Past president of the Society for General Systems Research (now the International Society for the Systems Sciences), 1976.

See Heinz von Foerster by Stuart Umpleby


Bringing the mechanical and physical understandings of a physicist to cybernetics, von Foerster was the first to draw attention upon the limits of the machine metaphor, and also the conceptual aparatus of the mechanical universe (eg, the absurd connection between self-organisation and entropy). Has also acted as an entrepreneur for cybernetics both in gaining funding for conferences and in providing a home for other workers in the field. His greatest contribution was to see that the concepts of cybernetics were essentially reflexive, and to apply them to themselves as their own subject matter, thus generating “second order cybernetics”, the cybernetics which includes the observer in its observation.


Heinz von Foerster (1911 - 2002)

Stuart A. Umpleby, The George Washington University, Washington, DC 200052
November 2001

The Years Before lllinois

Heinz has been a central figure in the field of cybernetics since its beginning. During his student days in the 1930s he became involved with the Vienna Circle, a group of philosophers that included Wittgenstein, Schlick, Menger and Carnap. From them he developed an interest in the fundamental difference between the world as it is and its symbolic representation in language or equations. He wanted to learn more about the observer. However, World War II intervened, and he spent those years in various laboratories in Germany working on plasma physics and microwave electronics. Luckily he survived the war unscathed in mind or body. After the war he helped set up the first post-war radio station in Vienna and was in charge of its science and art program until 1949. In those days of rejuvenation, he returned to the old riddle of the nature of the observer. With the encouragement of the psychiatrists Victor Frankl and Otto Potzi, he published a short monograph on a quantum mechanical theory of physiological memory. During a visit to the United States he met Warren Mc Culloch who not only had the data for his theory of memory but who also introduced him to the campus at Urbana.

Through Mc Culloch, at conferences about “Cybernetics: Circular Causal and Feedback Mechanisms in Biological and Social Systems” sponsored by the Josiah Macy Jr. Foundation, he met the people who laid the conceptual foundation for understanding the really complicated systems - teleological systems and self-organizing systems. The people attending these conferences included Gregory Bateson, Julian Bigelow, Margaret Mead, John von Neumann, Norbert Wiener and Ross Ashby. Heinz was so fascinated by the ideas that emerged at these meetings that after seven years of research at the University of Illinois in microwave tubes and ultra-highspeed oscillography, he went on sabbatical leave to learn more about the neurophysiology of his enigmatic observer. After one year under the tutelage of Warren Mc Culloch at MIT and Arturo Rosenblueth in Mexico he returned to the University of Illinois and established the Biological Computer Laboratory to study computational principles in living organisms.

The Biological Computer Laboratory

Almost from the beginning it was apparent that the Biological Computer Laboratory (BCL) was not an ordinary university research group. One of the most amusing episodes in the history of BCL was the series of events that led up to Heinz's being mentioned in the cartoon strip Pogo, a distinction for scientists even rarer than the Nobel Prize (see Figure 1). Someone at the National Institutes of Health wanted a mathematical model of the population dynamics of white blood corpuscles. By working on this problem Heinz became interested in the dynamics of populations, both those whose elements interact and those with elements that do not interact. He figured that data on human population growth would be the most complete set of data for a population with elements capable of communication. The result was an article in Science in 1960 by Heinz Von Foerster, Patricia Mora and Lawrence Amiot called “Doomsday, Friday, 13 November, A.D. 2026.”[2] They found that the equation which best fit the data was not an exponential but rather a hyperbolic equation. There is a major difference. If population is an exponential function of time, population will become very large as time increases, but within a limited period of time the population will remain finite. A hyperbolic function, however, has asymptotes. That is, there will be a time at which population will go to infinity. Applying the method of least squares to parameterize the equation led to the date 2027, hence the title of the article.

projects.isss.org_uploads_main_heinz-1.jpgFigure 1: Reference to the Doomsday article in the Pogo comic strip

