Enriching the model-building phase of soft systems methodology. (Research Paper).
INTRODUCTIONThe importance of worldviews, perceptions and conceptual tools in design situations is becoming increasingly acknowledged as the view on design as a mental activity is recognized and emphasized (Dahlbom, 1992; Fairtlough, 1982; Mirijamdotter, 1998; Stolterman, 1991). While the soft systems thinking approaches, in general, are characterized by a subjectivist approach to social science, there is one approach in particular, Soft Systems Methodology (SSM), that has paid special attention to people's perceptions of reality, their worldviews, and how these perceptions and worldviews affect design situations.
SSM aims at extending the area in which systems ideas can be used to find a structure, and hence help improve or solve real world problems in the realm of social systems; SSM has, as such, been argued to be `the most self-conscious (and certainly the most rigorous) attempt at an interpretive systems methodology' (Jackson, 1982 p. 22). This is done by taking seriously the subjectivity which is the crucial characteristic of human affairs and to `treat this subjectivity, if not exactly scientifically, at least in a way characterized by intellectual rigour' (Checkland, 1981, p. 30). This atempt is visible throughout the methodology and based on the assumption that individuals interpret situations differently and in accordance with what they find meaningful. Further, what is perceived as meaningful is based on our particular "image" (Boulding, 1956b) of the world, or Weltanschauung (Checkland and Davies, 1986), which in turn is formed by individual preferences, background, knowledge and experience, and which, in general, we tend to take for granted. SSM teases out such world-images and examines their implications. Because of this, the concept of Weltanschauung is argued to be the most important in the methodology (Checkland, 1981, p. 18).
However, in order to handle the complexity that a person's Weltanschauung implies, the content of the term has been reduced quite drastically in the model-building part of SSM. Instead of including the whole Weltanschauung of a person, Checkland suggests that it is enough to try and draw out the underlying rationale of why a person finds it meaningful to carry out a certain activity or transformation. Even if this is most helpful, it has been argued by Bergvall-Kareborn and Grahn (1996a) that the concept has been too narrowly restricted to do full justice to the diversity found in different perceptions. Mingers points to another weakness related to the concept Weltanschauung by arguing that when stressing the difference between different Weltanschauungs, SSM takes them at face value and does not analyse the similarities between them or seek an explanation for this in terms of the structure of society: `It therefore lacks an explanation of why these particular W's have developed and thereby how they might be changed' (Mingers, 1980, p. 48).
It has also been argued by many authors (Jackson, 1982; Mingers, 1980, 1984; Naughton, 1979; Prevost, 1976; Schregenberger, 1982) that the methodology has a tendency to result in conventional and regulatory, rather than radical, agendas for change. Some authors (Jackson, 1982; Mingers, 1984; Prevost, 1976) argue that this is an inherent characteristic of the methodology due to its functionalistic and/or subjective character. Jackson (1991), for example, argues that there is no attempt within SSM to make sure that conditions for genuine debate can be provided. Because of this `[t]he kind of open, participative debate which is essential for the success of the soft systems approach, and is the only justification for the results obtained, is impossible to obtain in problem situations where there is fundamental conflict between interest groups which have access to unequal power resources' (Jackson, 1991, p. 133).
A contrary view (Checkland, 1982; Mingers, 1980; Naughton, 1979). is that this remains unproven and that it is equally likely that the regular tendency stems from the context within which these studies have been carried out, people's inherent resistance to change, or lack of insight and originality on the part of those using the methodology, rather than from an intrinsic defect in the approach itself. As an example of the last reason Naughton points to his own experience and says that when making the transition from root definition to conceptual models `inexperienced analysts invariably plump for bureaucratic-type models' (Naughton, 1979, p. 70).
I agree with Checkland and Naughton and argue that there is nothing in the methodology per se which forbids or makes radical changes impossible. Further, while Jackson (1982) is quite right when he says that there is no real attempt within SSM to make sure that conditions for genuine debate can be provided, and that debates characterized by constrained discussion can prevent radical changes from being expressed, I cannot find any methodology that has succeeded in eliminating differences between stakeholders related to power, recourses and intellectual capacities. Jackson is also partly right when he says that in social systems `political or economic factors often act as the main catalyst of change' (Jackson, 1983, p. 113). However, I argue that these changes, even when they have their source in political and economic changes, reflect changes in Weltanschauung among people and groups, and that many times it is changes in values and attitudes that drive the political and economical changes, rather than the other way around. Just think of the abolition of slavery in North America, women's right to vote, and the more recent opinion against circumcision, or maiming, of young girls in Africa. These are all, I would argue, radical changes which have come about due to changes in people's Weltanschauung, and in settings characterized by constrained discussion.
Hence, I sympathize with Jackson in his endeavour to achieve an open and participatory debate characterized by `unconstrained discussion' (Jackson, 1982). However, I think this endeavour should be viewed as a vision rather than something all methods and methodologies need to achieve or fulfil before we can use them, otherwise we will find that we have very few methodologies at our disposal. I also argue that, while a constrained discussion can prevent radical changes from being expressed, unconstrained discussions in themselves are no guarantee for radical change.
Nevertheless, since many people perceive the methodology as regulative and argue that in practice radical changes often do not occur as a result of using the methodology, it is important to see how this can be changed. In this paper, I will do this by concentrating on the second reason for SSM's tendency towards conservative and regulative changes by exploring one way of helping the users of SSM to break away from traditional ways of viewing an issue or problem situation. I will also explore if it is possible to enrich the concept of Weltanschauung, used in the modelling phase, without making it too complex to deal with in a real-world situation. This is done by introducing a concept termed `qualifying function' (Dooyeweerd, 1997). The concept, which aims at identifying or defining the leading or guiding function for a thing or activity, originates from a Dutch philosophy developed by Herman Dooyeweerd, here referred to as the Dooyeweerdian philosophy.
Thus, the aim of the paper is to address some of the critical comments made regarding SSM by incorporating selected theories and ideas from the Dooyeweerdian philosophy with the aim of strengthening SSM, not to try and combine Dooyeweerdian philosophy, as such, with SSM. The paper should also be seen as a continuation of the multimodal developments of SSM, as set forth by Bergvall-Kareborn and Grahn (1996a, 1996b) and Mirijamdotter (1998), and has the following structure. It starts with a description of SSM, followed by a presentation of the philosophy of Herman Dooyeweerd, with a focus on the theories related to the concept of qualifying function. After that, the benefits of incorporating the concept of qualifying function will be discussed. Finally, the paper ends with a summary and conclusions.
SOFT SYSTEMS THINKING
SSM `is a cyclic learning system which uses models of human activity systems to explore with the actors in a real-world problem situation their perceptions of that situation and their readiness to decide upon purposeful action which accommodates different actors' perceptions, judgements and values' (Checkland, 1984, p. 98).
The methodology evolved through an action research programme at Lancaster University (Checkland, 1979b; Checkland and Jenkins, 1974), starting at the end of the 1960s, as a reaction against the inability of contemporary management science in handling complex real-world problems (Checkland, 1970, 1972, 1978, 1981, 1984, 1994; Checkland and Griffin, 1970; Checkland and Haynes, 1994).
According to Checkland (1970), Management Science was at that time characterized by extreme confusion, and there where two main sources of this. Firstly, there was considerable confusion among the practitioners of the many different varieties of Management Science. Secondly, there was confusion due to its lack of impact on the day-to-day reality of management. There was an enormous gap between what was theoretically possible, using all the techniques of Management Science, and what was actually put into practice.
While Checkland believed that the systems approach, in general, could help resolve the first confusion by revealing the pattern of the whole of Management Science and provide a way of structuring it, the general systems approach could not, at least not at that time, bring possibilities and actualities closer together. Systems Engineering, on the other hand, being the practical manifestation of the systems view, could. Systems Engineering, in the Lancaster sense, was `an attempt to use the insights of systems thinking to solve real problems, and to develop the subject by means of this close involvement in actual management problems rather than by pursuing the mathematics of General Systems Theory' (Checkland, 1970, p. 9).
Besides the above problems related to Management Science, Checkland also argued against the goal-seeking model of human action found in these theories as well as in traditional organization theories. Here, the manager is viewed as a purely rational decision-maker, pursuing organizational goals that often provide the standards against which progress will be judged.
Thus, in order to find a complement to contemporary management theories, Checkland began to investigate whether systems thinking approaches of that time, like Systems Analysis and Systems Engineering, could be used. This was done by studying what happened when these methodologies were applied to `soft' problems, such as those of policy-makers, administrators and managers. It was especially the methodology described by Jenkins (1969) that constituted the starting point for SSM in this work. However, these approaches were also found to be inadequate for managerial real-world situations due to their emphases on structured problems, and hence on finding efficient means of achieving known and defined ends. This was an inappropriate focus for managerial problems, characterized by Checkland as ill structured, fuzzy and `soft' and where the real difficulty lies in defining the problem itself (Checkland, 1979a).
