Guohua Bai (Ph.D. Guest Professor)

 

2010-02-28

My reading notes

 

 

·  Activity Theory 

Notes of Systems (cybernetic) Theory Study

Part 1 ; Part2; Part 3; Part 4


PART ONE


Ludwig Von Bertalanffy

The History and status of General systems Theory

Reprinted from George J.Klir, ed., Trends in General system theory, New york: Wiley-Interscience, 1972

In order to evaluate the modern "systems appoach," it is advisable to look at the systems idea not as an ephemeral or recent technique, but the context of the history of ideas.

Philosophy and its descendant, science, was born when the early Greeks learned to conider or find, in the experenced world, an order or kosmos which was intelligible and hence, controllable by thought and rational action.

Aristotle´s statement "The whole is more than the sum of its parts," is a definition of the basic system problem which is still valid. Arisotelia teleology was eliminated in the later development of Western science, but the problems contained in it, such as the order and goal-directednes of living systems, were negated and by-passed rather than solved.

Hierarchic order: Christian, Dionysius, leibnize.

Hegel and Marx: dialectic structure of thought and of the universe it produces: the deep insight that no proposition can exhaust reality but only approaches its coincidence of opposites by the dialectic process of thesis, antithesis, and synthesis. sixteenth-sventeenth centuries Scientific Revolution can be formulated by Gallleo as resolutive method: that is, to break down every problem into as many separate simple elements as might be possible. This conceptual paradigm of science forms the foundation to modern laboratory work. It was at the root of the enormous success of physics and the consequent technology.

To deal with order or organization, there were two highly successful principal ideas. One was the comparison with man-made machnes, e.g., Descartes´s bete machine. The theory of the living organism as machine in its various disguises, from a mechanical machine or clockwork in the early explanation in the seventeenth century, to the later conceptions of organism as caloric, chemodynamic, and cybernetic machine, provided explanations of biological phenomena from the gross level of the physilogy of organs down to the submicrosopic structures. Another was to conceive of order as product of chance, as expressed by the Darwinian´s idea of natural selection. The fight on the concept of organim in the first decades of the twentieth century indicated increasing doubts regarding the "paradigm" of classical science, that is, the explanation of complex phenomena in term of isolable of elements.

In the late 1920´s Von Bertalanffy proposed a concept "organismic biology" in order to provide a complete explanation of the vital phenomena which any single part and processes cannot provide.

"The propertie and model of action of higher levels are not explicable by the summation of the propeties and modes of action of their components taken in isolation. If, however, we ensemble of the components and relations exisiting between them, then the higher levels are derivable from the components" That is, in order to understand an organized whole we must know both the parts and the relations between them. Many discussions of reductionism were little adapted to deal with "relations" in system, they was concerned with one-way causality or relations between two variables. Here is the reason why, even though the problems of "system" were ancient and had been known for many centuries, they remained "philosophical" and did not become a "science"

Being a Biologist, Bertalanfy proposed the theory of "open systems", that is, sytems exchanging matter with environment as every living system does. Boulding, working independently and in different fieldes, from the direction of economics and the social sciences, had arrived at similar conclusions. In the meantime a different development of self-directing missiles, automation and computer technology, and inspired by Wiener´s work, the cybernetic movement became ever more influential and the basic model as feedback and dynamic system interactons were of wide applicability in many different disciplines.

General system theory consists of three parts: Systems science, systems technology and systems philosophy.

System science: Mathmatical systems theory

System science: that is, scientific exploration and theory of system in various sciences (e.g., physics, biology, psychology, social science), and general systems theory as the doctrine of principles applying to all systems.

A central notion of dynamical theory is that of stability, i.s, the response of a system to perturbation. (equilibrium states).

In external description, the system is considered as a "black box"; its relations to the environment and other systems are presented as graphically in block and flow diagram. The system description is given in terms of inputs and outputs; its general form are transfer functions relating input and output. Typically, these are assumed to be linear and are represented by discrete sets of values. In term of control theory, external description,typically, is given in term of communication (exchange of information between system and environment and within the system) and control of the system´s function with respect to environment (feedback), to the Winner´s defination of cybernetics.