There followed one of the most entertaining exchanges of letters ever to appear in Science. The idea that the human population could, through communication, form a coalition and engage in a game against nature was a particularly troubling idea. One demographer called attention to the widely accepted view that industrialization reduces rather than increases population. Heinz and his colleagues pointed out that, if an inverse relationship between population and technological know-how is applied to the human population over the last couple of millennia, then “either Stone Age man was a technological wizard, who carefully removed his technological achievements so as not to upset his inferior progeny, or our population dwindled from a once astronomical size to the mere three billions of today.” [3] The BCL equation turned out to be considerably more accurate than other forecasts in predicting world population in 1970. The others were more conservative. However, 1975 data suggested that world population had moved ahead of even the BCL equation. [4]

In addition to research the Biological Computer Laboratory also had a significant impact on the students at the University. On even the largest college campuses there is usually a small group of students who are innovators in campus activities. They are the students who write for the campus newspaper and lead political or reform movements. These students usually know each other, and they often can be found in special projects courses with the most stimulating faculty members. At the University of Michigan the Mental Health Research Institute with James G. Miller, Anatol Rapoport, Kenneth Boulding, John R. Platt, and Richard L. Meier was a focus of innovative activity. Tom Hayden and Carl Oglesby were students of Kenneth Boulding. Boulding once said that Students for a Democratic Society was born in his living room as a result of a seminar in economics. The Biological Computer Laboratory served a similar function at the University of Illinois. Over the years Heinz's students produced a Whole University Catalogue, a book on Metagames, an Ecological Sourcebook, and a large volume on the Cybernetics of Cybernetics.

It is easy to understand why there was always a feeling of excitement around BCL once one understands Heinz's views on education. Heinz notes that most of contemporary education is designed to make students react to a question in exactly the same way. Tests are given to determine how successful the system has been at making the student a completely predictable member of society. The higher the score, the more predictable the student. In other words the purpose of education is to turn nontrivial systems into trivial systems. Heinz, following Herbert Brun, defines an illegitimate question to be one for which the answer is known. A legitimate question is one for which the answer is not known. Hence formal education is mostly concerned with illegitimate questions. At BCL the emphasis was on learning to ask legitimate questions. [5]

 Said Heinz there are two kinds of questions,
 To some there are answers in lessons,
 But the questions that count,
 The ones to surmount,
 Are the questions that not yet are questioned. [6]

To be “recruited” to BCL one only had to be attentive to the life of the campus. Because of his highly entertaining as well as thought?provoking manner of speaking, Heinz was a frequent lecturer on campus. One of the first times I heard him speak was about 1964 at the weekly luncheon series at the YMCA. In his talk Heinz predicted that in the years ahead people would make three discoveries. First, they would discover that the earth is finite. That is, population growth cannot continue indefinitely. Second, people would learn that power resides where information resides. Third, human beings would discover that A is better off if B is better off.

Events have tended to follow these predictions. In 1968, Paul Ehrlich published The Population Bomb, and gradually people became more aware of rapidly increasing population and the impossibility of sustaining the high growth rate for very long. [7] Regarding the second prediction, the 1970's brought greater attention to global communications - satellites, television, computer networks - and also revelations about the covert activities of the CIA and the FBI. In 1975 documents were made public which showed that during World War ll the Allies were able to listen to the message traffic of the German and Japanese high commands. [8] Alan Turing was a central figure in this work. It was no accident that World War II was such a successful war for the United States. This new perspective on World War II helps to explain the interest of the intelligence agencies in cybernetics research. The U.S. government had learned that power resides where information resides. The achievement of Heinz's third prediction - people will realize that A is better off when B Is better off - seems to lie within our grasp. The Cold War is over. Europe is becoming united, and nations are working together as never before.

In about 1974, I mentioned to Heinz the three predictions he had made a decade earlier. He had forgotten about them, and he attached little significance to them. He said he had put them together a short while before giving the talk because he thought they would amuse the audience. But for me, those three predictions remain an example of Heinz's depth of insight, broad human concern, and faith in the eventual good sense of his fellow human beings.