Besides this, these approaches, later referred to as `hard systems approaches', also shared management science's view of reality as objective, neutral and value-free as well as the goal-seeking model of human behaviour. Checkland reacted against this and instead pointed out that people interpret situations differently, depending on what they find meaningful. What is perceived as meaningful is dependent on an individual's background, pervious knowledge, experience and so on. Therefore, a situation perceived as problematic by one person need not be interpreted accordingly by another. Further, depending on the way we interpret a situation, we form intentions; i.e., in the light of our interpretation we decide to do one thing rather than another. To explain human behaviour it is necessary to know the underlying concepts of intentions that were acted upon; we need to know the underlying purpose. By bringing these things into the open we can also better understand and learn from each other. This also points to the fact that learning is personal and needs to be adjusted to the knowledge and experience of each individual if it is to be really helpful. Furthermore, real-world managerial problems are multifaceted and cannot be pressed into predefined form; they are characterized by the fact that their objectives are hard to define, the decision-making process is uncertain, and suitable measures of performance are at best qualitative.
Based on the experience gained by applying the hard systems thinking approaches to soft and unstructured problems, the Lancaster group started to redefine Systems Engineering until it was clear to them that a new methodology had evolved: Soft Systems Methodology. This methodology still depended largely on the hard systems approaches, but tried to place these theories within the arena of human affairs by relating it to the work of Sir Geoffrey Vickers (1970, 1983a, 1983b) on `appreciative systems' and to Churchman (1971, 1984) and his writings on Inquiring Systems.
Besides these theories the interpretative tradition is also represented in SSM by both hermeneutics and phenomenology. However, among the above-mentioned theories, it seems that it is Churchman who has made the greatest impression on Checkland when it comes to the notion of Weltanschauung (Checkland, 1981). It is also in relation to this concept that one main difference between SSM on one hand, and Systems Analysis and Systems Engineering, on the other, or more generally, between soft and hard systems thinking, can be found. The difference is that while the hard approaches believes in, and aims to describe, an objective world, and hence, according to the language of the soft approach only takes one Weltanschauung into consideration, the soft approach acknowledges and considers many different perceptions and Weltanschauungs.
In striving to develop a new methodology that could extend the area in which systems ideas could be used in the realm of social systems, the guiding principle for the Lancaster group was that `only concepts used in actual problems would be incorporated into the findings, and that the outcome should be a methodology for using systems ideas to find a structure in, and hence solve, real-world problems of a `soft' or ill-structured kind' (Checkland, 1975, p. 278). According to Checkland, the unquestioned prime value embodied in `a systems approach, and the implicit value system of SSM is that continuous, `never-ending learning is a good thing' (Checkland, 1981, p. 285). Besides this, a number of other underlying assumptions can also be found, like the assumptions that people in general want to solve their problems or improve problematical situations and that in order to do this some kind of consensus or accommodation between stakeholders needs to be achieved. Further, it is assumed that many problems and conflicting situations involve disagreements related to needs, objectives, measures of performance, etc., and have their base in either misunderstandings and/or differences in interests or culture. These misunderstandings and differences are usually due to lack of understanding or information or to different and incompatible Weltanschauungs. Either way, by understanding each other and these differences better we can solve, or at least improve, many problematic situations.
Other SSM characteristics arising from this are that SSM views the design process as a learning process and points at the importance of making plain many different perceptions of the problem situation, as well as the underlying rationale or `hidden rationality' (Stolterman, 1992) of these perceptions. It stresses the importance of trying to break away from self-imposed constraints and frames of mind, and, finally, it tries to reach accommodation through discussion. To summarize, the purpose of SSM can be stated as follows: to improve real-world situations by orchestrating changes of appreciation through a cyclic learning process.
The SSM process, which according to Checkland `is, in fact, a method for the more formal articulation of the workings of what Vickers terms the process of appreciation' (Checkland, 1979a, p. 325), consists of four main phases (1): finding out; systems modelling; comparison; and, taking action, illustrated in Figure 1. It is not the phases in themselves that makes SSM particular because as Checkland himself points out, they are all `everyday mental acts' (Checkland, 1981, p. 214), and can be found in most methods and methodologies aimed at change. Rather, its particularity is the way in which these phases are perceived, due to the epistemological and methodological Weltanschauung of the originators of SSM.
[FIGURE 1 OMITTED]
Even though it is possible to start the process at any phase (it is the relation between the phases rather then their order that is important), it usually starts in the first phase by an exploration of a real-world situation of concern (the left centre of the figure), initiated because someone perceives that situation as problematic and wants to do something about it. The purpose of the exploration is to provide a better understanding of the situation in question and it is usually summarized in a so-called `rich picture' (Checkland and Scholes, 1990).
From the rich picture, issues, judged by the analyst or someone else to be relevant for improving the problem situation, are selected and modelled using systems concepts (the upper right-hand corner of the figure). These models, depicted as square boxes, illustrate different perceptions or interpretations of the real-world situation under study and represent activities that logically need to be performed in order to reach a certain purpose. Because of this, they are referred to as conceptual models of `human activity systems' (HAS) (Checkland, 1971, 1981; Checkland and Scholes, 1990), and the class human activity systems comprises all activities that are carried out by human beings. In order to form a `whole' or a system, these activities are linked to each other by some principle of coherence or some underlying purpose or mission. Further, these models should be neither accounts of the real world, nor Utopian designs, but rather epistemological devices which help to structure a debate.
In the third phase, the models of human activity systems are set against actual perceptions of the situation, based on individuals' appreciative settings and, to some extent, depicted in the rich picture (the right centre of the figure). Through the comparison, and the debate it creates, new insights are revealed and appreciative settings may be changed, hopefully in such a way that accommodations between different interests and views can be reached -- accommodations that emerge as both feasible and desirable and can lead the way towards actions to improve the situation. This represents the fourth phase and is depicted at the bottom of the figure. The implementation of agreed-upon changes, or actions to improve the situation, then becomes the new problem situation and this way the methodology comes full circle. However, if changes cannot be agreed upon, a more extended examination of relevant systems will be necessary.
The purpose of the modelling phase is to tease out different perspectives of the problem situation and to structure the thinking of the same. In order to do this, some `precise' techniques have been developed, which consist of root definition (RD); PQR (2); CATWOE and conceptual models (CM) of human activity systems. Root definition means naming, in a short statement, a system of purposeful activity. The formal rules for a well-formulated root definition are that it should contain the elements of the mnemonic word CATWOE (Smyth and Checkland, 1976) and/or include the elements of PQR (Checkland, 1999). PQR refers to the statement `Do P by Q in order to contribute to achieving R' and answers the three questions: What to do (P); How to do it (Q); and Why do it (R)? (Checkland, 1999, p. A23f.), while CATWOE stands for:
* Customers: the persons that would be beneficiaries or victims of the system;
* Actors: the persons who perform the transformation process;
* Transformation: an input-output process by which some entity is changed to some new form of that same entity;
* Weltanschauung: a worldview which makes the transformation meaningful (3);
* Owners: the persons who can stop the transformation;
* Environmental constraints: elements which affect the system but which cannot be controlled.
The aim of the conceptual model is to represent the minimum number of logically contingent activities that need to be carried out in order to accomplish what has been defined in the root definition, and which the actors could, in principle, carry out right away. These models are not models of parts of the real world, as was pointed out above, but rather `ideal types' in Weber's sense. This means that each model is formed by the one-sided accentuation of a certain perception of the world, that is, by a worldview or Weltanschauung. Since the Weltanschauung is different for different people and since a particular Weltanschauung of a particular person changes through time, Checkland argues that there is no such thing as the real human activity system. Hence, in a sense, `human activity systems do not exist; only perceptions of them exist, perceptions which are associated with specific Ws' (Checkland, 1981, p. 219). This means that both the concepts human activity systems and conceptual models cannot exist without the concept Weltanschauung.
Further, every concept in the root definition should find expression in the conceptual model, and the conceptual model should reflect all aspects of the root definition, but no others. The aim is to achieve a pairing of root definition (what the system is) and conceptual model (what the system must do in order to be the system named in the root definition) which are mutually consistent (Checkland, 1981). Until recently, it was argued that conceptual models ought to represent whats rather than hows since they originate from root definitions and these latter usually define a what (Checkland, 1981; Wilson, 1992). It was also argued that, in order for the comparison phase to result in an outcome that is as effective as possible, the what of the conceptual model should be compared with the how of the real world.
Schregenberger, however argues that `we cannot expect a CM to be `deduced', or in a deterministic sense, `derived' from a RD, since the modelled original (the human activity system) cannot be considered to behave deterministically. CM's derived from RD's cannot be more than `systematic associations'; representations of abstract stored knowledge about analogous originals. Two different persons will derive different CM's from a given RD; since they will have different abstract concepts about the real world, being as they are to a large extent the result of formal education and training' (Schngerberger, 1982, p. 97).