System technology: The problem arise in modern technology and society , including both hardware ( control technology, automation, computerization, etc.) and software (application of system concept and theory in social, economical etc., problems).

System Philosophy: the reorientation of thought and world view following the introduction of system as a new scientific paradigm.

first, we must find out what is meant by "system" and how systems are realized at the various levels of the world of observation. This is systems ontology. What is to be defined and described as system is not a question with an obvious or trival answer. It will be readily agreed that a galaxy, a dog, a cell, and an atom are "systems". But in what scense and what respects can we speak of animal or a human, society, personality, language, mathematics, and so forth as "systems" ?

Real system which entities perceived in or inferred from observation and existing independently of an observer, conceptual systems which esscetially are symbolic constructs, and abstracted system as sunclass, that is conceptual systems corresponding with reality. However, the distinction is by no means as sharp as it would appear. For example, when we consider on the interactions of component, a social system is just as real as an animal, or human being.

"Systems" epistemology. As against reductionism and physicalism, the problems and modes of thought occuring in the biological, behavioral and social sciences require equal consideration, and simple "reduction" to the elementary particles and conventional laws of physics does not appear feasible. The investigation of organized wholes of many variables requires new categories of interaction, transaction, organization, teleology, and so forth.

We see science as one of the perspectives that man, with his biological, cultural, and linguistic endowment and bondage, has created to deal with the universe into which he is "thrown," or rather to which he is adapted owning to evolution and history. The third part of systems philosophy is concerned with the relations of man and his world.


Bai's Notes

Cybernetics View of Future

If human being lived more instinctively, they might be able to care for future generations in a more direct and probably simpler manner. But we have developed into a species which relies on speculation and self-interest as its guid.

we are just passing through an episode in history where people in ..... But future generations have as yet no adequate way of even hinting what they want. We systems scientists are not thwarted by such a restraint. we are brave heroes and heroines who love the challenge.

We huamns and our minds must plan future societies with a multitude of concepts that defy precise definition:

happiness,well being,beauty, peace,serenity and justice.

the cyberneticians furnish an important application of Singer's point of view by showing that teleolohical concepts are extremely fruitful in the study of neurological and machine behavior, and that such concepts can be treated experimentally.

there have been many so-called solutions to the mechanist-vitalist paradox, but most have erduced to a mere affirmation of the fruitfulness of analyzing experence from both mechanical and teleological viewpoints.

he has developed a schems of science in which he has demonstrated this capability. in fact, he has provided a transformation principle which enables the scientist to pass from mechanical to teleological explanation without intruducing any non-scientific forces.

There is an ever-prevalent danger that the definition will be oriented with respect to the particular aims of its formulators.

The end point of the process may come when each field returns to its own work and ignores the potential contribution of other disciplines. the real danger is complete loss of integration which at the present time seems essential in the study of purposive behavior.

Cybernetics analyzes all purposive behavior and provides an exact notion of communication and the transmital of information.

a definition of purpose is no more nor less than a method of looking at the world-a scheme of organizing phenomena.

The general idea Rosenblueth and wiener want to develop is that an object behaves purposefully if it continues to pursue the same goal by changing its behavior as conditions change.intensive function :

cybernetics extensive function : relatively invariant behavior in wide range of environments purpose : exhibiting different types of behavior even though the environment remains constant.

the interests of cybernetician are usually restricted to what we have called intensive function.

the history of science to data has shown that the use of quantification permits a more rapid reduction of experimental error, and consequently, where greater precision is desired, attempts are made to convert qualitative morphologies into quantitative morphologies. But for many purposes it is inefficient to convert the qualitative, even though it can be done. it is only where distinctions between qualities become open to uncontrolled error that quantification is needed.

the crucial point of comparison between this schem and the one presented by rosenblurth and wiener lies in the relationship of purposeful individual and his environment. Rosenblueth and Wiener require that the purposeful individual receive information from the environment, i.s., they equire that the purposeful individual respond to the enviroment as well as to changes in the enviroment produced by the individual itself as well as by other forces. We do not require such a feed-back relationship for purposive behavior, though we recognize it to be involved in much purposeful behavior; we omit this requirement because of psycholological examples of purposeful behavior where such transmitting of information either is not present or is not of concern to the experimenter.