Lest this description of the activities at BCL leave the reader with the impression that the laboratory led an untroubled existence, I should say a few words about the difficulties that Heinz faced. BCL was a leader. It was chronically ahead of its time. There was an exuberance at BCL that some interpreted as lack of seriousness. Quite a few people thought that anyone with an interest in physics, linguistics, art, music, dance, and anthropology must be a dilettante. And more than a few people suspected that calling attention to perception was somehow subversive. While the lack of understanding was unfortunate, it did not greatly matter as long as Heinz maintained his reputation as a successful grantsman. But then came the Mansfield Amendment. Most of the early work on cybernetics had been supported by the Office of Naval Research and the Air Force Office of Scientific Research. But in about 1968 the Mansfield Amendment put an end to research projects supported by the Department of Defense that were not clearly related to a military mission. It was intended that the National Science Foundation and other agencies would pick up the support of projects that had been funded by DOD. The problem of course was that these agencies did not have people who were familiar with the work in cybernetics. There followed several frustrating years of searching for new sources of support. Meanwhile Ross Ashby and Gotthard Gunther had retired and left the University. Finally in 1975 Heinz retired and moved to California. The University decided not to hire new faculty members to continue the work of BCL. For those familiar with the laboratory, it was a heartbreaking end to a remarkable episode in the history of science.

How to do Cybernetics

BCL left behind a rich legacy. In its day, it was one of very few educational institutions training people in cybernetics. Between 1958 and 1975 operating under 25 grants, the laboratory produced 256 articles and books, 14 masters theses and 28 doctoral dissertations. The topics covered epistemology, logic, neurophysiology, theory of computation, electronic music and automated instruction. These materials are available in a microfiche file compiled by Kenneth Wilson. But merely to note the range of ideas developed at BCL does not capture the feelings regarding the laboratory. Among the graduates of BCL that I know there is a strong feeling that the work produced at BCL has not received the attention that it deserves. When we go to conferences on cybernetics and systems theory, we find that many people are still struggling with issues that were resolved at BCL, often quite elegantly, long ago. We believe that the field could progress more rapidly if the work done at BCL were more widely known. Of course other groups were also doing important work during this period. Great progress was being made in computer science, artificial intelligence, operations research, simulation languages, the brain sciences and other related fields. But at BCL there was more emphasis on epistemology than at other cybernetics research institutes. The different epistemology has led to a sort of gap between the BCL point of view and other systems theorists. People with a BCL background experience this gap as frustration when we try to discuss a wide range of theoretical issues. Our views are often rejected for reasons that seem to us uninformed and unpersuasive. It is apparent that there is something rather complex going on.

In order to understand the importance of Heinz's ideas and why they have so far experienced such limited acclaim, it is necessary to understand how science operates. Most scientific work follows well-trod paths. The questions dealt with are widely shared. The methods used are commonly practiced. Consequently the results of most scientific work are readily understood and there is widespread agreement about their significance within the community concerned.

But scientific work that leads to the establishment of a new area of inquiry is different in kind. To understand this difference I find it useful to refer to Thomas Kuhn's list of components of a “disciplinary matrix”. In the epilogue to The Structure of Scientific Revolutions Kuhn identifies at least four elements of a scientific field. [9]

(1) There are “symbolic generalizations” such as F = ma and E = IR. (2) Models and analogies include the idea that electric current is similar to water flowing in a pipe and the idea that the molecules of a gas behave like tiny elastic billiard balls in random motion. (3) “Values” could include the following propositions: Quantitative predictions are preferable to qualitative ones. Predictions should be accurate. Theories should (or need not)be socially useful. (4) “Exemplars” are the concrete problem solutions that students encounter from the start of their scientific education, whether in laboratories, on examinations, or at the ends of chapters in scientific texts. Exemplars show scientists how their job is to be done.

Work at The George Washington University on a “disciplinary matrix” for the field of cybernetics has led to two additional components.

(5) “Guiding questions” state the principal concerns that motivate the development of theory. For example, early in his career Warren Mc Culloch asked the question, “What is a number that man may know it, and a man that he may know a number?” (6) “Techniques” are the methods an author uses to persuade the reader to his point of view. Techniques can be mathematical or verbal. Examples are set theoretical proofs, regression analysis, computer simulation, laboratory experiments with animals, survey research, thought experiments, and historical examples.

Major changes in science seem to occur through the formulation of new “guiding questions”. Heinz, for instance, has been preoccupied with the nature of the observer. Progress toward the resolution of a guiding question usually takes the form of new “exemplars”. The role of exemplars in the development of a scientific field is crucial. Kuhn equates his concept of “paradigm” with exemplars. Exemplars can be used to identify the groups within a scientific field who practice the discipline differently.

