Printer Friendly

Soft systems methodology and 'soft' philosophy of science.


In his academic work Soft Systems Methodology: A 30-year Retrospective written about 10 yeas ago, Checkland (1999, p. A43) theoretically summarizes his main systems thinking in soft systems methodology (SSM) and says that SSM is 'a rigorous approach to the subjective'. This work means that Checkland has finally completed the philosophical base of SSM. It causes a heated debate in China. Some scholars argue for the rigorous and subjective method or methodology for it fits to deal with complex social problems democratically and humanistically, while others are against his philosophical base for it believes that systems exist only in our minds or inter-subjectively and he negates that the society or management has its existence on its own right objectively. Some scholars argue for its localism and pluralism for they fit to the new experimentalism and pragmaticism, while others are against its 'foundationalism' because Checkland thinks that the 'soft' methodology is the general case of which 'hard' methodology is only a special case. I think in order to resolve these problems, the best way is to put SSM into a broader context of scientific methodologies including philosophy of social science to confirm the position of SSM in systems methodologies and more generally in scientific methodologies.

So the paper first analyses philosophically the transition from hard systems engineering to SSM. Then it discusses how the hard-soft division and transition from the former to the latter are represented in and push forward scientific methodologies. Finally, it discusses the status of SSM in systems methodologies to accommodate the above problems.


Classical systems engineering (sometimes called hard systems engineering after the rise of SSM) is a new research area which is an engineering methodology to analyse, devise, produce and manage large engineering systems, complex systems and human-machine systems. It arises from the research of the above problems in Radiobroadcast Company of America in the 1930s. Those methods, together with operational research, got a large success in military affairs in World War II. The RAND Corporation calls them 'systems analysis' and A. D. Hall, an engineer of Bell Telephone Laboratories, calls them 'systems engineering' in his famous book A Methodology for Systems Engineering published in 1962 which has given a primary synthesis of the discipline. Reflecting on the discipline of HSM, we can mainly get four basic presuppositions of the methodology of hard systems engineering (HSM for short):

(1) Systems exist objectively. If some systems that we need are not there, we can design them by systems engineering, and then they exist in the world outside us. All things in the world can be seen as system. The problem for us is how to describe them, study them and improve them.

(2) Systems can be well defined by clearing up their goals. In that way, systems can be seen as engineering to achieve the goals. For example, to send human being to the moon in 10 years is a system engineering of America in the 1960s called 'Apollo Program'.

(3) There is an optimal solution in every system. All we should do is to compare different projects according to the degree of their attainability of the system's goals and choose the best one. In order to find out the optimal solution, we can use mathematical methods such as operational research, linear programming, optimization method, decision theory and so on. H. A. Simon calls the process of optimization 'goal seeking' and Hall calls it 'six-step method of systems engineering' (defining the problem, selecting objective, synthesizing systems, analysing systems, selecting the best alternatives, planning for action).

(4) The basic dimension of hard systems engineering is the logic dimension. A. D. Hall suggests three dimensions (temporal, intellective and logical). The logical dimension expresses the logical connection of the 'six-step method': the former is the necessary condition for the latter, while the latter is the sufficient condition of the former in the series. As a result, all the six steps form a logical chain of goals and means in order to find the optimal solution. Karl Popper once called these epistemic steps 'the logic of the situation', and C. G. Hempel codifies the main features as: (1) Agent A was in situational circumstances of type C (psychological and physical) at time T. (2) A was a rational agent at T. (3) Any rational agent, when in circumstances of kind C, will with high probability do X. (4) Therefore, A did X at T (James, 1968, p. 261). Here doing something with high or highest probability is the same as the optimal solution of individual A instead of the social complex collectivity. So there are no dimensions of historical, cultural or social political analysis at all in hard systems engineering (Hall, 1969).

It is interesting to note that either Hall or Simon says frankly and bluntly that HSM with their four basic presuppositions talked above comes from a famous philosopher John Dewey. Simon says, 'generally speaking, intelligence activity precedes design, and design activity precedes choice. The cycle of phases is, however, far more complex than that this sequence suggests ... Nevertheless, the three large phases are often clearly discernible as the organizational decision process unfolds. They are closely related to the stages in problem solving first described by John Dewey: What is the problem? What are the alternatives? Which alternative is best?' (Simon, 1960, pp. 19-20). Hall emphasizes that those three questions are connected together. He argues that 'Dewey says that a problem well put is half-solved. Thus the problem statement itself may contain within it possible solutions. No problem situation is ever completely indeterminate, completely confused or completely obscure. Defining problems calls for as much creativity as solving them' (Hall, 1962, p. 94). Here 'Dewey's Model' and the six-step model of Hall are shown in Figure 1(A), and SSM Model is shown in Figure 1(B) below (A) in order to compare them. Those are the basic ideas of classical systems engineering that Dewey expresses in his book How We Think about 100 yeas ago (Dewey, 1910). How profound Dewey's idea is! Unfortunately, when systems engineering and operational research are introduced to China, Dewey's philosophy in HSM was thrown away, left only its applied mathematics. As a result, most systems engineers and teachers do not understand the background, nor can they understand why there is a paradigm transform from hard systems thinking to soft systems thinking.