Suitability of definition is always to be judged relative to aims , and our fears is that a strict forcing for definition usually in one area upon another discipline will eventually work against integration where integration is so urgently needed.


Wilby Jennifer (1994)

A critique of Hierarchy Theory.

Jouranl of Systems Practice, Vol. 7, No. 6, pp 653-6670

Critical Systems Thinking (CST) has five commitments: Complementarism and critical reflection on both the methodological and the theoretical levels and emancipation (Jackson,1991).

A critique of one model from within the standpoint of another methodology is an important implementation of CST.

The originating of theories has been integrated by Jackson (1985) and summarized by Schecter (1991) into three categories:

# Technical interest: To what extent, and how, does the theory (or can the theory) erve the technical interest of prediction and control. i.e. the hard systems approach ? # The practical interest: to what extent, and how, can the theory serve the practical interest of understanding, shared meaning, language, way of observing, worldviews, and roles in achieving consensus, i.e., the soft systems approach.

# The empowerment interest: to what extent, and how, can the theory serve the emancipatory interest, i.e., people's interest in being free from unjust power relation? this is added dimension of the critical systems approach.

Simon believes that most of the complex systems found in nature can be described in four interwined hierarchic sequences. His first hierarchy contains sets of component molecules, atoms, nuclei, electrons, and elementary particles. The second hierarchy contains living organisms, tissues, organs, cells, macro molecules, and organic compounds-and interwines, at its highest level, with the molecules of the first hierarchy. The third hierarchy is genes, chromosomes, and DNA, while the fouth hierarchy contains human societies, organizations, small group, individuals, thinking process, and "elementary information processes- where the junctions with the tissues and organs of neurobiology largely remain to be dicoverd (Simon, 1973).

Miller's (1978,1992) living systems theory i a biological description of entities that are "complexity structured open systems" (Miller, 1992). As in all cybernetic systems, these entitie have inputs, outputs, transformations, and feedback involving the processing of matter, energy, and information. Miller's living systems theory "demonstrates that living systems exist at eight levels of increasing complexity: cell, organs, organisms, groups, organizations, communities, societie, and upranational sytems" (Miller, 1992). Within each of these eight levels are the same 20 subsystems, and these units perform the cybernetic function of input, output, transformation, and feedback for each level in the hierarchy. Living system is a 1100-page description of these level and subsystems and is an effort to integrate all the social, biological, and physical sciences that apply to structure or process at any of the levels.

Other Hierarchy models e.g., Beer's or De Raadt MMM model , Activity theory model can be discused.

A critique to this diverse of hierarchy theory is that it does not have a unified or single set of definitions (its subjective bias, superficious feeling), the lack of a specific methodology for the application of the theory (not serve the technical interest), and the lack of a capable mathematical structure for all of hierarchy's middle-number systems.(p655). finally, it creat a views in the social sciences that the social structure to be top-down, authoritarian, if not dictatorial systems design (p665). There are two major philosophical quetions raised by hierarchy theory. The ontological question concerns whether hierarchies exist external to the observer or whether they are a reality held entirely within the consciousness of the observer. Most would hold the former viewpoint. The epistemological question is ??? (methodological quetion?).


Sterman, John D. (1994)

Learning in and about complex systems

Systems dynamics Review, Volum 10 No.2-3. PP291-329

Many advocate the development of systems thinking - the ability to see the world as a complex system, in which we understand that "you can't just do one thing," that "everything is connected to everything else". ...... The development of systems thinking is crucial for the survival of humanity. There are many schools of systems thinkings (see Richardson, 1991; Lane,1993). Some emphasize qualitative methods, others fromal modeling. As sources of method and metaphor they draw on fields as diverse as anthropology, biology, engineering, linguistics, psychology, physics, and TAoism, and seek applications in fields still more diverse. All agree, however, that a systems view of the world is still rare.

