, particularly those concerned with exports. She notes:
 

"As a result of the methodology used a shift of focus occurred during the final stages of the research when it became clear that there were political problems with the creation of the Ministry of Foreign Trade, the atomization of institutions in the sector and the legal framework that were hindering the flow of information amongst institutions in the sector." (Cubero Venegas, 1991).

Peter Nielsen's 1990 Lancaster dissertation addressed the choice of methodologies appropriate to different situations, and concentrated upon SSM, which he viewed as a species of action research, in a "dynamic and situational approach" to individual and project-level problems. Ledington's 1989 dissertation concentrated upon intervention process itself, proposing a "conversations framework" for analyzing interactions, which emphasized positions, power lines, and the interactions among personalities. The study included guidelines for managers, and "an expanded analysis of the process of 'action-based' research."

Finally, Paul Forbes described the history and dissemination of SSM in a lengthy and philosophical Lancaster dissertation (Forbes, 1989). Aimed at generating suggestions for the further development of SSM, the study included "an hermeneutic interpretation of the historical documents" created within the Department of Systems at Lancaster, an analysis of contributory theories such as Vickers' Appreciative System, Kuhn's theories of scientific knowledge, Rogers' works on the dissemination of innovations, and Kolb's work on learning.

To this point, we have discussed several academic theses and dissertations which have either used Soft Systems Methodology, or which have analyzed and evaluated it. We have also mentioned a few instances of its use within the field of librarianship, notably in Australia. These concentrations reflect the focus of the current dissertation. However, no discussion of the literature of SSM would be complete without mention of the Journal of Information Systems. Begun under the editorship of Peter Checkland, with the title, Journal of Applied Systems Analysis, this publication continues to serve as a forum for practioners of SSM, and well as for technical and theoretical articles about SSM in particular, and the systems movement in general. Articles have ranged from discussions of particular projects, to treatments of particular techniques such as rich pictures and conceptual diagrams, to explanations of SSM terms such as "Weltanschauung". This journal remains the locus classicus for practitioners or theorists interested in the methodology.

¹This chapter was presented with slight modifications under the title, "Soft, Appreciative, and General Systems: Idealism in Action", at the Third Canadian Conference on Foundations and Applications of General Science Theory, "Universal Knowledge Tools and Their Applications", June 3 - 5, 1993 Oakham House, Ryerson Polytechnic University, Toronto, Ontario, Canada.

²"Briton gives science lecture". The Varsity. Friday October 26, 1956.

³"Industry can frustrate enjoyment of luxuries, U.K. scholar warns." Globe & Mail, October 23, 1956.

⁴For a discussion of Beer's viable systems models, see de Raadt (1990).
 
 
 

BIBLIOGRAPHY
 

Copyright © Christopher Brown-Syed 1995-2001. Disclaimers.