This is also in line with Mingers' argument that from each root definition a number of different conceptual models can always be conceptualized; `there can never be a completely specified expansion of an RD into one and only one CM, unless the RD explicitly includes all the main activity verbs, in which case it could be argued that it is specifying a how anyway' (Mingers, 1990, p. 23). Because of this, Mingers suggests that conceptual models should not be seen as whats but rather as fairly general hows, detailing the way in which the what, expressed in the root definition, is to be carried out. Due to this, but also in accordance with their own practical experience, Checkland and Tsouvalis argue that 'CMs should, therefore, better be seen as `hows' rather than `whats' (Checkland, and Tsouvalis, 1997, p. 168).
Further, Checkland and Scholes argue that it is important that the formulation of these three, or four, techniques (RD, CATWOE, PQR and CM) are consistent with each other, `since credibility (and the participants' confidence in the process) can be diminished if some smart person in the situation points out a basic logical flaw in the model' (Checkland, and Scholes, 1990, p. 41). In this effort, the concepts of transformation process and Weltanschauung can be seen as a kind of glue which helps to keep the techniques in coherence with each other. These two concepts also form the base for the whole of the modelling phase; together they constitute the foundation for both the root definition and the conceptual models.
The importance of the concept of Weltanschauung and its central position within SSM have also been argued by other authors (Checkland and Davies, 1986; Fairtlough, 1982; Jackson, 1982). Within SSM, the concept is defined as a set of assumptions taken as given in communication between people which can help an observer understand social situations. Hence, it helps explain a situation, rather than the actors in that situation. This can be illustrated by an example given by Checkland (1981), where he says that when we observe people voting we do not see marks being made on pieces of paper but rather human beings taking part in the democratic process. Hence, whether we realize it or not, we view `raw' data via a particular mental framework, worldview or Weltanschauung and we attribute meaning to the observed activity by relating it to this larger framework that we supply from our minds. Even more, the observed activity is only meaningful to us in terms of a particular Weltanschauung, which in general we take for granted.
However, as a response to Fairtlough's (1982) article, pointing at eight different meanings associated with how the term Weltanschauung was used in SSM, Checkland and Davies (1986) distinguish three categories of the term: W1, W2 and W3. W1 is a taken-as-given set of assumptions which makes a particular root definition meaningful and is there only to help in model building. It has nothing to do with the dynamic flux of events which represents social reality and should be stated as pure and as simply as possible. W2 is related to a version of the problem situation and, thus, related to the taken-as-given assumptions in W1 in the sense that W2 makes W1 relevant. Finally, W3 is of wider concern and related to the social reality in which the problem situation is embedded. W3 is linked to our beliefs and assumptions about reality and makes us understand social situations. This means that while W1 is related to the modelling phase, W2 and W3 are not; they are both related to the problem situation existing in the real world. Because of this, neither W2 nor W3 is included in the techniques developed for SSM (Checkland and Davies, 1986; Fairtlough, 1982). Checkland and Davies also say that the concept Weltanschauung should be reserved for W2 and W3, while W1 can be referred to as simply W.
While I can understand the distinction made by Checkland and Davies, I feel that this distinction tends to drain the methodology of the richness that the concept Weltanschauung stands for. Since it is only W1 that is used in the modelling phase, and this concept is quite limited and restricted, much of the richness and therefore also understanding is lost. Evaluation of a broader notion of Weltanschauung, which could be found in W2 and W3, is also needed. The weakness of limiting the modelling phase to only W1 has also been mentioned by Bergvall-Kareborn and Grahn (1996a).
To illustrate my point I will give two examples from randomly chosen case studies of SSM. The first example is from a case study carried out in the East Berkshire District Health Authority, where the purpose was to investigate and recommend ways of measuring the performance of community medicine in this district (Checkland and Scholes, 1990, p. 100).
Root Definition:
A DHA-owned [District Health Authority] system, staffed by professional officers accountable to the DHA which, in the light of existing provision of health care (NHS [National Health Service] and non-NHS), plans and organizes the delivery of health care to defined populations using current health technology. The system manages the delivery via both ongoing services and specific projects, operates according to principles laid down by DHSS and Region and within the budget allocated. The system responds also to ad hoc issues arising outside the framework described. Its reporting meets the requirements of the `NHS planning system.' CATWOE T: population in given health state [right arrow] population in improved health as a result of this system's contribution W: organized provision of health care is feasible and desirable, it can be planed and organized
What do the T and the W really mean in the above statements? Why is it meaningful to plan and organize for improved health of people and why is it meaningful to improve the health of people at all? The W does not answer these questions; it does not address the issue of why improved health, or planning for it, is desirable and meaningful to achieve. Rather, it focuses on how to achieve improved health (through organized provision of health care). Besides this, the W also seems a bit too general because, in order to understand the given statement, we also need to understand the underlying rationale behind the assumption that organized provision of health care will, in fact, improve the health of people.
The second example is from Checkland and Tsouvalis (1997, p. 5) and looks as follows:
Root Definition:
A professionally-manned system in a manufacturing company which, in the light of market forecasts and raw material availability, makes detailed production plans for a defined period. CATWOE T: need for production plan --> need met W: rational planning of production is desirable and is a possibility
Similar questions can be asked in relation to this second T and W. Why is there a need for a production plan and what makes rational planning desirable and possible? The fact that someone wants to do T and think it is possible still does not tell us why that person would like to do T. Hence, in both of these cases the W in CATWOE does not really make plain what makes the transformation meaningful. Further, while it is important to state both what is to be done and how it is to be done, it is equally important to state why it should be done, and in these two examples the why seems to be lost, both in relation to the what and how. Finally, the Ws also seem too general and, as Boulding says, `we always pay for generality by sacrificing content' (Boulding, 1956a, p. 197). One result of this is that the more general a W is stated, the larger the number of possible interpretations that can be generated and, according to Checkland (1981), it is very difficult to make conceptual models out of root definitions including multiple transformations and Ws. Because of this, it is important to find some way to make the T more specific.
THE PHILOSOPHY OF HERMAN DOOYEWEERD
The philosophy of the Cosmonomic Idea, as it was called originally, was mainly developed by Herman Dooyeweerd (1894-1977), but is also related to the work of people such as Groen van Prinsterer (1801-1876) and Abraham Kuyper (1837-1920). Currently, however, the philosophy is usually referred to simply as Dooyeweerdian philosophy. It builds on the worldview coming from the Judeo-Christian tradition and assumes an ordered reality where structures and laws, to a large extent, already exist. These structures and laws are seen, not as constraints, but rather as guidelines helping the individual to understand the world around him and directing him when intervening in this reality through the process of human formation.
Hence, the philosophy aims at describing reality in order for us to understand it better. It consists of five fundamental and coherent theories: (1) the theory of religious ground motives; (2) the modal theory; (3) the theory of time; (4) the entity theory or the theory of individuality structures; and (5) the social theory.
In the following, I will focus on the parts of the philosophy that directly relate to the concept qualifying function. This means that only parts of the modal theory and parts of the entity theory will be covered. While both theories mainly deal with the law side of reality, there is a difference in focus between them. The modal theory describes modal functions and structures and offers a way of doing justice to both the diversity and the coherence in reality. It focuses on the general, is related to theoretical thought and tries to answer the question `How does something function?' The entity theory, on the other hand, focuses on classes of objects and activities found in reality, is related to concrete experience rather than to theoretical thought and asks `What is something?' This focus on the law side, rather than on the subject side, is visible in most of Dooyeweerd's work, and is commented on by Glas in the following way: `Dooyeweerd has said a lot about the structural side; however, he has never given a satisfactory account of the individual side' (Glas, 1995, p. 77). However, despite this focus Dooyeweerd both recognizes and strongly argues that everything that we experience in the world around us has both a law side and a subject side.
The Modal Theory
The modal theory was the first part of the philosophy to be developed and it came about as a result of Dooyeweerd's (and Vollenhoven's) quest for a reformed epistemology and ontology. In this search, it was the notion of `sphere sovereignty' (Bratt, 1998; Dooyeweerd, 1997; Henderson, 1994; Kalsbeek, 1975), introduced by Kuyper in order to protect different types of institutions in society from being dominated by the state, that eventually inspired and guided Dooyeweerd in his development of the theory (Dooyeweerd, 1973, 1997; Henderson, 1994; Kalsbeek, 1975). However, even though Kuyper saw the state as the strongest source of domination in society and therefore focused his interest on the relation between the state and other types of institutions, the idea behind the concept is more general and seeks to protect each type of institution in society from being dominated by any of the other types.