Learning about complex system when you also live in them is difficult.

We are all passengers on an aircraft we must not only fly but redesign in flight.

I argue that successful approaches to learning about complex dynamic systems require (1) tools to articulate and frame issues, elicit knowledge and beliefs, and creat maps of the feedback structure of an issues from that knowledge; (2) formal models and simulation methods to assess the dynamics of those maps, test new policies, and practice new skills; and (3) methods to sharpen cientific reasoning skills, improve group processes, and overcome defensive routines for individuals and teams....

All learning depends on feedback. George Richardson (1991) shows how beginning in the 1940s leading thinkers in economics, psychology, sociology, anthropology, and other fields recognized that the engineering concept of feedback applied not only to servomechanisms, but to human decision making and social settings as well. Forrester, in Industrial Dynamics (1961), asserted that all decisions (including learning) take place in the context of feedback loops. Later, Powers (1973,351) wrote:

Feedback is such an all-pervasive and fundamental aspect of behavior that it is as invisible as the air that we breath. Quite literally it is behavior - we know nothing of our own behavior but the feedback effects of our own outputs. To behave is to control perception. Learning as an iterative cycle of invention, observation, reflection, and action (Schön 1992). Learning as an explicit feedback process has appeared in practical management tools such as Total Quality Management.

The single feedback loop shown in figure 1 describes the most basic type of learning. The loop is a classical negative feedback whereby decision makers compare quantitative and qualitative information about the state of the real world to various goals, perceive discrepancies between desired and actual states, and take actions that (they belive) will cause the real world to move toward the desired state. Even the initial choices of the decision makers do not close the gaps between desired and actual states, the system might eventually reach the desired state as subsequent decisions are revised in light of the feedback received.

The feedback loop shown in fig 1. obscures an important aspect of the learning process. Information feedback about the real world is not the only input to our decisions. Decision are result of applying a decision rule or policy to information about the world as we perceive it. The policies are themselves conditioned by institutional structures, organizational structures, and cultural norms. These in turn are governed by the mental models of the real world we hold. As long as the mental models remain unchanged, the feedback loop is what Argyris (1985) calls single-loop learning, a process whereby we learn to reach our current goals in the context of our existing mental models. Single-loop learning does not result in deep change to our mental models-our understanding, our worldview, of the causal structure of the system. Our world is actively constructed - modeled- by our sensory and cognitive structures. This mental model long ago evolved structures to build these models automatically. Usually we are totally unaware these mental models even exist.

Argyris (1985) denoted double-loop learning shown in Fig.2. Here information feedback about the real world not only alters our decisions within the context of existing frames and decision rules but feeds back to alter our mental models. As our mental model change, we creat different decision rules and change the strategy and structure of our organizations. The development of ystems thinking is a double-loop learning process in which we replace a reductionist, partial, narrow, short- term view of the world with a holistic, broad, long-term, dynamic view and then redesign our policies and intitutions accordingly. Such learning involves new articulations and new decision rules, not just new decisions.

For learning to occur, each link in the two feedback loops must work effectively, and we must be able to cycle around the loops quickly relative to the rate at which changes in the real world render existing knowledge obsolete.

Much of the literature in psychology and other fields suggests learning proceeds via the simple negative feedback loops.

time delays between taking a decision and its effects on the state of the system are common and particularly problematic.

The self-reinforcing feedback (positive feedback) between expectations and perception has been repeatedly demonstrated in a wide varity of experimental tudies. Sometimes the positive feedback assists learning by sharpening our ability to perceive features of the envoronment, Cause and effect are often distant in time and space, actions have multiple effects, and the delayed and distant consequenses are often different from and less salient than proximate effects - or simply unknown.