Those with a BCL background use different exemplars and hence are using a different paradigm. Heinz's articles are filled with fascinating exemplars. Three of these accompany this article as illustrations.

A central concept within the field of systems science is the notion that a whole can be somehow greater than the sum of its parts (see Exemplar 1, at the end of this paper). Yet this very important idea is often not precisely understood. Heinz has resolved the mystery with two simple illustrations, one from mathematics and one from biology.

The story of the magnetic cubes in a box demonstrates what Heinz calls “order from noise” (see Exemplar 2). I believe that it is also the most readily understood illustration of a self-organizing system. In order for more complex systems to evolve, two things must happen, New variety must be generated, and appropriate selection must take place. A self-organizing system is not a living system. Rather, a self-organizing system contains both organisms and their environments. As the system moves toward its stable equilibrial states, it “selects” the stable relationships. The example of the magnetic cubes in a box illustrates how this very general process can also generate variety. As the separate boxes find stable relationships, they form chains. The chains of boxes can themselves be the elements in the next step in the process of self?organization. Thus Heinz's example of the magnetic cubes in a box illustrates one way that new entities may emerge.

The idea of a self-organizing system as a closed system (the interaction rules do not change during the period of observation) is one of the most important and powerful ideas in systems theory. The principle can be applied in many ways. For instance, anticipating the stable states of a system is an alternative to trend extrapolation as a forecasting method.

 Regression fits systems just trivial,
 But humans are systems convivial,
 To say what will last<
 From times that are past,
 Will lead to conclusions peripheral.

The gentleman in the bowler hat can be regarded as proof of the need for multi-valued logics (see Exemplar 3, at the end of this paper). A principal concern of systems theory is the relationship between an organism and its environment. However, it is not sufficient to speak only of an organism and its environment. The logical structure of the concept “environment” is more complex than can be described by a dyadic relationship. In order to establish the concept “environment,” there must be at least two elements observing this environment. And they must be sufficiently alike in order to serve as mutual witnesses for any objective event. Only knowledge that can be shared belongs to the environment. Observers without shared knowledge inhabit different universes. The logical structure of environment is a triadic relation because it involves three entities: an observer, A; a witness, A'; and that which is witnessed, B. What is called the environment can be called “together?knowledge” for which the Latin expression is conscientia. Heinz has suggested that it was probably the triadic logical structure of this concept that gave philosophers over the last three thousand years difficulty when they tried to reason to a simple true-false, two-valued, Aristotelian logic, where at least a three?valued logic is required. [10] Of course It can be claimed that consciousness Is a single person's affair and that you do not need a witness in order to be conscious. However, note that our consciousness is produced by the “together-knowledge” of our different senses. The ear is witness to what the eye sees. Touch confirms what the eye reports.

For those for whom mental activity is an important part of life, a person who can invent images with such great organizing power becomes the object of tremendous affection. Cybernetics, like any science, is a different way of seeing. “Doing research” on a different way of seeing means creating examples of how the world looks from the new point of view. If the field of cybernetics has not progressed as rapidly as we would have liked; the reason may lie in the fact that more of us have not realized that a new field requires new exemplars.

Second Order Cybernetics and the Change of Science

Heinz believes that in order to deal with current concerns, science must change to include attention to the observer. He calls the new point of view “second order cybernetics”. For several definitions of the difference between first order cybernetics and second order cybernetics, see Table 1. [11] Let me explain this new point of view in my own words. Science can be thought of as having moved through three stages in how it deals with objectivity. In the early days of science researchers were concerned with inanimate objects such as balls, pendula and planets. We could call this an era of “unquestioned objectivity”. When science progressed to the study of human behavior, it was found that control groups were necessary to eliminate interaction effects between the observer and the subjects. That is, the experimenter would now have two groups of subjects rather than just one. For example, one group would be given the medication to be tested and the other group would receive a placebo. The difference between the responses of the two groups was assumed to be due to the drug. Effects due to the attention received by the subjects from the doctor could thereby be eliminated. This second stage in the treatment of objectivity could be called “constructed objectivity”.