In trying to answer what constituted or distinguished separate life-spheres such as the state, church, school and family, and by what right these spheres should be sovereign and free to operate without being trespassed by any other, Dooyeweerd gradually identified distinct modal aspects related to human experience and thought.
Or as Dooyeweerd says in an interview:
During one of these walks in the dunes I received an insight (ingiving) that the diverse modes of experience, which were dependent upon the various aspects of reality, had a modal character and that there had to be a structure of the modal aspects in which their coherence is reflected. The discovery of what I called `the modal aspects of our experience horizon' was the point of connection. (In Henderson 1994, p. 38)
However, besides providing a methodological criterion for what was meant by Kuyper's concept of sphere sovereignty and, hence, providing a theoretical base for the concept, the modal framework also helped distinguish between different areas of knowledge and science. At the end of his life, Dooyeweerd had identified 15 irreducible aspects of human experience which formed the base of the modal theory (see Figure 2). These were the arithmetic (numerical), spatial, kinematic, physical, biotic, sensitive (psychic), logical, historical (cultural), lingual, social, economic, aesthetic, juridical (political), ethical and pistic (credal). These different aspects are referred to by a number of collective names, but I will try to limit myself to the labels dimensions, modalities or spheres in the following.
[FIGURE 2 OMITTED]
Since the modalities are based on human experience and thought, I will illustrate their meaning by relating them to a situation that I hope most readers will feel familiar with. Imagine a big and old spruce tree. The tree will have a certain number of branches and needles, representing the arithmetical aspect. It will also occupy a certain amount of space -- a space not so easily defined and demarcated if one considers the tree with all its root systems. This refers to the spatial modality. The kinematic aspect is visible in the molecular movement present in the tree, while the physical modality is noticeable through the feeding and breathing of the tree where it consumes water, sunlight and carbon dioxide and produces oxygen through the process of photosynthesis. The biotic modality can be illustrated both in the splendour of the tree but also by its eventual death.
Standing close to the tree we can touch the spruce and feel its sharp needles, observe the colour and form of the tree and smell the scent of it, all of which point to the sensitive aspect. Looking at the tree we can distinguish the branches from the stem, and the stem from the roots. This distinction, or analysis, illustrates the logical modality. The tree also has a historical aspect that makes it possible to investigate when it first came to a country and how it has spread during the years. Not only can we feel, observe and smell the tree, we can also talk and write about it, pointing at the lingual aspect. The social aspect related to a spruce tree is perhaps most noticeable, at least in most Christian countries, during Christmas. At this time, many people either go to the woods and cut themselves a tree, or they buy one, which they later bring inside the house and decorate, all in accordance with both social and historical traditions. The tree constitutes a gathering point for many social activities, especially for the younger ones.
During this time, the economic aspect of the tree also becomes visible, at least for those people standing in the street selling or buying Christmas trees. Here, the aesthetic modality is also present and will influence the price of the tree. The juridical aspect is also present in the buying and selling of the tree, since the owner of the tree is the one who can decide its destiny and call for legal protection against people who would steal or damage the tree. The ethical aspect is visible through the people who own forests and spruce trees and who allow relatives, friends and neighbours to cut themselves a tree for free. Finally, the pistic modality can be illustrated in our different beliefs on how the tree has come in to being; e.g., through a slow process of evolution made possible by the Big Bang or by the creation of a God.
The purpose with the modalities is not to exhaustively describe all of reality including human nature, nor do the modalities, or their sum, represent the whole of the universe. Instead, the modalities reflect the rich variety in human life and their intent is to remind us that both reality and humanity comprise a variety of discrete dimensions which cannot be reduced into one single dimension. The modalities also indicate that there is an interrelationship between the different dimensions that needs to be considered.
Further, the list of modalities is not a final one; on the contrary, Dooyeweerd constantly emphasized that the modalities, as well as their number, should be open for corrections and elaboration. In his later development he also showed this by including an additional aspect: the kinematic. This was originally part of the physical modality, but Dooyeweerd argued that while the physical modality was governed by energy, the kinematic was characterized by movement and, hence, distinguished from the physical modality. However, not all members of the school associated with Dooyeweerd agree with this distinction. A further example of differences in the interpretation of the modal order of reality is that some scholars (though not part of the Dooyeweerdian circle) assert that the pistic aspect represents nothing more than a sign of sensitive expression. If this is true, something that Dooyeweerd strongly rejects, the pistic modality could be reduced to the sensitive and should therefore not be a modality of its own (Kalsbeek, 1975).
While many times it is impossible, scientifically, to prove who is right and who is wrong in these arguments, the main principle stated by Dooyeweerd is that each modality should be fundamentally different from all the others. If this is not the case and if one modality is considered to be a variant of another modality and thus reducible to it, it should not be considered as a separate modality.
Although we can distinguish all of the aspects in temporal reality, they never appear in isolation, but always in an inseparable and mutual coherence. It is only in order to more fully understand the different modalities and their individual nature, that is, the element that distinguishes each aspect from the others and makes it irreducible, that we need to isolate them. The element distinguishing each aspect from all the others is called the nucleus or kernel and it is the nucleus that `characterizes or qualifies' (Kalsbeek, 1975, p. 84) a modality.
In Figure 3 all the modalities, together with their kernel, are listed. Hence, the numeric modality is characterized by its kernel discrete quantity, which is different from symbolic representation, the kernel of the lingual modality. As with the modalities, the kernels always indicate a how, a manner of being, never a concrete something. Hence, continuous extension may never be identified with the space we see around us. Space is always related to a concrete something in which all the aspects are found, not just the spatial (Kalsbeek, 1975). Besides having their own kernel, each modal sphere is also ruled by its own types of laws. These laws can be of two different types: determinative and normative. While the determinative laws always exert their own fulfilment, the fulfilment of the normative laws is contingent on people's inclination to follow them. However, Dooyeweerd argues that, as with the determinative laws, the normative laws `cannot be accounted for as a relation between or within people but must have an external origin or ground in order to have authority over us' (in Henderson 1994, p. 42). Finally, one very important thing to recognize with the normative modalities is that in a way they embody both their kernel and its negation. This means that both love and hate resides in the ethical modality. Being ethical or not being ethical are both ethical issues.
[FIGURE 3 OMITTED]
The Entity Theory
While the modal theory focuses on the general, the entity theory focuses on what is specific with different classes of entities and activities in reality, and it is within this latter theory that the concept of qualifying function belongs. The theory aims at gaining theoretical access to the structure in which things and events, also called individuality structures, present themselves to US.
An individuality structure is `a concrete, whole entity or event which has special qualities distinguishing it from all other individuality structures' (Kalsbeek, 1975, p. 42). However, the concept of individuality structure refers to the structure of a class of entities or activities, not individual entities and activities. Hence, all dogs have the same individuality structure, which is different from that of cars or that of schools, and all individuality structures have a qualifying function.
While all individuality structures function in all 15 modalities, it is only human beings and individuality structures related to human beings, like families, schools and businesses, that function as a subject in all of the modalities. A tree only functions as a subject up to the biotic modality, while a dog functions as a subject up to the sensitive modality. Hence, a dog functions as a subject in the arithmetical, spatial, kinematic, physical, biotic and sensitive modality, but it only functions as an object in the other 10 modalities. A dog does not analyse situations, master a language based on symbolic representation, or live by any particular belief system. The dog does, however, as an object, function in all of the 10 modalities above the sensitive. As an object, the dog can be fought for in a court of law, its aesthetic qualities can be appreciated and it can be described with the help of language.
In the same way as all individuality structures have subject and/or object functions they also have founding and qualifying functions. The founding function of an individuality structure is related to the aspect that has created the specific individuality structure under study, or made it possible for it to come into being. The qualifying function, on the other hand, refers to the function or modality that characterizes a particular individuality structure and gives it a certain identity. Further, the qualifying function also determines the inner structure and `indicates the intrinsic destination of a thing' (Dooyeweerd, 1997, vol. III, p. 60). Hence, it is the qualifying function that determines the form that all of the 15 modalities need to take in order to become a specific individuality structure. Because of this, the concept can be seen as the manager, foreman or pilot of the internal structure of the certain individuality structure and the concept is therefore also referred to as the guiding or leading function. Finally, all four concepts have an ontological character by trying to establish what something is, not merely how we interpret something.
To better illustrate the meaning and significance of the concept of qualifying function I will continue with the example of the tree. According to Dooyeweerd, the qualifying function of a tree is to be found within the biological modality, for while the physical and chemical processes within a tree are of vital importance, it is the biological function that qualifies the tree as a structural whole. It is this modality that characterizes it more than any other. However, the qualifying function should not be confused with external teleological relations such as what end or purpose a tree can serve in reference to other beings. The fact that a tree can give shade for humans and animals and that birds can build their nests in it lies outside of the internal structure of a tree in so far that, while the internal structure of a tree favours and enables it to be used for certain purposes, such as the ones mentioned above, using a tree for these purposes does not change the internal structure of the tree. Hence, the inner structure, or destination, of a tree, or any other individuality structure, is implicitly distinguished from its external teleological relations according to Dooyeweerd.