After World War II, when systems engineering was applied from military engineering to civilian enterprises, a research community led by Peter Checkland in Lancaster University found that some of the situations are quite different from military operational research. It is difficult to use the objective language in human affair systems because there are no such common goals existing in the social world. Based on different Weltanschauung, different cultural backgrounds and different interests, participants involved into the problem situations have their own ambiguous or unambiguous aims for these situations and always contradict each other. Checkland once gave us a very interesting example. What is the objective of Common Agricultural policy (CAP) of European Economic Community (EEC)? The Treaty of Rome boldly declares that the CAP has three equally important objectives: to increase productivity in the agricultural industry; to safeguard jobs in the industry and to provide the best possible service to the consumer (Checkland, 1999, p. A6). However, problems cannot at all be resolved according to the above definition, because the situations are in contradiction. The next example is Manhattan Project. It refers specifically to the period of the project from 1942-1946 to produce atom bombs under the control of the U.S. Army Corps of Engineers. The scientific research was directed by the American physicist J. Robert Oppenheimer. As a military engineering, the project has its well-defined problem. But after the atom weapons have been produced successfully, the situation changes. Must we use these bombs to kill several hundred thousand people including so many women and children in Japan? The problem is unstructured and has caused a lot of debate even today. Facing problems as complex as the above ones, it is necessary to distinguish hard systems engineering methodology from SSM. The characteristics of SSM are as follows:

(1) There are no such objective systems outside our minds in the situation that can be described, researched and dealt with. The systems we face in SSM are neither of nonteleological natural systems nor original from human designed well-defined artificial technological systems, but are human affair systems or human activity systems. They originate from self-consciousness and genuine freedom of choice in selecting. To know these 'systems' is not to describe them, but to use hermeneutics to interpret them.

(2) In these systems, there is no clearly consentaneous objective, so that we cannot half-solve them even when we discover their problem situations. It is not to say that the participants as agents have no objective or aims, just the reverse, human beings can choose their action freely and purposefully, and thus form their problem situations. But what are the situations? What goals the systems must attain to? How to improve the problem situations? Different participants have different feeling, interpretation and modelling based on their worldviews and cultural backgrounds. SSM emphasizes that, due to the complexity and variety of human activity systems, there will be always pluralistic problem situations, pluralistic systems modelling and pluralistic solving.

(3) There are no optimal solutions in SSM. When we search for the solution of problem situations, all the varied solution models have been existent and contradicted each other. The approach to solve problems in SSM is to choose the models which are closely relevant to the real-world problem situation, exploring their underlying weltanschauung and cultures, comparing them with each other and comparing them with the real world, to cause debate within the relevant people in order to develop an accommodation among these relevant people and to improve some of the problem situations. Therefore, SSM inevitably causes a learning circle which itself is an organized learning system.

(4) There are two basic dimensions in SSM. The first one is the 'logic-based stream' of analysis. The second is the 'social-cultural stream' of analysis (see Figure 3(A)). Scientific research and operation research have twosideness. They are socially cultural and rationally logical, and they are scientific and humanistic.

Based on the above four points of views, Checkland (1999) gives a Figure 1n the book Soft Systems Methodology: A 30-Year Retrospective which is very similar to that of Dewey and Hall in form but essentially different in content. In order to compare these two clearly, I put them together in one figure (see Figure 1):


First, according to Checkland, Figure 1(A) is about the objective world, while Figure 1(B) is about our subjective world and the problem situations it causes. Second, Figure 1(A) gives top priority to 'defining the problem' and defining objective, while Figure 1(B) is neither problem definition nor objective definition at all. Third, Figure 1(A) is a functionalist model, which is to attain to the aims or goals, and to get the optimal solution, while Figure 1(B) is an interpretational model, which is to interpret the worldviews and cultures of the relevant actions in the system. Fourth, Figure 1(A) has three circles to denote that the system has three elements, among which every two elements has feedback connects described by the double arrow-line. But there is no feedback line departing from the problem solution to the indeterminate situation. It is not an occasional situation. Hall says that 'if the optimal system is good enough, the whole process is over' (Hall, 1962, p. 92). So, different from Figure 1(B), Figure 1(A) is not a learning circle. Finally, different from Figure 1(B), there is no cultural dimension in Figure 1(A), which relates to the scientism of some systems engineers and of the philosopher Dewey in some sense. Dewey thinks that desires, interests and their bases, namely, traditions and cultures are not the premises of valuation because social cultures can be changed and all of them can be reduced into some kind of energies in scientific test and valuation. He says, 'desires and interests must themselves be evaluated as means in their interaction with external or environing conditions ... Desires, interests and environing conditions as means are modes of action, and hence are to be conceived in terms of energies which are capable of reduction to homogeneous and comparable terms. Co-ordination or organization of energies, proceeding from the two sources of the organism and the environment is, thus, both means and attained result or 'end' in all cases of valuation, the two kinds of energy being theoretically capable of statement in terms of physical units'(Dewey, 1939, p. 53). According to this reductionism, if there is some 'dimension' of social consideration, it will be reduced to the logic dimension of natural science. So Figure 1(A), as a hard systems engineering methodological model, is single dimensional model. On the other hand, Figure I(B) shows that 'this version of SSM as a whole recognizes the crucially important role of history in human affair', and is 'also a cultural and political stream which enabled judgements to be made about the accommodations between conflicting interests which might be reachable by the people concerned and which would enable action to be taken'(Checkland, 1999, pp. A14-A15).