Things became quite unwieldy, however, when social scientists began to do large-scale social experiments. During the mid-1960's a great wave of social legislation was passed in the United States as part of Lyndon Johnson's Great Society program. After a few years it became apparent that many of the programs were not working exactly as anticipated. There then developed a peculiar commonality of interest between conservatives and social scientists. The conservatives wanted to stop the social programs. The social scientists argued that before implementing a new social program on a nationwide basis, experiments should be run in selected communities to determine the effects of the program. These experiments were frequently quite controversial. The purpose of the experiment was usually defined differently by different people. The experimenters were considered suspect because they did not live in the community. And people demonstrated a remarkable ability to reject findings they did not agree with no matter how “scientifically” the experiment was conducted.

Table 1

Author First Order Cybernetics Second Order Cybernetics
Von Foerster The science of observed systems The science of observing systems
Pask The purpose of the model The purpose of the modeler
Varela Controlled systems Autonomous systems
Umpleby Interaction among the variables in a system Interactions between observer and observed
Umpleby Theories of social systems A theory of the interaction between ideas and society

These experiences led Mitroff and Blankenship to develop seven guidelines for conducting a holistic experiment.[12] The guidelines are a radical departure from current methods in the social sciences. The usual social science experiment today, consistent with the second stage in the treatment of objectivity, consists of testing a single hypothesis using a set of subjects and a control group. However, the guidelines for a holistic experiment say that at least two points of view should be used. Furthermore, the experimenters should be included within the class of subjects, and the subjects should be included within the class of experimenters. These guidelines amount to a new scientific method that is compatible with the new epistemology and its emphasis on the role of the observer. This third stage in the treatment of objectivity can be called the period of “contested objectivity”.

 Some systems display little interaction,
 But politics makes payments to each faction.
 When establishing agreement,
 Becomes a great achievement,
 Will "objectivity" give sufficient satisfaction?

The new epistemology or second order cybernetics encounters at least two kinds of resistance. Some people simply do not understand the question. Hence, they are unable to appreciate the answer. The concern with epistemology is not their concern. Their attention is elsewhere. So the discussion is irrelevant. This was my situation during my early association with BCL. I began working with Heinz in 1971 in order to better understand the work of Ross Ashby, which did seem relevant to my concerns at the time. Heinz kept talking about the observer, a point that seemed rather obvious to me and not worthy of lengthy discussion. I believe it took about eighteen months for me to begin to get an inkling of what Heinz was talking about. I find now that second order cybernetics provides me with a scientific basis for understanding things that I previously would have thought were outside the reach of science. For example the rise of humanistic psychology and the human potential movement and the concern with values and ethics can be explained in terms of the observer's being aware of his relationship with his environment and the desire that people have to develop communities of common understanding.

The second type of opposition that the new point of view encounters is more difficult to cope with. This opposition is less a matter of ignorance than trained disbelief. It is not malicious, but It can be virulent. Its self?confidence springs from the belief that the concern with the observer is a new form of introspectionism, the point of view that psychologists abandoned when they adopted behaviorism in the 1960's in their effort to be more scientific. It may well be that science tends to oscillate back and forth between positions ? a concern with the observer, then with the outside world, then with the observer, etc. But the pendulum never returns to exactly the same position. The old position is redefined as a result of intervening experience. Nevertheless it is extremely difficult to persuade a person whose early career involved the abandonment of one position and the adoption of a new one to embrace a point of view that appears to him to be quite similar to a position he long ago rejected. For dealing with this second kind of opposition to the new epistemology I believe our best hope lies in the correspondence principle.

Heinz introduced me to the correspondence principle - any new theory should reduce to the old theory for those cases in which the old theory is known to hold. The principle assumes that science grows somewhat like concentric circles. Of course science can also grow by the development of theories to explain completely new phenomena unrelated to previously explained phenomena. But if the correspondence principle can be applied, it brings with it several advantages. First, if the new theory is consistent with the old theory for those cases already investigated, a large body of support for the new theory is readily at hand. Second, and most important for our purposes, if the correspondence principle can be shown to hold, then those scientists who have devoted their professional lives to the development of the old theory have not labored in vain. That is, the new theory does not threaten to invalidate their work, merely to extend it. Thus the correspondence principle reduces the threat of the new theory to those who have developed the old theory. Of course the correspondence principle does not completely reduce the emotional threat. Nor does it seem to make the paradigm shift noticeably easier. Nevertheless I have found it to be a useful debating point that seems to ease tensions. The principle tends to turn an either/or situation into a both/and situation.

projects.isss.org_uploads_main_heinz-2.jpgFigure 2: Second order cybernetics fits the correspondence principle.