While natural individuality structures, or systems, like a tree, have their origin in the origin of the universe, designed individuality structures, or systems, are the result of conscious human activity. Because of this the qualifying function of designed individuality structures, like that of clothes, becomes closely related to the intentions and visions of the designer. This difference affects the concept of qualifying function, even though Dooyeweerd argues that it does not, in any fundamental way, change it. Even for designed objects there still exists a most fundamental difference between their qualifying functions and subjective or teleological ends. While there is nothing to prevent us from using a wineglass as a coffee cup or a soup bowl or a table as a seat, we are usually conscious that, according to its entire inner structure, they are not designed with these purposes in mind. And, regardless of how it is used, its internal structure will not change. `No, in empirical reality the objective structure of a thing is constant so long as the latter exists' (Dooyeweerd, 1997, p. 147).
A focus on designed individuality structures does, however, reveal a `problem' related to the concept of qualifying function, namely that in the course of time the objective, or original, function of a thing seems to change. Old armours and weapons originally intended for war or hunting may now be used as decorations or as symbols communicating historical conditions. There are also some individuality structures that are so narrowly bound to a certain office, social status, family or person that apart from such they cannot be used in accordance with their individual qualifying function. This points to the subject-object relation mentioned above and can be exemplified by items such as a throne, an admiral's uniform or a wedding ring. Everyone, provided they have access to the item, can sit on a throne, wear an admiral's uniform or a wedding ring, but this would still not cause the things to function according to their objective, or original, function.
Thus, in relation to designed systems three different modes related to the qualifying function need to be defined. The first mode is called `intentional representational relation' or `objective qualifying function' and is defined as the `objectification, in the structure of a real object, of the intentional object, as the latter is conceived in the design of a thing'. This means that when one designs something the qualifying function of the designed object affects its structure and through this the qualifying function is objectified into the object. The second mode is called `unfolding relation' or `opening relation' and is said to be the `subjective unfolding or opening in human experience of the closed objective thing-structure'. By this is meant the experience or knowledge of the objective qualifying function of a certain individuality structure from a historical period. Finally, the last mode, called `actualization relation' is defined as the `actualization of the objective thing-structure by human activity, which uses the thing according to its objective and open qualifying function' (Dooyeweerd, 1997, Vol. III, p. 148). This means that a certain individuality structure is used in accordance with its objectified qualifying function.
To illustrate the three modes let us consider the clothing of knights. In medieval feudal society knighthood had an actual military function. At the end of the Middle Ages, knighthood lost its military significance, but retained its position as a rank endowed with political power. The French Revolution destroyed this position, whereby knighthood disappeared as a particular privileged social rank. However, later generations `can still experience knightly attire as objective socially qualified things of a by-gone historical period. Their objective qualifying function is the objectification, in the real clothing material, of an intentional conception of the designer of these costumes. This state of affairs remains unchanged by the transitions of cultural development' (Dooyeweerd, 1997, Vol. III, p. 147). However, the socially qualified objective function of knightly costumes can no longer be actualized in today's society because the feudal class system has disappeared.
INCORPORATING THE CONCEPT OF QUALIFYING FUNCTION INTO SSM'S MODEL-BUILDING PHASE
Since there is a difference in purpose between Dooyeweerdian philosophy and the underlying philosophy of SSM, and hence between the concept of qualifying function and the modelling phase of SSM, I want to make this difference clear. I also want to explain why I consider it possible and fruitful to bring some theories and ideas from the Dooyeweerdian philosophy into SSM. This will be done by arguing that this can have positive effects on SSM.
In the above it has been argued that the Dooyeweerdian philosophy strives to say something about reality; to describe or depict it in order for us to better understand it, and that the concept qualifying function, in its originality, is thought to represent an actually present function within an object or activity. The underlying philosophy of SSM, especially noticeable in its modelling and comparison phase, is characterized by its focus on people's perceptions of reality, rather than on reality as such. The aim with the modelling phase, for example, is to tease out different perceptions of a certain problem situation and to depict these in so-called root definitions and conceptual models. These root definitions and conceptual models are not supposed to be models of the real world, but rather models relevant for debating about the real world. Hence, while the focus for the Dooyeweerdian philosophy is on reality as such, the focus for SSM is on people's perceptions of reality.
Further, while Dooyeweerd argues that given laws and structures exist both within the natural and the social world, Checkland only acknowledges given laws and structures within the natural world. This can be seen in many of his statements and also in his `Systems Map of the Universe' (Checkland, 1971, 1981). In the latter, he places social systems in between natural and human activity systems, with the argument that it would be wrong to place social and cultural systems solely within the boundary of human activity systems because even though they are the context of virtually all human activity they are also, at the same time, natural systems. The relation to natural systems is through human beings and `due to the fact that man is a gregarious animal who has a basic need for the supports provided by his fellow human beings in community life' Checkland (1971, p. 110). Hence, Checkland only acknowledges laws and structures within social reality arising from the human being as a natural system.
Personally, I believe that some type of structure exists even within social reality, and the great similarities that exist between countries and nations when it comes to basic values strengthen my belief. I also believe that, even though the world is not divided into physics and biology or psychology and sociology, there is a relation between reality and our perception of it. Hence, even though it is people who have decided that some things should be called linguistics, while other things are named ethics and aesthetics, and even though it would be possible both to rename and redefine these notions, I believe that the notions have come about because, somehow, there are aspects reflecting these notions in the world around us. But whether these notions have their origin in man or should be sought outside the realm of man, I cannot answer.
Further, even though I believe in the existence of certain aspects this does not in any way mean that I see the whole of reality as given. Rather, it indicates a number of aspects where we, as individuals and groups, have the possibility to intervene and form reality through the process of human formation. Hence, even though language and social norms, of some kind, exist in most, if not all, cultures they can take quite different forms. And even though I believe in ontologically present aspects I do not believe that it is epistemologically possible to ever find a method or methodology that can give us full and true knowledge about the world around us. This means that besides analysing `how the world is' and debating about `how the world ought to be' we should also ask ourselves `how the world could be' and `how we want it to be'. Addressing the two latter questions we enter into the area of design, because design focuses on reality as it can become or be made to be, rather than on reality as it is. Design is not a predictable or determinable activity, and it does not lead to insights into reality as it must be of necessity; rather it focuses on creating the not yet existing (Lowgren and Stolterman, 1998; Strijbos, 1998).
Before discussing how the modal theory and the concept of qualifying function can enrich the model-building phase of SSM there are two comments that I wish to make in relation to Dooyeweerd's view on the notion of qualifying function. Firstly, Dooyeweerd argues that `in empirical reality the objective structure of a thing is constant so long as the latter exists'. I would say that this is a rule to which there are exceptions. Imagine a wineglass from ancient Greece. It might have been designed as part of a fine service to be used at specific holidays, rituals or other social or religious events. As such, it would be the social or the pistic modality that would represent the objective qualifying function, also called intentional representational relation of the glass, and guide its inner structure. However, by the time we discover the wineglass in modern times it might be impossible to use the glass in accordance with its original intent because we might not have full knowledge about the rituals or special conditions surrounding the use of the glass. Using the terminology of Dooyeweerdian philosophy, we might not be able to unfold or actualize the objective qualifying function of the wineglass. While this might change the way we viewed the glass it would not change the structure of the glass itself. If, on the other hand, the glass were to be broken and would have to be glued together or if it had been lying in a place where dangerous material had blended into its original material this would both make it impossible to use the glass in accordance with its qualifying function but, it would also, I argue, change the objective structure of the glass as such.
While the above illustration or argument might be more of an exception than a rule when it comes to natural or designed physical things, it is much more common when it comes to social systems. These systems, contrary to natural and designed physical ones, can much more easily change their structure with the passing of time and in relation to people's perception of them.
The following statement, made by Checkland when discussing designed systems, can help to illustrate how I view the concept of qualifying function. He says that: `We could if we wished use a piece of rock, a natural system, to knock nails into pieces of wood to make a hen coop. But we would perform the task better if we used a hammer, a physical system designed with fitness for purpose in mind' (Checkland, 1981, p. 119). The idea that when we design something the vision and intentions of the designer are objectified into the designed object and through that gives the object a certain form and function is very appealing to me. It also feels closely related to the concept of qualifying function. To illustrate my point I will take the thought experiment given by Tsouvalis and Checkland (1996) where an unconventional householder chooses to view a bicycle as a kind of fencing, and therefore makes a fence from a line of bicycles buried up to the axles. While the authors are right when they say that no one can stop the householder from viewing, or using, bicycles as a type of fence, in the same way as we can use a book or a piece of rock as a hammer if we want to, one can hardly, at least not in my mind, argue that bicycles have been designed to fit that purpose. However, even though most designed objects are designed with a certain purpose or function in mind, this function or purpose is not always easily recognized by other people.