To summarize, Classical systems engineering originally comes from the research of the epistemic process in philosophy of science. The development from Classical (or hard) systems engineering methodology to SSM is a transformation of paradigms which, from the philosophical point of view, is the transformation from objectivism, scientism and functionalism to subjectivism, constructionism and hermeneutics. Non-understanding to this philosophical transformation causes the negation and omission of the importance of SSM.


Reviewing the development of philosophy of science in the 20th century, we discover surprisingly that the paradigm transformation of philosophy of science experiences the similar process to that of hard-to-soft systems methodology, even though we do!not know whether systems methodology simulates philosophy of science or vice versa. Because the object of scientific investigation is of objective nature, we ought to suppose that natural systems exist independently outside our minds and can be described by scientists. Checkland says, 'I am claiming only that natural systems are the evolution-made, irreducible wholes which an observer can observe and describe as such, being made up of other entities having mutual relationships'. '[but] as wholes ... this remains true' (Checkland, 1981, p. 113). So, there is a common criterion of truth for testing about scientific hypothesis and laws, which makes the Classical scientific methodology similar to HSM and consistent with it. Analytic empiricism including logical empiricism and Popper's empiricism arising in the 20th century just hold the same position as HSM. Analytic empiricism is called the classical school (received view) of philosophy of science just like that the HSM is called the classical school of systems methodology. But what is science? How does science operate? These are standard problems of philosophy of science. They are not only a kind of logical and experiential problems, but also clearly a kind of social problems. In other words, science is a partly socially institutionalized rational action. The knowledge system of human being, according to Checkland's systems typology, is a kind of human activity system which is intent to improve problem situations of knowledge and their application in society. Usually they are realized by scientific communities socially and are influenced by the worldviews and cultures of scientists heavily. These approaches show an important aspect towards the social dimension of science, which Checkland and SSM emphasize all along. Analytic empiricism omits and negates this aspect, which is why it loses the dominant position in philosophy of science. The situation begins with Thomas Kuhn's work in his book Structure of Scientific Revolutions (Kuhn, 1962) which challenges the analytic empiricism. As Kuhn says, 'its main themes might better be described more generally as a study of the problems raised by the transition to what's sometimes called the historical and sometimes just the 'soft' philosophy of science' (Kuhn, 2000, p. 90). After the 1960s, the historic-social school of philosophy of science including pragmatic perspectives arises. They become the mainstream of philosophy of science from the beginning of the late 20th century until now. So, it seems that 'hard' philosophy of science and 'soft' philosophy of science really co-exist and they coexist with HSM and SSM for a long time.

Similar to the division of HSM and SSM, let us compare 'hard' philosophy of science (the classical school) with 'soft' philosophy of science (the historic school) in the following three problems: the epistemic process; evaluation and choice of scientific theories and explanation and interpretation of science. Do they undergo some kind of transformation of paradigms? If so, can we analogize them with the 'hard'-'soft' transform in systems thinking?

Let us first discuss the epistemic process and the problem of theoretical evaluation and choice. In analysis empiricism, there are two kinds of cognitive models: one is observational facts--theoretical hypothesis--confirmation of hypothesis--new observation series of logic empiricism, and the other is supposed by Karl Popper, which is a series of the epistemic process, a famous linear formula: [P.sub.1]--TT--EE--[P.sub.2]. Popper says that scientific research begins with problem ([P.sub.1]), and in order to resolve that problem, tentative theories arise (TT) and then it comes to the crucial step, namely criticizing and falsifying the tentative theories to eliminate errors (EE). If the theory is successfully against the falsified, the problem will be resolved and a new problem arises ([P.sub.2]). First, the problems in science in Karl Popper's mind are usually well defined and widely recognized by scientists, such as the 23 problems of mathematics listed by David Hilbert in 1900 and the 'two clouds' of physics pointed by Lord Kelvin (1) in the same year. They are recognized by mathematicians and physicists at that time. Especially at the stage of the so-called 'normal science', most problems are the so-called puzzle-solving with their definitions and solving procedures decided by different paradigms. Popper does not negate them. Second, Popper points out that '... the sequence is not circular, the latter problem is different from that of the former in general. It is the result of the new situation'(Popper, 1979, p. 255). Third, Karl Popper improves his formula of the cognitive process of science in 1968, using problem situation instead of problem, changing tentative solving to multiple tentative in the following way:

But how can the multiple theories be evaluated and chosen? According to Popper, it is determined by severe experimental scrutiny according to falsification logic only. I think it loses the stream of cultural analysis ahistorically in understanding human activities in science. As a result, it is not possible to give social, cultural or political analysis or personal weltanschauung analysis to the multiple tentative resolvings of the problem situation and the theory choice. So, falsificationism of Popper and positivism of logic empiricism belong to the category of 'hard' systems thinking.

