- Some History
The people considered to be the founders of the General Systems movement are Ashby, Bertalanffy, Boulding, Fagen, Gerard, and Rappoport [Banathy]. The general systems movement was initially concerned with finding solutions to emerging complex problems; affecting humanity by identifying principles that are common to several disciplines. These principles were described by von Bertalanffy as abstractions and conceptual models rather than superficial analogies. These abstractions and models were thought to enable the collaboration of specialists from different fields to approach problems which no single specialist or discipline could. The General Systems Movement evolved into the study of systems, i.e. Systems Science, as can be seen from the evolution of the movementis name: the society for the Advancement of General Systems Theory; the Society for General Systems Research; International Society for Systems Research; International Society for the Systems Sciences. During this shift that now spans half a century, the systems sciences movement have subsumed several models such as cybernetics, etc.
The process through which the systems sciences porpose to approach complex problems is called systems inquiry. Banathy defines three interrelated domains comprising systems inquiry: Systems Theory, Systems Philosophy and Systems Methodology. System theory evolved from General Systems Theory (GST, von Bertalanffy). Systems theory aims at defining common principles of systems while understanding their specificities. System theory promotes transdisciplinary research to approach complex problems. Systems philosophy provides an explicit worldview to underlying the theories and methods of the systems sciences. Systems methodology defines ways of approaching problems depending on the nature of these problems (engineering problems, human systems problems etc).
- The Key Ideas of Systems Thinking
The best summary I have found of the key ideas in systems thinking are Morin's 7 principles necessary for reasoning about complex issues (freely translated from French):
1. Le principe systemique ou organisationnel: The systemic principle through which we consider the whole as well as the parts. The inter-relationships of the parts define so called emergent properties; which cannot be found in any of the parts. The whole is thus more than the sum of the parts but also less than this sum because the whole is constrained by the inter-relationships of the parts. Stated as it is above emergent properties seem to be related to stuff in the system. However, as noted by Weinberg, emergent properties should not be thought of as stuff in the system but as emerging from an observer's relationship with the system as a whole [Weinberg 75, p. 60].
2. Le principe hologrammatique: The hologrammatic principle which states that not only does the whole contain the parts but that each part is a reflection of the whole. For example, a person is a part of society but the whole society is reflected within each person through their language, culture norms, etc.
3. Le principe de la boucle retroactive: This is the cybernetic feedback loop through which a system regulates itself thus leading to its autonomy. The classical example of feedback loop is the thermostat regulating the temperature of some substance. The feedback may be negative as in the case of the thermostat where reactions are dampened and positive in which case reactions are amplified as in the case of violence cycles.
4. Le principe de la boucle recursive: This is the generative cycle through which a system generates its parts by is also generated by its parts, also called autopoietic process by Maturana and Varela. Thus, society produces individuals but individuals generate society.
5. Le principe d'auto-eco-organisation: The principle which defines open systems in the sense of von Bertalanffy. An open system is autonomous but has interactions with its environment. The system draws energy and information from the environment in order to maintain its autonomy. Thus the system is inseparable from its environment. Morin further notes that the a key idea of auto-eco-organizing systems is that their life is inseparable from death. Thus, living organizm depend on the life and death of their cells. Life and death are complementary and antagonistic notions.
6. Le principe dialogique: The association of two antagonistic notions which seem to exclude each other but which co-exist. As an example, consider life and death (above), order and disorder, peace and war etc.
7. Le principe de la reintroduction du connaissant dans toute connaissance: As stated by Maturana and Varela: Everything said is said by someone.This is also the key idea of second order cybernetics, i.e. the notion that any observation is made by an observer. Thus any model is a model made by someone and will not be the same as a model done by someone else. In order to understand the model we usually also need to understand the modeler. This notion can be easily linked to constructivism which states that reality is not imposed on people but that every individual (re) constructs their understanding of reality.
In general, Morin states that thinking about complex issues requires an act of distinguishing and linking rather than separating and isolating as is done in rational cartesian analysis. The idea here is to distinguish between entities in order to understand them but not to treat them in isolation from other entities. An entity cannot be understood without understanding its relationships with other entities.
Le Moigne defines three more principles attributed to complex systems
- The time transformation, meaning that a system is tranformed through time in an irreversible way.
- The investigation focuses primarily on behavior (actions) rather than on apparently invariant objects.
- The teleologic principle which states that to understand a system we need to understand its purpose. Note that we replace the term teleology and its derivatives by the term teleonomy as suggested by Checkland to avoid the confusion with the vitalists use of teleology to explain the behavior of natural phenomena.
The other set of ideas are found in the quotes, insights, jokes and so called laws found in Gerald's Weinberg's extraordinary book called "An Introduction to General Systems Thinking".
By definition systems science principles can be used in any scientific field. However, these principles need to be adapted to each field because they cannot be used as simple generalities. When we do this we are in danger of producing yet another specialty which runs contrary to the basic ideas of general systems thinking. We need to apply the principle of distinguishing and linking so that the reloationships between the new model and related models as well as the general principles from which the model was derived are explicitly stated.
Our application area is the understanding and design of information systems for human organizations. In this respect we are mostly influenced by the general systems science principles and by so called soft systems approaches [Checkland], [Vickers] [Conklin], [Shum]. See the defitinition of wicked problems in complexity.