Since the aim of this paper is to address some critical comments made towards SSM, not to try and combine SSM with Dooyeweerdian philosophy as such, and since design is focused on the new and not yet created, I will use the concept of qualifying function in a more exploring and interpretative way than was intended in its originality. I will use it as a tool to discuss which function and modality best relate to the design vision of the designer(s) rather than to identify a specific function present in already existing entities and activities. While SSM is already strong on teasing out different perceptions, the concept of qualifying function has the potential to further strengthen this process. Using it this way has the following benefits.
Incorporating the concept of qualifying function into the modelling phase of SSM can help people in a particular problem situation to view the situation from new and different perspectives. The importance of the latter has been argued by many authors (Ackoff 1993; Checkland, 1981; Flood and Jackson, 1991; Morgan, 1986) and cannot be overemphasized in a design situation. Ackoff (1993), for example, argues that the identification and removal of self-imposed constraints provide two out of three steps in creative behaviour. The third is an exploration of the consequences of having done so. This, however, often requires that we can change, break away from, or at least question our present Weltanschauung -- something that has proven to be a very difficult task, which also often appears very threatening to people. As an example of how hard it can be to question previously unquestioned worldviews, consider the well-known example of the shift from the Aristotelian to the Copernican Weltanschauung or Weltbild (4) related to the structure of the Universe.
In Total Systems Intervention, Flood and Jackson (1991) do something similar by relating different systemic metaphors, similar to those of Morgan's (1986) eight metaphors of an organization, to different systems methodologies. They argue that these metaphors can highlight aspects of a particular problem situation and hence guide the choice of appropriate systems problem-solving methods. Or as Midgley puts it, `[p]articipants can use them to think in different ways about the issue with which they are concerned' (Midgley, 1997, p. 259). In addition, they argue that consciously looking at problem situations through different metaphors should help a manager in creative thinking. However, the benefit with the modal framework is that it is based on a theoretical framework that gives each modal aspect a defined meaning and establishes coherence between the different spheres.
SSM is both aware of the importance of being able of put on different kinds of thinking caps and has also developed techniques (the formulation of root definitions, CATWOE and conceptual models) for how this can be done. Hence, the concept of qualifying function is not meant to replace these techniques. Rather it is meant to complement the already existing techniques in the identification, and hopefully removal of self-imposed constraints. The concept is also useful when exploring alternative design possibilities.
To better illustrate how the concept of qualifying function can be used to complement the modelling techniques in SSM, the concept is incorporated into an SSM study carried out in the East Berkshire District Health Authority by, among others, Peter Checkland (5). The purpose of this study was to help the East Berkshire Community Medicine Department (CMD) with the problem of measuring its performance, which mainly consisted of providing epidemiological data and actively being involved in helping to manage the delivery of health care programmes in the district. To manage this Checkland felt that they needed to understand `not only the ideas (and controversy) underlying the different conceptions of community medicine but also the basic mechanisms operating at district level in the NHS in the provision of health care' (Checkland and Scholes, 1999, p. 97).
Incorporating the concept of qualifying function into this study would include selecting and discussing possible functions that can qualify the National Health Care Service, the East Berkshire District Health Authority and/or the East Berkshire CMD. This can be seen as a way of introducing W2 and W3 into the modelling phase and let them set the stage for further modelling.
If we select the National Health Care Service it is probably the biotic modality that feels like the most logical choices for representing the inner structure of the system. Taking this modality as the qualifying function one would argue that it is this modality that characterizes the kernel of the activity carried out at our hospitals and health care systems. It is the biological modality, with a focus on curing diseases and saving lives, that represents the scientific discipline to which doctors and nurses belong and it is within this modality that they have their knowledge, responsibility and sovereignty. This view can be noticed in the argument made by the authors when they say: `If, as is often the case, there is more than one way of carrying out a surgical operation, a DGM [District General Manager] cannot instruct a surgeon to adopt the cheapest way, that must be left to the surgeon's professional judgement' (Checkland and Scholes, 1999, p. 94). This argument is based on the underlying assumption that the biological modality constitutes the main responsibility or function of the medical staff, and that this is where they have sovereignty. If this was not the case the statement would not make sense.
However, it is always important to consider a number of possible functions or modalities to qualify a specific system or holon since the main purpose of the activity is to provide insights into our underlying assumptions regarding the problem situation at large and to try to find new ways to view the situation. In the case described above, modalities such as the social, economic, juridical, ethical or pistic can all provide interesting insights into our assumptions on health care.
Taking the ethical modality, for example, as the qualifying function, one could argue that curing biological diseases is only one element of the healing process and therefore not a broad enough focus for the health care system. Instead of focusing on individual parts or diseases health care should focus on caring for the whole person with a special emphasis on caring rather than curing. Hence, by letting the ethical modality guide and characterize the health service one might take a broader perspective on health and argue that in order for people to improve their health we need to see them, and care for them, as whole human beings, not as broken parts like a bad knee, an aching back or a high blood pressure.
This change in the qualifying function from the biological modality, focused on curing diseases and saving lives, to the ethical modality, focused on human compassion and care, would also most probably affect both the activities and the structure of the hospital. Discussing and deciding whether it is right or wrong for hospitals to provide help in dying for patients that are in great pain and where there is no hope for recovery would, for example, be seen very differently. From a biological standpoint this would be seen as a real `sin' since it would go against the very purpose of the system. From an ethical viewpoint, on the other hand, helping patients take control of their situation instead of letting them die in pain and fear could instead be seen as a human and caring action. As to the structure of the hospital differences regarding where the hospital should be located (where the air, temperature and humidity are especially favourable or close to family and friends), how the rooms should be designed and the importance of having playgrounds, family rooms and social gathering spaces might also differ depending on what qualifying function is chosen.
Further, the concept of qualifying function can also be used to facilitate the process of generating different possible views on what makes a particular T meaningful, i.e., generating different W1' s. To exemplify this in relation to the above study I have chosen to focus on a model described by the authors as `a rather detailed model of a general (primary task) system to plan and organize (but not deliver) health care to a defined population' (Checkland and Scholes, 1999 p. 97ff.). Following the authors, the system has the subsequent root definition and CATWOE.
Root Definition:
A DHA [District Health Authority]-owned system, staffed by professional officers accountable to the DHA which, in the light of existing provision of health care (NHS [National Health Service] and non-NHS), plans and organizes the delivery of health care to defined population using current health technology. The system manages the delivery via both ongoing services and specific projects, operates according to principles laid down by DHSS and Region and within the budget allocated. The system responds also to ad hoc issues arising outside the framework described. Its reporting meets the requirements of the `NHS planning system'. CATWOE C: defined population A: professional officers T: population in a given health state --> population in improved health as a result of this system's contribution W: organized provision of health is feasible and desirable; it can be planned and organized O: DHA E: Structure DHSS/Regional District; `NHS planning system' as a reporting mechanism; budget
Using the qualifying function to elicit different possible W's by generating different views on what makes a particular T meaningful would in this case mean that we focus on why it is seen as meaningful to plan and organize for improvement of people's health. Searching for possible functions that could qualify the above activity or transformation does this.
Seen from an ethical point of view one might desire to improve the health of people because one wanted to limit the suffering that sickness can cause, both to the sick and to their relatives and friends. It is also possible to strive towards improved health as an issue of equality or justice: trying to balance the seeming injustice of nature in allowing some persons to be born strong and healthy while others are born weak and sick. Or one might want to improve the health of people for economic reasons, arguing that it is more economic, in the long run, to invest money in health care, since this most probably will result in a healthy population that can work and contribute to the production of products and services.
Besides facilitating the process of generating different possible views on what makes a particular T meaningful, the concept of qualifying function can also help to clarify an already stated T or W. In relation to the above statements it becomes clear that there can be many different views on why it is seen as desirable to improve the health of a given population. By clarifying these different views new root definitions with more specific W's can be created. This is important considering Checkland's (1981) statement that it is very difficult to make conceptual models out of root definitions including multiple transformations and W's.
Using the qualifying function in the two ways described above can improve our understanding and help question our underlying assumptions related to established systems such as the health care system. It can also help us with the process of generating different possible views (W's) on what makes a particular T meaningful as well as help to clarify an already stated T or W. However, besides these three benefits the concept also has the potential to address some of the critiques directed towards SSM.