Of course, the main problems discussed by philosophy of science are how it can be developed, tested, evaluated and chosen, during which logic analysis and the experimental test are indispensable. But scientific research itself is a particular kind of human activity system or human affair system. 'These are less tangible systems than natural and designed systems' (Checkland, 1981, p. 110-111). In order to understand the epistemic process and its structure, it is necessary to make social-cultural analysis, political analysis and even personal worldview analysis. I use the two streams model of Checkland to analyse the situation of scientific affairs and soften Popper's formula in the following way (see Figure 3). First, I use problem situation [P.sub.1] not only to express well-defined problems in science but also to express ill-defined scientific problems including some kinds of social science problems. Second, we can give multiple tentative theories or hypothesizes with cultural analysis including science community analysis, social system analysis and even political analysis instead of viewing them as arising from random phenomena or free imagination and creation called discovery context belonging to psychology outside the research of philosophy of science. Third, the theory choice of science in rationality is one of the most important problems. It is not to be resolved on the logic criterion only. Some kinds of value criteria and the personality of the scientists are also important factors.

The work belongs to the historic-sociological approach of philosophy of science, particularly adopted by Thomas Kuhn and Paul Feyerabend. As we know, on how to evaluate and choose a good theory between all the competitive theories in the cognitive processes, analytic empiricism insists that the main criterion of theory choice is to see how accurate the consequences, deducible from a theory, agree with the results of experiments. But different theories may fit this criterion exactly in the same way. So to choose a theory, the experience criterion is only an underdetermined factor. Then some philosophers of science extend their criteria to include factors as accuracy, being consistent with other areas internally and externally in logic, having broad scope to explain phenomena, being simple to express, being fruitful to get new research findings, validity in application and so on. But all those criteria belong to the logic-based stream of analysis in Checkland's terms. Thomas Kuhn says even though we all agree with these criteria, different scientists have different opinions, different weights and different understandings to these criteria based on their cultural backgrounds and personalities.

Thomas Kuhn says, 'Kepler's early election of Copernicanism was due in part to his immersion in the Neoplatonic and Hermetic movements of his day; German Romanticism predisposed those it affected toward both recognition and acceptance of energy conservation; nineteenth-century British social thought had a similar influence on the availability and acceptability of Darwin's concept of the struggle for existence'. This is what Checkland says as the other stream of analysis that models being put forward and chosen is influenced by culture. Kuhn continually says: ' Still other significant differences are function of personality. Some scientists place more premium than others on originality and are correspondingly more willing to take risks; some scientists prefer comprehensive, unified theories to precise and detailed problem solution of apparently narrower scope. Differentiating factors like these are described by my critics as subjective ... let me for the moment accept it. My point is, then, that every individual choice between competing theories depends on a mixture of objective and subjective factors, or of shared and individual criteria. Since the latter have not ordinarily figured in the philosophy of science, my emphasis upon them has made my belief in the former hard for my critics to see' (Kuhn, 1998a, p. 263). The concepts of 'mixture of objective and subjective factors' and the combination of 'shared and individual criteria' of Kuhn are the same as the two streams of SSM. Figure 3 expresses some ideas which I call 'soft' thinking of philosophy of science and compares it with the two streams model of SSM. I believe most areas of science, including ecosystems science and social science, can fit with them, based on hermeneutics, historism and holism. SSM and the historic approach of philosophy of science have a common characteristic, which are their pragmatic perspectives. It seems to me that pragmatism has different tendency to different authors, it can also be divided into 'hard' pragmatism and 'soft' pragmatism. For example, Dewey was inclined to hard thinking at least at the beginning of the 20th century. He reduces the value dimension to the scientific dimension as I have talked about in the previous section. But William James' consideration shows some kind of soft thinking. He exactly recognizes science and value as two steams of analysis. He says, 'moral questions immediately present themselves as questions whose solution cannot wait for sensible proof.

A moral question is a question not of what sensibly exists, but of what is good, or would be good if it did exist. Science can tell us what exists; but to compare the worths, both of what exists and of what does not exist, we must consult not science, but what Pascal calls our heart. Science herself consults her heart when she lays it down that the infinite ascertainment of fact and correction of false belief are the supreme goods for man. Challenge the statement, and science can only repeat it oracularly, or else prove it by showing that such ascertainment and correction bring man all sorts of other goods which man's heart in turn declares. The question of having moral beliefs at all or not having them is decided by our will' (James, 1897, Section 9).

To summarize, SSM inspires philosophy of science to take science not only as the result of exploring nature, but also as the product of social and cultural backgrounds. It is better not to take the multi-tentative hypotheses as stochastic and free creates but to give them cultural and social-constructional analysis. Criteria of theoretical evaluation not only include the logical and empirical criteria but also contain that of valuation. As to the choice of theories, philosophers of science ought to consider the role of personality and preference of scientists. The logical approach and cultural approach of philosophy of science have both been studied by philosophers for a long time, and it is time to integrate them together.