The multiple feedbacks in complex systems cause many variables to be correlated with one another, confunding the task of judging cause. However, people are poor judges of correlation. People can generally detect linear, positive correlations given enough trials if the outcome feedback is accurate enough. However, they have great difficulty in the presence of random error, nonlinearity, and negative correlations, often never discovering the true relation (Brehmer 1980). Among the failure of scientific reasoning most inimical to learning is the tendency to seek evidence consistent with current beliefs rather than potential disconfirmation. Experiments show the tendency to seek confirmation is robust in the face of training in logic, mathematics, and statistics. search strategies that focus only on confirmation of current beliefs slow the generation and recognition of anomalies that might lead to learning, particularly double-loop learning (paper gives an example of Wason card task with E,K,4,7 and let the subject use smallest number of cards you should turn over to test the rule that 'If a card with vowel on one side have even numbers on the revere'. Wason found that most of the subjects selected E and 4 as the answer, very few selected the right answer E and 7).


Clemson Barry (1994)

A parable of two princes: An effective Approach to National Development.

Jouranl of Systems Practice, Vol. 7, No. 6, pp 619-631

Two opposite strategies: (1) initially optimum policies in the absence of a strong feedback system and (2) initially poor policies with a strong feedback system (real-time information system). The entire approach is based upon the viable sytem model of Stafford Beer.

The key issues in learning or adaptation generally and development in particular.

In any feedback control system, there is a delicate problem of timming. A control system that reacts too slowly allows catastrophe to occur (e.g., the moving car fails to stop and smashes into the one ahead of it) However, control action that are too fast( or merely timed wrongly) can put the system into stat of oscillation).

Von Beer's handbook for development:

Principle 1: In managing any socioeconomic system, time lags in knowledge defeat regulation.

The usual choice is between good measures late and poor measures earlier. Social scientists, by training and inclination, almost always opt for the good measure and the result is a bit time lag while the information i collected. Managers, however, have different needs than the scientists. If the manager knows immediately that his action makes things a little better or little worse, than that manager can continuously improve. The manager, unlike the scientist, doesn't need great accuracy; he needs only to get the sign right. Further, the scientist normally doesn't face severe time constrains, but timing is of essence for the manager. Thus the manager should be willing to sacrifice almost everything else in the data for the sake of speed of knowing.

principle 2: focus on nested sets of integral wholes

You can't understand the role of the human arm by removing the arm for study-you have to study it in the context of the human body.

If the units of analysis are not integral wholes in some important sense, then we will not be able to understand or regulate them.

Principle 3: Use simple recursive models.

Principle 4: Focus on dynamics

Principle 5: Use computer as a filter to find the important data for managers

Managers suffer from a flood of data, many of them irrelevant. The critical data are often overlooked simply because they are mixed in with masses of trivial or irrelevant material. The viable system model includes the communication channels to knit the entire system together into one coherent whole and to ensure that the subunits are working together appropriately and contributing to the objectives of the whole.


Sushil (1994)

Flexible Systems Methodology.

Jouranl of Systems Practice, Vol. 7, No. 6, pp 633-650

System concepts and methodologies have been developed as a response to the ever- increasing complexity of sociotechnical and managerial systems.

Different approaches have found favor in different situation and have their own strengthes and limittions. The task of designing or electing methodology for a particular problem situation is becoming more difficult with an ever-increasing choice of techniques. Efforts to expand system thinking: Bertalanffy's general system theory; Ackoff's idealized interactive planning; Checkland SSM as learning paradigm;.....

This paper proposes a unified spectrum of systems approache to capture the essence of spectral and relational thinking and provides a flexible approach to using the isomorphies and systems techniques and methods.

The value of combing various systems modeling approaches can be effectively seen in the area of DSS and AI which are based on cohesive and symbiotic framework of quantitative as well as qualitative approaches.

Paradoxes

Hard vs Soft Systems Thinking

Hard systems thinking is based on an "optimizing" paradigm, whereas soft system thinking is based on a learning paradigm. Both paradigms have been criticized by each other.

Quantitative vs Qualitative Analysis

Quantitative is said to be tangible and objective against subjective bias;

Qualitative is said to be full perspective of the problem situation against simplistical and partial set and milead user.

Individual vs System Structure

Flexible Systems Methodology's Paradigm is build on a 'spectral and integrative' paradigm . Fig 1. Systems Continuum : a spectral paradigm (Tedious description how to use the methodology)

<----- Get Part 2 of Digest or have a rest ------->


Return to Reading Notes