Firstly, including the concept of qualifying function in the modelling phase, as was done above, could be one way of addressing the critique given by Mingers (1980) that SSM does not notice similarities between different W's or seek an explanation for this in terms of the structure of society. As in the illustration related to the British coal-mining industry after the Second World War, Checkland (1981) points to the difference in Weltanschauung between Lord Robens, Chairman of the National Coal Board, and one of the working miners, but does not consider the fact that most people in Lord Robens' position will have similar appreciations, and these will be very different from those of most working miners. Hence, SSM is said to lack an explanation of why these particular W's have developed and thereby how they might be changed. While neither the concept of qualifying function nor the modal framework constitutes a critical sociology, as is suggested by Mingers, they have the possibility to at least point to similarities and differences in focus between and/or within different groups of people.
In the study discussed above related to the NHS it could be argued that while doctors and nurses, due to their education, interests and responsibilities, might be more apt to view situations from biotic and ethical perspectives, administrators and economists might instead focus on economical and juridical issues and give prominence to these aspect in their discussions.
Secondly, expanding the T and/or the W also has the potential to enrich the conceptual models. As early as 1979 Naughton argued that `inexperienced analysts invariably plump for bureaucratic-type models' (Naughton, 1979, p. 70) when making the transition from root definition to conceptual models. In a more recent study Mirijamdotter (1998) argues that when the model builder lacks personal experience of the subject or transformation being modelled the conceptual models tend to become impersonal, general, poor on normative issues and sometimes not very informative. These statements are also in agreement with the findings of Mingers (1992), Schregenberger (1982), Vickers (1983a), Woodburn (1985) and even Checkland (Checkland, 1999; Checkland and Tsouvalis, 1997). However, teasing out the qualifying function for the transformation (T) has the potential to enrich the conceptual models in that it adds knowledge to the T and by that helps the model builder find relevant activities.
Thirdly, incorporating the concept of qualifying function into the root definition can also strengthen `the fragility of the "bond" between a conceptual model and a root definition' (Checkland and Tsouvalis 1997, p. 162) and by that strengthen the defensibility of the conceptual model. Defensibility here refers to the second of the two validity questions posed by Checkland in order to answer how well one can tell a `good' design from a `bad' one: the first being `whether a model is actually "relevant" or not', and the second, `whether a given model is competently built' (Checkland, 1995, p. 52f.).
In the early stage of the methodology the `Formal Systems Model' (Checkland, 1981) was developed for this purpose, i.e., to check `whether a given model is competently built'. Later on the formal systems model was, however, discarded on the grounds that it blurred the distinction between the language of the real-world problem situation and the systems thinking language of the same (Checkland and Scholes, 1990; Checkland and Tsouvalis, 1997). Instead, another technique for the same purpose has been introduced. This technique aims at modelling the transformation process of the root definition in such a way that it fulfils the criteria of effectiveness, efficiency and efficacy (Checkland and Scholes, 1990). Later, the criteria ethics and elegance were also added (Checkland et al., 1990). These criteria are usually referred to as the 3 or 5 E's, and function as the measures of performance by which the conceptual model is judged.
Recently, a third technique has been proposed, which suggests that all the main verbs of a conceptual model should be incorporated into the root definition (Checkland and Tsouvalis, 1997). While this technique helps to bind the root definition and the conceptual model closer together, by showing that each activity in the conceptual model stems from the words used in the root definition, it also has as a consequence the loss of `the richness gained by the pairing of being and doing which the root definition-conceptual model relationship provides' (Checkland, 1981, p. 292; Checkland and Tsouvalis, 1997, p. 160).
While the first technique uses general system thinking concepts as a means to evaluate the technical defensibility or validity of a conceptual model, the second uses more logical and economic concepts. The third technique, on the other hand, focuses on the internal consistency between the modelling tools (root definition, CATWOE and conceptual model). However, as was argued above, this validation technique binds the modelling tools so close together as to almost make them alike. Hence, the concept of qualifying function could constitute an alternative technique for checking the technical defensibility of a conceptual model.
It could of course also be argued that the concept of qualifying function, while being based on the 15 modalities, can create a boundary for our thought and hence constrain rather than free our minds. While I have to agree that this is a possibility, my defence is that all concepts, tools, methods or methodologies have this weakness. While they help and guide us in how to think, what to do and how to do it, they also automatically have the possibility to constrain us. As soon as you highlight one direction the others easily get pushed into the background. However, I argue that the help they offer far exceeds the constraint they represent. Further, this weakness or constraint inherent in methods and methodologies can be limited by being aware of its existence and by working with many different and diverse methods.
SUMMARY AND CONCLUSION
In this paper I have pointed to the central role of the concept of Weltanschauung or worldview within SSM. While the concept plays a central role within the whole of the methodology its presence is most noticeable within the model-building phase. Here, the aim is to tease out different perceptions of the problem situation and to structure the thinking of the same. In order to do this some precise techniques have been developed (root definition, CATWOE and conceptual model). Hence, the techniques do not aim to describe part of the real world but rather to accentuate certain perceptions of the world.
However, in order to handle the complexity that a person's Weltanschauung implies, the content of the term has been reduced quite drastically in the model-building part of SSM. Instead of including the whole Weltanschauung of a person, Checkland suggests that it is enough to try and tease out the underlying rationale of why a person sees it meaningful to carry out a certain activity or transformation. Even if this is most helpful it was argued that the concept becomes too narrowly bounded to do full justice to the diversity found in different perceptions. In order to further enrich and stimulate the process of teasing out different perceptions I introduced a concept called qualifying function.
The concept originates from the philosophy of the Cosmonomic Idea, presently termed the Dooyeweerdian philosophy, and refers to the function that characterizes a particular thing or activity. Hence, it can be seen as the manager, foreman or pilot of the internal structure of a particular individuality structure, and is therefore sometimes also referred to as the guiding or leading function. Finally, this function or characteristic is also related to the modal framework developed by Dooyeweerd that comprises the following 15 irreducible aspects: the numerical (arithmetic), spatial, kinematic, physical, biotic, sensitive (psychic), logical, historical (cultural), lingual, social, economic, aesthetic, juridical, ethical and credal (pistic). This means that the qualifying function of a particular individuality structure is related to one particular modality.
Incorporating the concept of qualifying function into the modelling phase of SSM has the following benefits. Firstly, it helps people in a particular problem situation to view systems, situations and activities from new and different perspectives by complementing already existing methods and techniques. Secondly, it can be used to clarify a stated Transformation (T) and/or Weltanschauung (W) as well as to point at other possible ways of interpreting these. This is important considering that Checkland argues that it is difficult to make conceptual models out of root definitions including multiple T's and W's. Thirdly, the concept can be used to related different W's to roles and structures in society and by that increase our understanding of a stated W. This also, to some extent, addresses the critique given by Mingers that SSM lacks an explanation of why particular W's have developed. While neither the concept of qualifying function nor the modal framework constitutes a critical sociology, which is what Mingers calls for, they have the possibility to at least point to similarities and differences in focus between and/or within different groups of people. Fourthly, the concept provides an alternative technique for model validation by relating the root definition and the conceptual model closer together. Finally, it has the potential to enrich the conceptual models by directing the transformation process (T) and by that hopefully make the models less conservative.
I would like to end the paper by pointing out that this paper constitutes only a first step in the exploration of how the concept of qualifying function can be related to a specific design methodology, in this case SSM, and the benefits this can have. However, in further studies the concept and its discussed benefits need to be tested in real-world situations and, hence, applied in empirical studies.
* Correspondence to: Birgitta Bergvall-Kareborn, Department of Informatics and Systems Science, Lulea University of Technology, S-971 87 Lulea, Sweden. Published online 12 December 2001
(1) These phases are called `perceiving', `predicating', `comparing' and `deciding' by Checkland (Checkland 1981, p. 214).
(2) In earlier writings PQR has always been referred to as XYZ, but in Checkland (1999) it is changed, with the argument that Y might be confused with why represented in the question `why carry out the transformation'?
(3) In Checkland (1981, p. 225) Weltanschauung is said to make the root definitions meaningful to the actors in the situation.
(4) Weltbild is a more limited concept and is related to our understanding of the physical universe.
(5) This study was also used as an example in the section above on `Soft Systems Thinking' in order to illustrate the limitation of only considering W1 in the modelling phase.
REFERENCES
Ackoff RL. 1993. Idealized design: creative corporate visioning. International Journal of Management Science 21: 401-410.
Bergvall-Kareborn B, Grahn A. 1996a. Expanding the framework for monitor and control in soft systems methodology. Systems Practice 9: 469-495.
Bergvall-Kareborn B, Grahn A. 1996b. Multi-modal thinking in soft systems methodology's rich pictures. World Futures 47: 79-92.
Boulding KE. 1956a. General systems theory: the skeleton of science. Management Science 2: 197-209.
Boulding KE. 1956b. The Image: Knowledge in Life and Society. University of Michigan Press: An Arbor, MI.