Hermeneutics is the philosophical base of soft systems science, but what is the status of hermeneutics in science or in philosophy of science? Hermeneutic (or interpretation) and explanation are two kinds of methods to understand the world. They can both be used in natural science and social science. Explanation applies scientific laws especially causal laws to account phenomena, while interpretation introduce hermeneutics to understand human will, intention and social rules to 'read' human behaviours as a text. SSM is hermeneutic or interpretative mainly, because its critical step is to interpret or read the social regulations, cultural backgrounds and weltanschauung underpinning and influencing human behaviours in order to understand the problem situation and various models. This approach is strange to philosophy of natural science which originates to describe and analyse natural phenomena several decades ago. As the development of philosophy of science and its extending from natural science to social science, we have to face hermeneutic problems. The research of natural phenomena is mainly based on natural facts and their laws gained by observation, experiments and mathematical application. In order to confirm them, scientists must retest them by doing similar observation and experiments at other places and in different ways to examine their logic as wide scope as possible. It seems unnecessary to interpret natural phenomena and natural laws by hermeneutics, because natural concepts and natural laws are the same to different culture backgrounds. However, this is not always the case. For the Greeks, concepts of heavenly objects are divided into three kinds: stars, planets and meteors. We also have categories with those names today. For the Greeks, the sun, the moon, Jupiter, Mars and so on are put into the same kind of planets, which must be interpreted by their culture background; otherwise the meaning is quite different and may cause some ambiguity.

In some sense, hermeneutics has no different roles between natural science and human science. Most natural concepts are shared by the same paradigm in certain scientific community which itself is a kind of culture or subculture. The transform of paradigms and the training of the next generation to hold the new paradigm are realized by the hermeneutic method or 'reading the text'. Even though in modern textbooks of physics, there are different interpretations to the same Schrodinger equation: Copenhagen school has its own probability interpretation, while Einstein school has its own determinist interpretation. Once a paradigm or hermeneutic basis has been constructed and come to the period of normal science, scientists can use the hermeneutic basis to do experiments, observe, describe natural phenomena and solve puzzles. It is not necessary to reinterpret their paradigm or hermeneutic basis again. So hermeneutics has its special but limited position in natural science rather than in social science. Kuhn says, 'the natural science, therefore, though they may require what I have called a hermeneutic base, are not themselves hermeneutic enterprises' (Kuhn, 1998b, p. 133).

Compared to natural science, the function of hermeneutics is quite different in social sciences. In natural science, the main work is to discover causal laws to explain phenomena, but in social sciences, the basic facts and basic research domains are human actions and human activities, which are dominated by intention, purpose, some kinds of rules or norms and even conventions and cultures. If you describe thunder in nature, you must not interpret it as the anger of Thor. But if you see some people put a piece of paper into the box, even though you picture it in detail, you still cannot understand the activity, because you must interpret the intention of this event, which is whether he is casting a vote or doing something else. Social facts have their own meaning dimension. Only when you use the meaning dimension to interpret or 'read' the meanings, intentions, norms, institutions and social cultures, can you say that you characterize these social facts. Social experiential facts are socially constructed, dominated by subjective agents and their values. They are not only received by perception or observation, but also through 'experience', 'empathizing' and interpretation. Interpretation accounts the meaning of human behaviours and human activities. So, social experiential facts are the entanglement of facts and values, and of perceptions and concepts. Intentions and purposes, and experiences and interpretations are individually subjective phenomena. Different people will have different interpretations to the problem situation and different modelling to improve the situation according to their different cultures, different interests, different life backgrounds and different personalities. So social experiential facts are, in some sense, local, distinguishing and non-repeating. That is why SSM emphasizes this point and puts it in its central position. Checkland says, 'there were potentially as many description of the relevant system as there were concerned actors in the problem situation' (Checkland, 1981, p. 278). This approach is very important for philosophy of social science. No historical facts or opinions are exactly the same, therefore the recognized criteria of social facts are only 'soft' criteria. How to determine these criteria is quite a new problem of philosophy of social science.

There are two kinds of methods and two kinds of structures to explore and understand the society. One is the explanatory method, which is to discover social laws to deduce social facts. It seems that Checkland omits this methodology. We will talk about it in the next section. The other method is the interpretative method, which is to unscramble or 'read' the special and local social phenomena, interpret their meaning and find out the whole structure in order to understand and interpret various local social phenomena. It is a circle from local phenomena to the whole structure and reversely from the whole phenomena to understand the local and the parts. It is called the hermeneutic circle. The whole method of SSM is hermeneutics that can give powerful support as a practical base to philosophy of social science. So, it can be said that SSM is one resource of philosophy of social science in theory and in practice.

Then how can SSM help relevant agents to resolve relevant problems to improve the problem situation they are involved in? The basic approach is to use systems thinking or holons thinking to construct and choose some of the models, interpret these models, use the models to guide a debate about the problem situation through the leaning circle to get the accommodated projects which improve the situation and are systemically desirable and culturally feasible and then realize the projects. The whole process includes social communicative rationality and discourses ethics of J. Habermas. We hope to search out the way to arrive at the rationality of social sciences through the research of SSM.

