Unlearning the Obsolescent


A short but punchy essay by Brodey which insists that dealing with irrelevance and obsolescence should be an important aspect of designing physical and educational environments.

One would not expect to find a course called “Unlearning the obsolescent” in a university catalogue or a company brochure. Yet the requirement that each student or employee know how to free himself from yesterday’s truths is critical. Obsolete knowledge, as I am using the term, remains true knowledge; it has merely become irrelevant.

The above paragraph has the certainty and brevity that would make it ideal for a catalogue or a brochure. But the bright student would immediately say, ‘There is a logical fallacy here: if I recognize that an item of truth is now obsolescent, surely I have already unlearned it. Such a course must be for pointing out the latest obsolescent facts’. To counter this valid objection, we had better add a last sentence to our course notice. ‘This course acknowledges that the metabolism of noise into information; and information into knowledge and knowledge into noise is necessary to survival.’ This last sentence is a salute to Parkinson’s Law. It would intrigue those who might easily benefit from the course, and also engage those who could challenge in a meaningful way. The others would simply say ‘nonsense’.

In the first meeting of the ‘unlearning course’ we would have to explain that sentence we just added. Most of our students are going to read it as they would read about the conservation of energy, sure that our new course is indeed needed. They would not at first comprehend that the course is aimed at placing the classical ways of modelling phenomena within the perspective of other possibilities. As we change models what was noise becomes information and what knowledge becomes noise. ‘The model provides a familiar way of structuring phenomena into truths—familiar but antique. It is now recontacted by elegance in measuring movement even of such large high inertia ecologies as our solar system. Space travel makes old knowledge irrelevant. We were taught to use conservation by the teachers whom we had as kids, it catapulted us into our new learning.

The model of the universe that postulates final cause is built into our descriptive language. It cannot be escaped. Know which model is used and its way of fitting our purpose. Rule I: Have fun. Remember that each purpose and its practical way of structuring (modelling) a phenomenon will organize its truths. When you are struck with an irrelevant truth, change the modelling context and re-examine the territory that the old truth labelled.

To illustrate: The whole concept of information, the language and truths popular today among information engineers, was designed to help decide how to quantify information flow along a channel. The range of possible messages and codes and the nature of the channel were all predetermined. The game was played with the deck of the cards and the possible moves known. Which move would be made remained unknown. Probabilities could be specified and novelty determined. But the meaning or effect of the message on the receiver and the effect of his return responses were not ready to be studied then. Feedback, complexity, and context were omitted. Limits were set that would generate a truth that might be used to establish telegraph charges and such. The work of Weaver and Shannon is important. Weaver’s essays suggested that the theory would grow into a wider framework important to understanding meaning and effectiveness. Their start is relevant to the comprehension of dialogue, but not when it is used to dismiss the broader modeling problems the authors were then unable to formalize.

Some of the students will ask: "What has this to do with unlearning the obsolescent?" Anyway, once you understand the limitations of the formal definition of information, you have a simple description that is so well defined that it will allow you to build a richer definition by simple expansion. Expanding old forms is one way to go. But they have forgotten to think their way into our rule: Each way of modeling a phenomenon will organize its truths. Each epistemological structure, each pattern-recognition system is, in the most basic sense, a code that acts as a carrier that makes certain kinds of relationships translucent. Try to find the code that will allow labeling then measuring the territory most relevant to the control purpose for which the code is designed. If our purpose is to describe what we have observed objectively to others who share our language, then we will get a code that is easily transmitted in descriptive form. We do our experiment so that it will fit on simple graphs. But if we seek to find a code that allows us to control the system, we must recognize that this may not create truths that are easily transmitted in the classical descriptive forms—or with the classical simple static graph as a means of display.

Unlearning obsolescence, then, in its simplest form, means helping high school students or older experts overcome the simple models given them in second and third grade or before the experts had begun to conceptualize models, systems, information, and control. When I was a kid, it was the belief of the teacher, and perhaps true, that saying condensation of water vapor caused rain was about the best you could do to show him with a kettle and a cool glass how rain formed. Children then had been taught to ask ‘What causes?’ questions, and the teacher answered with ‘cause and effect’, only later introducing other ways of labeling relationships.

Grandpa, or brother, each presents the child with the kind of relational model he learned in his childhood. This creates what is simple—unless he unlearns it. And when the child uses this model, its use is reinforced by being understood. The lag in learning modeling skills prevents our redefining what is simple as our world changes.

The recursive system of modeling, with no easy way out of the loop, puts a time lag on our skill at manipulating epistemologies. We are still stuck in developing our formalization of how to vary our basic pattern recognition systems.

It furthers each way of structuring (or modeling) phenomena will organize their truths. The epistemology that a young child is taught determines to some degree what he can label as simple. Children who have been taught the science of the cause-effect model will most easily absorb information about what causes phenomena—because that’s simplest. They are unaware that their truths are true within a particular code, for that code is for them omnipresent and automatic. Yet, daily life cannot be economically represented this way.