Bratt JD. 1998. Abraham Kuyper: a centennial reader. Paternoster Press: Carlisle.
Checkland PB. 1970. Systems and science, industry and innovation. Journal of Systems Engineering 1: 3-17.
Checkland PB. 1971. A systems map of the universe. Journal of Systems Engineering 2: 107-114.
Checkland PB. 1972. Towards a systems-based methodology for real-world problem solving. Journal of Systems Engineering 3: 87-116.
Checkland PB. 1975. The development of systems thinking by systems practice: a methodology from an action research program. In Progress in Cybernetics and Systems Research, Trappl R, Hanika FdP (eds.) Hemisphere: Washington, DC; 278-283.
Checkland PB. 1978. The origins and nature of `hard' systems thinking. Journal of Applied Systems Analysis 5: 99-110.
Checkland PB. 1979a. The problem of problem formulation in the application of a systems approach. In Education in Systems Science, Bayraktar BA, Muller-Merbach H, Roberts JE, Simpson MG (eds). Taylor & Francis: London; 318-326.
Checkland PB. 1979b. Teaching and testing systems methodology. In Education in Systems Science, Bayraktar BA, Muller-Merbach H, Roberts JE, Simpson MG (eds). Taylor & Francis: London; 190-200.
Checkland PB. 1981 Systems Thinking, Systems Practice. Wiley: New York.
Checkland PB. 1982. Soft systems methodology as process: a reply to M. C. Jackson (Part 3 of `Three replies to Jackson'). Journal of Applied Systems Analysis 9: 37-39.
Checkland PB. 1984. Systems thinking in management: the development of soft systems methodology and its implications for social science. In Self-Organization and Management of Social Systems, Ulrich H, GJB (eds). Springer: Berlin; 94-104.
Checkland PB. 1994. Systems theory and management thinking. American Behavioral Scientist 38: 75-91.
Checkland PB. 1995. Model validation in soft systems practice. Systems Research 12: 94-104.
Checkland PB. 1999. Systems Thinking, Systems Practice: Includes a 30-Year Retrospective. Wiley: Chichester.
Checkland PB, Davies L. 1986. The use of the term `Weltanschauung' in soft systems methodology. Journal of Applied Systems Analysis 13: 109-115.
Checkland PB, Forbes P, Martin S. 1990. Techniques in soft systems practice part 3: monitoring and control in conceptual models and in evaluation studies. Journal of Applied Systems Analysis 17: 29-37.
Checkland PB, Griffin R. 1970. Management information systems: a systems view. Journal of Systems Engineering 1: 29-43.
Checkland PB, Haynes MG. 1994. Varieties of systems thinking: the case of soft systems methodology. System Dynamics Review 10: 189-197.
Checkland PB, Holwell S. 1993. Information management and organizational processes: an approach through soft systems methodology. Journal of Information Systems 3: 3-16.
Checkland PB, Jenkins GM. 1974. Learning by doing: systems education at Lancaster University. Journal of Systems Engineering 4: 40-51.
Checkland PB, Scholes J. 1990. Soft Systems Methodology in Action. Wiley: New York.
Checkland PB, Scholes J. 1999. Soft Systems Methodology in Action: A 30-Year Retrospective. Wiley: New York.
Checkland PB, Tsouvalis C. 1997. Reflecting on SSM: the link between root definitions and conceptual models. Systems Research and Behavioral Science 14: 153-168.
Churchman CW. 1971. The Design of Inquiring Systems: Basic Concepts of Systems and Organization. Basic Books: London.
Churchman CW. 1984. The Systems Approach. Dell: New York.
Dahlbom B. 1992. The idea that reality is socially constructed. In Software Development and Reality Construction, Floyd C, Zullighoven H (eds). Springen: Berlin; 101-126.
Dooyeweerd H. 1973. Introduction. Philosophia Reformata 38: 5-16.
Dooyeweerd H. 1997. A New Critique of Theoretical Thought. Edwin Mellen Press: Lewiston, UK.
Fairtlough G. 1982. A note on the use of the term `Weltanschauung' (W) in Checkland's Systems Thinking, Systems Practice. Journal of Applied Systems Analysis 9: 131-132.
Flood RL, Jackson MC. 1991. Creative Problem Solving: Total Systems Intervention. Wiley: Chichester.
Glas G. 1995. Ego, sell and the body: an assessment of Dooyeweerd's philosophical anthropology. In Christian Philosophy at the Close of the Twentieth Century: Assessment and Perspective, Griffioen S, Balk BM (eds). Uitgeverij Kok: Kampen; 67-78.
Henderson RD. 1994. Illuminating law: the construction of Herman. Dooyeweerd's philosophy. Doctoral thesis, Free University, Amsterdam.
Jackson MC. 1982. The nature of `soft' systems thinking: the work of Churchman, Ackoff and Checkland. Journal of Applied Systems Analysis 9:17-29.
Jackson MC. 1983. The nature of `soft' systems thinking: comments on the three replies. Journal of Applied Systems Analysis 10: 109-113.
Jackson MC. 1991. The origins and nature of critical systems thinking. Systems Practice 4: 131-149.
Jenkins GM. 1969. The systems approach. Journal of Systems Engineering 1: 3-49.
Kalsbeek L. 1975. Contours of a Christian Philosophy: An Introduction to Herman Dooyeweerd's Thought. Wedge: Toronto.
Lowgren J, Stolterman E. 1998. Design av Information-steknik; Studentlitteratur: Lund.
Midgley G. 1997. Mixing methods: developing systemic intervention. In Multimethodology: The Theory and Practice of Combining Management Science Methodologies, Mingers J, Gill A (eds). Wiley: Chichester; 249-290.
Mingers J. 1980. Towards an appropriate social theory for applied systems thinking: critical theory and soft systems methodology. Journal of Applied Systems Analysis 7: 41-49.
Mingers J. 1984. Subjectivism and soft systems methodology: a critique. Journal of Applied Systems Analysis 11: 85-103.
Mingers J. 1990. The what/how distinction and conceptual models: a reappraisal. Journal of Applied Systems Analysis 17: 21-28.
Mingers J. 1992. Questions and suggestions in using soft systems methodology. Systemist 14: 54-61.
Mirijamdotter A. 1998. A multi-modal systems extension to soft systems methodology. Doctoral thesis, Lulea University of Technology, Lulea.
Morgan G. 1986. Images of Organization. Sage: London.
Naughton J. 1979. Functionalism and systems research: a comment. Journal of Applied Systems Analysis 6: 69-73.
Prevost P. 1976. `Soft' systems methodology, functionalism and the social sciences. Journal of Applied Systems Analysis 5: 65-73.
Schregenberger J. 1982. The development of Lancaster soft systems methodology: a review and some personal remarks from a sympathetic critic. Journal of Applied Systems Analysis 9: 87-98.
Smyth DS, Checkland PB. 1976. Using a systems approach: the structure of root definitions. Journal of Applied Systems Analysis 5: 75-83.
Stolterman E. 1991. Designarbetets Dolda Rationalitet. Doctoral dissertation, Umea University, Umea.
Stolterman E. 1992. How system designers think about design and methods: some reflections based on an interview study. Scandinavian Journal of Information Systems 4: 137-150.
Strijbos S. 1998. Towards a new critical systems thinking: philosophical reflections on systems methodology and systems ethics. In Proceedings of the 42nd Annual Conference of the International Society for the Systems Sciences, Allen JK, Wilby J eds. Atlanta, GA; 1-9.
Tsouvalis C, Checkland P. 1996. Reflecting on SSM: the dividing line between `real world' and systems thinking world. Systems Research 13: 35-45.
Vickers G. 1959. The Undirected Society. University of Toronto Press: Toronto.
Vickers G. 1970. Freedom in a Rocki Boat. Allan Lane: London.
Vickers G. 1983a. The Art of Judgement: A Study of Policy Making. Harper & Row: London.
Vickers G. 1983b. Human Systems are Different. Harper & Row: London.
Wilson B. 1992. Systems: Concepts, Methodologies and Applications (2nd edn). Wiley: Chichester.
Woodburn L. 1985. Some developments in the building of conceptual models. Journal of Applied Systems Analysis 12: 101-106.
Birgitta Bergvall-Kareborn (1) *
Department of Informatics and Systems Science, Lulea University of Technology, Lulea, Sweden
 Printer friendly
 Cite/link
 Email
 Feedback
| |
| Author: | Bergvall-Kareborn, Birgitta |
|---|---|
| Publication: | Systems Research and Behavioral Science |
| Geographic Code: | 4EUSW |
| Date: | Jan 1, 2002 |
| Words: | 14512 |
| Previous Article: | `Devolution' as an opportunity to test the `synergism hypothesis' and a cybernetic theory of political systems. (Research Paper). |
| Next Article: | Analysing IT ethics. (Research Paper). |
| Topics: | |