To summarize, SSM tells us that cognition is a learning circle, understanding is hermeneutic, the action to improve problematic situation is systemically reasonable and culturally doable and conflicts of interests can be harmonized and accommodated. These ideas of SSM have become one of the theoretical and practical bases of social science and philosophy of social science.


As one of the methodologies dealing with general social problems, SSM has its important position in philosophy of natural (and social) science, but on the other hand, it also has its limitations. Generally speaking, it fits better to social affair systems which have strong subjectivity to be interpreted and have distinct conflicts in interests and in culture to be accommodated, but it is weak in dealing with some social and technological systems which have common interests and objective and universal criteria of values between its members. As a result, SSM plays its role only in part of scientific methodologies and management methodologies.

Originally, Checkland creates SSM to deal with some kinds of special social systems (ill-defined problem systems, non-optimized problem-solving systems and so on), but after a long-term research and consideration, he finally generalizes SSM, changing his definition of 'system' from an objective term to a subjective one. He says, 'Firstly, in getting away from thinking in terms of some real-world systems in need of repair or improvement, we began to focus on the fact that, at the higher level, every situation in which we undertook action research was a human situation in which people were attempting to take purposeful action which was meaningful for them. Occasionally, that purposeful action might be the pursuit of a well-defined objective, so that this broader concept included goal seeking but was not restricted to it' (Checkland, 1999, p. A7). Here 'goal seeking' means hard systems engineering which Checkland takes as occasional factors included in SSM. He says, 'the "soft" methodology is seen to be the general case of which 'hard' methodologies are special cases'(Checkland, 1981, p. 191). He continues to talk about this problem about 30 years later. He says, 'this means that the 'soft' approach does not throw away the 'hard' thinking of the 1950s. It subsumes it as a special case within the broader approach' (Checkland, 2010). Is this conclusion consistent with his view about the relationship between HSM and SSM? What is the position of SSM in systems methodology? I summarize his main ideas more broadly and get the following five points:

(1) SSM subsumes hard systems methodology (HSM) as a special case. The idea of the hard-soft relationship contradicts with his other formulations, such as:

(2) SSM and HSM fit to different areas respectively. Checkland says, 'in the literature it is often stated that 'hard' systems thinking is appropriate in well-defined technical problems and that 'soft' systems thinking is more appropriate in fuzzy ill-defined situation involving human beings and cultural considerations. This is not untrue' (Checkland, 1999, p. A10). But this point of view contradicts with the first one explicitly, because 'well-defined problems' and 'ill-defined problems' is a logical dichotomy, the former cannot be 'subsumed' by the latter, and the latter cannot deduce to the former, even though occasionally. Let me continue to list other views about that problem Checkland asserts.

(3) HSM is applied to the world, while SSM is not. Checkland says, in HSM, 'the word 'system' is used simply as a label for something taken to exist in the world outside ourselves', and 'the world can be taken to be a set of interacting systems'. But in SSM, 'the use of the word 'system' is no longer applied to the world, it is instead applied to the process of our dealing with the world' (Checkland, 1999, pp. A10-A11). Here, again, there is a Cartesian subject-objective dichotomy. If the real world has no system at all, and systems are only in our minds, how can the real world as a special case be subsumed by our minds? Checkland has some excellent systems thinkings which are summarized from a lot of literatures of systems science. For example, his concept of holon means 'a whole having emergent properties, a layered structure and processes of communication and control within principle enable it to survive in a changing environment' (Checkland, 1990, p. 22). But how can we understand holons as well as emergence, layered structure, communication, control, environment and survival are subjective and not existing in the world?

(4) In Checkland's famous book Systems thinking, Systems practice (Checkland, 1981, p. 122), there is an excellent section about 'systems typology' which shows systems classes in the following way: 'The systems map suggests that the absolute minimum number of systems classes needed to describe the whole of reality is four: natural, designed physical, designed abstract and human activity systems'. The relationship between the four systems classes is like this: natural systems, including human beings, originate from the universe and the processes of evolution. As human being, we may investigate, describe and learn from natural systems, create and use designed physical and abstract systems and seek to form human activity systems. In this picture, human activity systems can neither subsume natural systems nor can it subsume artificial systems (designed physical systems or technological systems). As a result, SSM mainly dealing with human activity systems and social communicative systems cannot replace and include HSM as a special part.

(5) In the same book (Checkland, 1981, p. 280), Checkland quotes the two dimensions classifying social sciences of Burrell and Morgan (1979). The abscissa (horizontal coordinates) is from subjective to objective, and the vertical coordinates is from the sociology of radical change to the sociology of regulation. Thus social sciences are divided into four quadrants. The first quadrant is about radical humanism such as anarchism and individualism. The second quadrant is about radical structuralism, including the social conflict theory. The third quadrant is occupied by functionalist sociology and social systems theory. The social theory implied in SSM would lie in the fourth quadrants with hermeneutics and phenomenology. In this classification, hermeneutic sociology with SSM, functionalist sociology with HSM and systems analysis locate in different quadrants separately, which means functionalist sociology with HSM cannot be taken as a special case included in hermeneutic sociology with SSM.