- Annotated bibliography
L. Von Bertalanffy, General System Theory. New York: George Braziller, 1968
This book is considered the seminal text of systems science. It consists of the collection of articles written by Von Bertalanffy from 1940 to 1968. These articles mainly presents the principles of General Systems Theory (GST) as they apply to Biology, psychology, the social sciences etc. It provides an excellent reference to the principles and philosophy of GST but does not provide specific methods for approaching the different fields of inquiry. It does, however, act as a mind opener for people who were trained in the traditional scientific methods but are receptive to new ways of looking at the world. The result is that the reader has more questions than answers after reading the book.
My favorite chapters are:
- Chapter 4 Advances in General System Theory
- Chapter 6: The Model of Open System
- Chapter 10: The Relativity of Categories
Checkland, and J. Scholes, Soft System Methodology in action. Chichester UK: Wiley, 1990.The first part of the book presents a synthetic description of Soft System Methodology (SSM) as it existed in the late 1980s. The final pages of this part make the link between SSM and information system design. These pages are useful in understanding that present day computers are not intelligent machines in the same sense that humans are intelligent and that computers do not attribute meaning to information but rather manipulate symbols. However, Checkland and Scholes acknowledge that the detailed linking of SSM to detailed design of computerised data manipulation systems has not yet been accomplished (p. 57). Also, The view of an information system as data manipulation system makes them focus on information flow models and data structures rather than objects and/or goals.
The second part of the book gives numerous detailed examples of SSM projects.
P. Checkland, S. Holwell, Information, Systems and Information Systems, making sense of the field. Chichester, UK: Wiley, 1998.In this book Checkland and Holwell position SSM with regard to what they call the orthodox model of organizations as goal seeking entities where the main activities are decision making, and of information systems as helping managers to make decisions. Checkland and Holwell attribute the origins of this prevalent model to H.A. Simon's work on decision making. Checkland and Holwell argue that this model takes the activity of defining the goals for granted, assuming that organizations have pre-established goals and that everybody in the organization understands and agrees with the stated goals. According to Checkland and Holwell most textbooks on information systems use this model implicitly or explicitly without challenging its validity or its effects on the research and practice of information systems development.
To this orthodox view that they refer to as positivist Checkland and Holwell oppose the interpretivist view. They attribute the origins of this interpretivist view to Sir Geoffrey Vickers (see Vickers 87 below). In line with this interpretive stance Checkland and Holwell define their model of data, information and knowledge. They depart from the popular model of data as unstructured facts, information as structured data and knowledge as contextualized and operational information by noting that information is captured by people according to their interpretations. They thus add the intermediate concept of capta, which is the information the part of the information that the observer chooses (knowingly or not) to accept from the environment.
P. Checkland, Systems Thinking, Systems Practice. Chichester, UK: Wiley, 1999.This book is a reprint of the Checkland's classical book by the same name which was originally published in 1970s. The first part of the book referred to as an appendix is completely new.
H.R. Maturana and F.J. Varela, The Tree of Knowledge. Boston, MA: Shambhala, 1998.The essence of this beautiful book is best exaplained in the first paragraphe of its preface: The book that you now hold in your hands is not just another introduction to the biology of cognition. It is a complete outline for an alternative view of the biological roots of understanding. From the the outset we warn readers that the view presented here will not coincide with those they are likely to be familiar with. Indeed, we will propose a way of seeing cognition not as a representation of the world out there, but rather as an ongoing bringing forth of a world through the process of living itself. Maturana and Varela show that the brain cannot be conceived of as an information processing machine, i.e. that the metaphor of the brain as a computer is wrong p. 169. The idea is that we are not simply reacting to outside influences but we construct our own reality. This is the essence of constructivism, which, by the way, is not mentioned explicitly in the book.
J.G. Miller, Living Systems. Niwot, CO: The University Press of Colorado, 1995.
M.D. Mesarovic, D. Macko and Y Takahara, Theory of Hierarchical, Multilevel, Systems, New York: Academic Press, 1970.
J-L, Le Moigne, La modelisation des systemes complexes. Paris: Dunod, 1990.
J-L, Le Moigne, Les epistemologies constructivistes, Que sais-je, PUF, 1999.
E. Morin, J-L. Le Moigne, L'intelligence de la complexite, L'Harmattan
H.A. Simon, The Sciences of the Artificial. Cambridge, MA: MIT Press, 1996.
G. Vickers, Policymaking, Communication, and Social Learning, New Brunswick NJ: Transaction Books, 1987.
G. M. Weinberg, An Introduction to General Systems Thinking. New York: Wiley & Sons, 1975.This book is filled with so much wisdom and wit that it is difficult to summarize it without extracting large quotes from it. Since I cannot quote half of the book here, I will quote and comment the absolutely best paragraphs. However, I really encourage anybody not only people interested in sytems thinking to read this book.
The chapter on Boundaries and Things beautifully shows that the separation between system and environment is difficult to do even for physical systems. Weinberg's example of considering human hair as part of the human body or as part of the environment is troubling but effective in showing that the distinction between system and environment is due more to our modeling aim than to reality itself. Moving beyond physical systems, Weinberg notes that when [...] we begin to encounter systems without well-defined physical boundaries, the boundary metaphor lures us more easily into attractive, but false reasoning.
T. Winograd, F. Flores, Understanding Computers and Cognition, 1986.
Shum et al. Graphical Argumentation and Design Cognition
To be reviewed:
- Societies and Journals
International Society for Systems Research (ISSS)
The International Society for General Semantics
L'Observatoire pour l'Etude de l'Universite du Futur
- Courses and Definitions
Eric Schwartz introduction to sytems thinking (in French)
Systemique (Rene Berger)
Summary by Gil Regev
RM-ODP | Goal Modeling | Business Plan | General Systems Thinking