By now, the students who had been captured by the catalogue description of my course would be uniformly annoyed because I am talking of childhood; they would think that this simply is a throwback to my period of training and practice as a psychiatrist. Rule II: Make fun of, exaggerate, what is difficult to grasp because there is no code. It takes courage to be absurd. Being absurd is a part of the move out of one frame of reference towards another. But to communicate and take the students with us into the new frame of reference, we must inspect the limits of their tolerance of the ambiguity. The no-man’s-land between models is noisy and rich in information which seems to create itself as we unlearn what was before omnipresent and unavailable to doubt. May I repeat, what is absurd or true depends on one’s model; shifts in model make conceptions absurd. When you can’t have a language for a shift in model and are groping at best, use the old words in a ridiculous way until you find relevant data. Expose your shift in model until someone senses in a crude way the territory you are recognizing, and gives you a location where there is data support. There are those who are expert at shifting the structure of their pattern recognition system. Find someone who shares code enough to help you refine it. This dialogue with laboratory or persons or both provides reinforcement and gradually and usually with pain lifts into awareness the obsolescent modeling so taken for granted as to be automatic. It is easier to unlearn the obsolescence that happens when a widely held system of modeling phenomena is found inadequate for generalizing powerful and beautiful, in the sense of highly relevant, truths.

I am interested in man’s control of his environment, and the need for skill in overcoming obsolescence so that we can adapt to a rapidly changing ecology, which we ourselves are producing. Evolution involves unlearning obsolescent patterns that are no longer adaptive for humans. Evolution involves intelligence patterns. The invention of new tools—for example, the computer, which gives the environment intelligence in man’s image—changes our intelligence requirement, just as the promise of cheap power changes our power requirements. Without adequate capacity to conceptualize the complexity of our new ecology and to control the rate of change of the system so that it remains within our adaptive range, we are in trouble, for man and his environment are as interdependent as smog and breathing. The model of ‘cause-effect’ or ‘who done what to whom’ is obsolete. The formalization of phenomena as if they were contextless is obsolete. Where to hit the smog cycle is a control problem, not a cause problem. 

One cannot stop industry from producing smoke without causing other problems. The description of phenomena as if they were timeless is obsolete. The timing of industrial, financial, and market shifts so they phase in to allow smog to be controlled requires scientific knowledge of timing that will not be available if we seek timeless truths. The conceptualization of phenomena into a mathematics that is based on truth tables—bimodal true-or-false relations and their probabilities—may mop up the old questions seen in the old structural format, but it has limited value in formulating the new questions which are concerned with the control of complex systems. If we intuitively handle the flowing together of features which are interdependent, why is it formally considered impossible and perhaps foolish to tackle? Starting with something simpler, one at a time, and building up to complexity doesn’t work.

But for complex interdependent nets, or what McCulloch calls anastomotic nets, we have no formal, no logic of relations. Our computers could help us, but first we must have some way of structuring what we humans do and how we handle complex phenomena. Our pattern recognition systems are complex, but we must begin with awareness that perhaps we can be simple if we can find the right way of modeling our complexity.


Bargellini, P. L. “Considerations on man versus machines for space probing,” Advances in Computers, Vol. 6, p. 195, 1965.

Brodey, W. M. “A cybernetic approach to the family,” Family Therapy and Disturbed Families, Science and Behavior Books, Palo Alto, (In Press).

— “Developmental learning and the education of the child born blind,” Etc., Vol. XXIII, pp. 293-306, 9/65.

— “Sound and space,” Journal of the American Institute of Architects, July 1964.

— “A systems approach to the voice insertion of data,” Western Electric Show and Convention, September 1964.

— “The time manifesto,” Conference on Interdisciplinary Perspectives in Time, Annals of the New York Academy of Sciences, (In Press).

— and W. S. McCulloch, “Reviewing the achievements in the biological sciences with special emphasis on understanding the function of the nervous system,” Great Ideas of Today, Encyclopaedia Britannica, (In Press).

Darwin, C. The Expression of the Emotions in Man and Animals, University of Chicago Press, Chicago, 1965.

Fantz, R.L. “Visual perception from birth as shown by pattern selectivity,” Annals of the N.Y.A.S., Vol. 118, pp. 783-866, May 25th, 1965.

Fogel, L. J., A. J. Owens, and M. J. Walsh, “Artificial Intelligence through simulated evolution,” John Wiley, New York, 1965.

Lewis, B.N. and G. Pask, Teaching machines and programmed learning, II: Data and direction, pp. 213-266, National Education Association, Washington.

McCulloch, W. S. Embodiments of Mind, MIT Press, 1966.

MacKay, D. M. “Brain and will,” The Listener, May 9th and 16th, 1957.

Pask, G. “My predictions for 1984," Prospect, Hutchins, 1962.

— “The thresholds of learning and control,” Data and Control, February 1964.

Rosenblueth, A. , N. Wiener, and J. Bigelow, “Behavior, purpose and teleology,” Phil. Sci., Vol. 10, pp. 18-24, 1943.

Shannon, C. E. and W. Weaver, The Mathematical Theory of Communication, University of Illinois Press, Urbana, 1964.

Smith, K. U. Delayed Sensory Feedback and Behavior, W. B. Saunders Co., Philadelphia, 1962.

von Uexküll, J. “A stroll through the worlds of animals and men.” Instinctive Behavior, tr. and ed. C. H. Schiller. International Universities Press. New York. 1957.