The theoretical development and the practical application of SSM in the past 30 years show the above 2, 4 and 5 are relatively appropriate, while 1 and 3 are obviously inappropriate. There is a serious problem about concepts that ought to be paid attention to. Almost all of us agree that 'today it is 'systems thinking' in its various forms which would be taken to be the very paradigm of thinking holistically' (Checkland, 1999, p. A3), and 'hard' systems paradigm' and soft systems paradigm are taken as basically different paradigms (Checkland, 1999, p. A10), but what does it mean? According to Thomas Kuhn, different competing paradigms are incommensurable, paradigm changes do cause scientists to see the world of their research-engagement differently, just like the gestalt switch that what were ducks in the scientist's world before the revolution are rabbits afterwards (Kuhn, 1962, p. 111, 150, 151). Although it is some kind of visualized expression, different competing paradigms have different theoretical frames, different standards of science, different vocabularies and different lexical taxonomic terms. There are no logical connect between them or the so-called logical continuation broken. If the 'hard' methodologies as the special cases can be deduced from the 'soft' methodologies as the general cases, there will be no paradigm differences of 'hard'--'soft' division at all.

Based on Checkland's 'systems typology' (Checkland, 1981) and Burrell and Morgan's 'multiple paradigm model analysis of social theories', Michael C. Jackson develops a metamethodology of 'total systems intervention' called 'critical systems thinking and practice' to deal with the relationship of various systems methodologies. He calls this meta-methodology 'creative holism' in order to treat various systems methodologies in a holistic and balanced way. The work done by M. C. Jackson is metatheoretical, meta-methodological or in metaparadigms. The problems 'critical systems thinking and practice' discussed are not about the systems themselves (which belong to first-order systems methodologies), but are about how to compare, evaluate, select and use various systems methodologies or systems paradigms. It belongs to second-order systems methodologies or second-order systems paradigms (Fan, 2010, this issue). On the other hand, the discussion is rationally possible, because the cognition of the incommensurability of various paradigms is only 'local incommensurability', not full incommensurability, otherwise the comparison and transition are irrational. According to Thomas Kuhn, two paradigms being incommensurable denotes that there are some kinds of taxonomic terms that have meaning-overlap to different theories and there are no common language to which both theories can use to translate the terms from one theory to others without residue or loss. That is why the original meaning of incommensurability is a 'local form', using the phrase 'no common measure' of mathematics to express them metaphorically. The taxonomic term 'planets' is a good example to show the incommensurable or untranslatable of Copernican theory and Greek astronomers (See the prior section). So, T. Kuhn says, 'incommensurability thus becomes a sort of untranslatability, localized to one or another area in which two lexical taxonomies differ' (Kuhn, 2000, p. 93). A matter worthy of note is that local incommensurability of paradigms means there are other aspects of paradigms: local commensurability. M. C. Jackson puts various systems methodologies into two dimensions (increasing complexity and increasing divergence of values) and six quadrants (simpleunitary, complex-unitary, simple-pluralist, complexpluralist, simple-coercive, complex-coercive) to analyse their positions and roles. The analysis is about the relations between different paradigms and different problem situations. If every different systems methodology is ideally suited to the analysis of a particular problem situation, and just falls to the quadrant they occupies exactly, then various systems paradigms are not incommensurable at all. So, if there are some kind of incommensurability in systems methodologies, M.C. Jackson's two dimensions and six quadrants cannot express them. For example, to the classification of systems methodologies of M. C. Jackson, 2000, pp. 357-362. Fig. 10.1 and 10.2, Jackson, 2003, Fig. 2.1 and 2.2, we can ask the following questions: How and why soft systems thinking do not cover the simple-coercive and complex-coercive problem situation? How and why systems dynamics and organizational cybernetics do not fit to the simple problem situation? Are they not related to others as the competing paradigm? If we want to solve or to accommodate these problems, we need to extend Jackson's matrix in the following way (see Figure 4). The belts formed with broken lines are called cross-paradigm belts that mean in the areas or scopes which have meaning-overlap of different paradigms arise around here in the same area, and transition of paradigms arises here, too. If the left vertical line and the right vertical line are put together in Figure 4, the matrix will become a cylinder. It shows every quadrant crosses and overlaps with each other. As a result, all the incommensurabilities can find a place in the cross-paradigm belts.


Here I use the cross-paradigm belt to deal with Jackson's grid of problem contexts, and the preliminary classification of systems approaches of Jackson, 2000, Fig. 10.2 can be treated in the same way.