The difference between the old cybernetics and the new cybernetics can be described as follows. Start with the assumption that first order cybernetics dealt with interaction among the variables in a system. Second order cybernetics, on the other hand, focuses attention on the observer as well as the system being observed. Draw a line going from zero to some large number (Figure 2). The points on this line indicate varying amounts of attention being paid to the observer. Classical science or logical positivism (not just first order cybernetics) dealt only with cases very near zero. Contemporary science deals with cases all along the line. Clearly the old conception of science is a special case, a subset of the new conception of science.</P>

Few scientists have made contributions as significant as Heinz Von Foerster. He has contributed to electrical engineering and neurophysiology, he revolutionized demography (13), and he has been a central figure in establishing the new field of cybernetics. There are still some systems theorists who claim that systems theory is not and should not be a scientific field. But Heinz had the daring to follow the idea of communication and control to its logical conclusion. Since science is a method of communication and control, cybernetics is in part a science of science. By concentrating on the observer it became apparent that a new kind of science is required. Heinz and his colleagues not only established cybernetics as a scientific field in its own right, they have also developed theories which will eventually lead to the change of science itself.

Heinz Is currently enjoying a very active retirement in a house which he, his wife Mai, and his son Andy designed, and with one helper built on Rattlesnake Hill close to Pescadero in California.

See also A Selection of Writings

Heinz von Foerster Exemplar 1Heinz von Foerster Exemplar 2Heinz von Foerster Exemplar 3

References

  1. This account is taken from Heinz Von Foerster, “Physics and Anthropology: A Personal Account by the New President of the Wenner-Gren Foundation”. Current Anthropology, Vol. 5, No. 4, 1964, p. 330.
  2. Heinz Von Foerster, Patricia M. Mora, and Lawrence W. Amiot,“Doomsday; Friday, 13 November, A.D, 2026,” Science, 132, 1960. pp. 1291?1295.
  3. Von Foerster, Mora and Amiot, “Population Density and Growth,” Science, 133, 16 June 1961, pp. 1932?1937.
  4. James Serrin, “Is Doomsday on Target?” Science, 189, 11 July 1975, p. 4197.
  5. Based on an article by Don Blyly, “Von Foerster: Austrian Professor Brings Creativity, Innovation,” Technograph, the student engineering magazine at the University of Illinois, February 1973, p. 11.
  6. Stuart Umpleby, “Limericks about cybernetics,” Cybernetics and Systems, Vol. 23, No. 2, 1992, pp. 229-239.
  7. Paul R. Ehrlich,The Population Bomb, New York:Ballantine, 1958.
  8. F. W. Winterbotham, The Ultra Secret, A Dell Book, 1975.
  9. Thomas S. Kuhn, The Structure of Scientific Revolutions, University of Chicago Press, Second Edition, 1970, pp. 182?187.
  10. Heinz Von Foerster, “The Logical Structure of Environment and its Internal Representation”, R.E. Eckerstrom (ed.), International Design Conference Aspen 1962, Zeeland, Michigan:Herman Miller, Inc., 1963.
  11. Stuart A. Umpleby, “Second Order Cybernetics and the Design of Large-Scale Social, Experiments,” Proceedings of the Annual Meeting of the Society for General Systems Research, Boston, February 1975.
  12. Ian Mitroff and Vaughan Blankenship, “On the Methodology of the Holistic Experiment: An Approach to the Conceptualization of Large-Scale Social Experiments”, Technological Forecasting and Social Change, 414, 1973, pp. 339?353.
  13. Stuart A. Umpleby, “The Scientific Revolution in Demography,” Population and Environment, Vol. 11, No. 3, Spring 1990, pp. 159-174.
 
heinz_von_foerster_by_stuart_umpleby.txt · Last modified: 2008/01/12 19:56 by daviding