I think that the relationship between HSM and SSM is neither the general-special relation, nor the relation of inter-consistency, but is the relation of inter-exclusive, inter-competitive (in some of the areas) and inter-complementary (in the other areas). It is to use multi-methods and cross-scope methods to resolve multi-problems and improve multi-problem situations human being face. It is not 'anything goes' of Feyerabend, but anything goes to resolve problems and improve problem situations if they can. In a sense, we insist pragmatism and do not refuse any hard systems engineering methodologies if they are available. I think there are two reasons: (1) What is human activity systems? In the classification according to the ideal type of Max Weber, it denotes a social whole in which many human activities are based on their self-consciousness and free will purposefully related to each other to form a set, but it cannot exist in isolation with its natural environments or other human activities with designed physical systems, namely technical systems or productivity systems using Marxist terms. So, concrete human activity systems always cross or combined with natural systems and artificial natural systems. For example, railway network systems and atomic power engineering systems are human activity systems, artificial natural systems and parts of eco-systems at the same time. The methodology to deal with those systems must be pluralistic and combinational. (2) Although human activity systems are formed by purposive and intentional activities of human being and, therefore, must use mental science of Delta or hermeneutics to deal with, as the result of some causality, human mental activities must have their mental causes or material causes, and it seems that human activity systems are dominated by some of their own objective social laws. The commutative activities of commodities between people can be seen as some kinds of activities based on the subjective preferences of human being, but the subjective preferences of thousands of people will form some kind of objective needs to determine the market operation. It represents the objective economic laws. So, the subjective hermeneutic method of SSM is necessary but not sufficient. It should be combined with other methods or methodologies to deal with the problem situations of human activity systems that are subjective, objective and the subjective-objective entanglement systems.

SSM has great significance not only in theory but also in practice, especially to the practices in institutional reform and openness. But many people in educational domains of management have not yet realized the management paradigm transformation from HSM to SSM in China. Academician Qian Xuesen says that systems engineering belongs to the 'discipline of technology' or 'discipline of engineering' (Qian, 1986, p. 266), and he has not differed SSM from HSM. Some people who teach management omit SSM, especially its philosophical base, namely Dewey and James' pragmatism, Weber Max's hermeneutic sociology, Habermas' social communicative theory and so on. It is the problem to be resolved.

DOI: 10.1002/sres.1022


Burrell G, Morgan G. 1979. Sociological Paradigms and Organizational Analysis. Heinemann Books: London.

Checkland PB. 1981. Systems Thinking, Systems Practice. Wiley: Chichester.

Checkland PB. 1999. Soft systems methodology: a 30-year retrospective. In Soft Systems Methodology in Action, Checkland P, Scholes J. Wiley: Chichester; A1-A66.

Checkland PB. 2010. Researching real-life: reflections on 30 years of action. Systems Research and Behavioral Science 27: 129-132.

Checkland PB, Scholes J. 1999. Soft Systems Methodology in Action. Wiley: Chichester. Dewey J. 1910. How We Think. D.C. Heath & Co: Lexington, Massachusetts.

Dewey J. 1939. Theory of valuation. In International Encyclopedia of Unified Science, Vol II, No.4. University of Chicago Press: Chicago.

Habermas J. 1981. The Theory of Communicative Action. Beacon Press: London.

Habermas J. 1990. Discourse ethics: notes on a program of philosophical justification. In Moral Consciousness and Communicative Action. MIT Press: Cambridge.

Hall AD. 1962. A Methodology for Systems Engineering. Van Nostrand Reinholdt: Princeton, NJ.

Hall AD. 1969. Three dimensional morphology of systems engineering. In Contributions to a Philosophy of Technology, Rapp F. (ed.). D. Reidel Publishing company: Holland.

Jackson MC. 2000. Systems Approaches to Management. Springer: London.

Jackson MC. 2003. Systems Thinking: Creative Holism. Wiley: Chichester.

James JL. 1968. The logic of the situation. Philosophy of Science 35(3):

James W. 1897. The Will to Believe and Other Essays in Popular Philosophy. John Wilson and Sons, Cambridge, USA. p. 22.

Kuhn TS. 1962. The Structure of Scientific Revolutions. University of Chicago Press: Chicago.

Kuhn TS. 1998a. Objectivity, value judgment, and theory choice. In Science Reason and Reality. Issues in the Philosophy of Science, Rothbart D. (ed.). Harcourt Brace: New York; 260-273.

Kuhn TS. 1998b. The natural and the human science. In Introductory Reading in Philosophy of Science, Klemke ED, Hollinger D, Rudge, DW (eds). Prometheus Books: Buffalo, NY.

Kuhn TS. 2000. The Road Since Structure. University of Chicago Press: Chicago.

Popper KR. 1979. Objective Knowledge: An Evolutionary Approach. Clarendon Press: Oxford.

Qian X. 1986. On Systems Engineering. Hunan Science and Technology Publishing Company: Hunan (in Chinese).

Simon HA. 1960. The New Science of Management Decision. Harper and Rowe: New York, NY.

(1) He said: 'The beauty and clearness of the dynamical theory, which asserts heat and light to be modes of motion, is at present obscured by two clouds.' the first one of cloud is Michelson-Morley experiment, the second cloud is the experiment of black body radiation.

Zhang Huaxia *

Department of Philosophy, Sun Yat-sen University, Guangzhou, China

* Correspondence to: Zhang Huaxia, Department of Philosophy, Sun Yat-sen University, Guangzhou 510275, P. R. China. E-mail:
COPYRIGHT 2010 John Wiley & Sons, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Research Paper
Author:Huaxia, Zhang
Publication:Systems Research and Behavioral Science
Article Type:Report
Geographic Code:9CHIN
Date:Mar 1, 2010
Previous Article:A revolution in the field of systems thinking--a review of Checkland's system thinking.
Next Article:An approach applying SSM to problem situations of interests conflicts: interests-coordination SSM.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters