CYBERNETICS FORUM THE PUBLICATION OF THE AMERICAN SOCIETY FOR CYBERNETICS

WINTER 1979 VOLUME IX No. 4

A SPECIAL ISSUE DEVOTED TO POSSIBLE CRISIS IN CYBERNETICS AND GENERAL SYSTEMS THEORY

IN THIS ISSUE:

Editorial

ls There A Crisis of ldentity in Cybernetics?, V. G. Drozin ...... 1

Articles

Notes on the Crisis of Cybernetics, Robert Lilienfeld ...... 3

"Don't Bite My Finger, Look Where lt Points": Some Comments on Lilienfeld's ldeo/ogical Analysis, Ari Ariely ...... 7

Review of Berlinski's "On System Analysis," Joseph P. Martino ...... 15

Some Thoughts on Systems Theory and lts Critics: A Letter To The Editor, Ervin Laszlo ...... 21

Linear and Elitist: How Our Critics See Us, Stuart Umpleby ...... 24

Cybernetic~: Search for a Paradigm, N. A. Coulter, Jr...... 27

About the Authors ...... 32

© 1979 American Society for Cybernetics BOARD OF EDITORS

Editor Charles H. Dym Frederick Kile V.G. DROZIN Oym, Frank & Company Aid Association for Lutherans Department of Physics 2511 Massachusetts Avenue, N. W. Appleton, W/54911 Buckne/1 University Washington, DC 20008 Lewisburg, PA 17837 Mark N. Ozer TECHNICAL EDITOR The George Washington University Kenneth W. Gaul Gertrude Herrmann Conference Calendar Editor School of Medicine and 111 0/in Science Building Health Seiences Buckne/1 University 1131 University Boulevard West, N2122 3000 Connecticut Avenue N. W. Lewisburg, PA 17837 Washington, DC 20008 Si/ver Spring, MD 20902 ASSOCIATE EDITORS Charles I. Bartfeld Doreen Ray Steg Harold K. Hughes School of Business Administration, Oepartment of Human Behavior & The Stare University College American University Oevelopment, Potsdam, NY 13767 Mass. & Nebraska Aves. N. W. Orexel University Washington, DC 20016 Philadelphia, PA 19104

N.A. Coulter, Jr. Oepartment of Surgery Akira lshikawa Curriculum in Biomedical Graduate Schoo/ of Business Paul Studer Engineering Administration School of Library and Information University of North Carolina Rufgers University Science, Stare University College School of Medicine 92 New Street of Artsand Science Chapel Hili, NC 27514 Newark, NJ 07102 Geneseo, NY 14454

OFF/CERS OF THE AMERICAN SOC/ETY FOR CYBERNETICS

PRESIDENT VICE PRESIDENT TREASURER Stuart Umpleby Doreen Steg Al Kreger The George Washington University SECRETARY OMBUDSMEN Washington, DC 20052 Roger Conant Klaus Krippendorff Rolf Wigand

TRUSTEES Stafford Beer Carl Hammer Daniel Howland Douglas Knight Heinz Von Foerster

PRINTED BY: DISTRIBUTED BY: Colonial Printing American Society West Market and 20th Street for Cybernetics Lewisburg, PA 17837 (717) 523-0702 Arnerican SUBSCRIPTION RATES: Society ASC Cybernetics Forum Journal of Cybernetics and Information Science 8oth Publications Domestic $35.00 per year Domestic $55.00 per year Domestic $80.00 per year for Foreign $40.00 per year Forelgn $60.00 per year Foreign $90.00 per year Cybzrne~ics Copyright © t 979 American Society for Cybernetics EDITORIAL ls There A Crisis of ldentity in Cybernetics?

At a recent Planning Meeting of the American Obviously, in order to establish a goal state and Society for Cybernetics, the question was twice to have repertory of decisions and their executions, raised: "What is Cybernetics?" Unfortunately, no both the controlling and the executive subsystem agreement was reached. Similar Situation exists in should first be aware of the properties of the con­ General Systems Theory. A. Rapoport, after 22 years trolled system. Thus, application of the scientific of editing General System, came to the conclusion method preceeds that of the cybernetic method. that: "General Systems is not a weil defined However, a person, after studying chemistry, cannot 'field' . . . " lf, only a few years after the creation of design a chemical factory. She or he also needs to Cybernetics, our inability to identify the field could learn chemical engineering. ln other words, pure be attributed to the "illness of growth," now, over science does not supply information in the form 30 years later, it may indicate more of an "organic needed for effective application of the cybernetic illness." Such Iack of clear definition is also a hand­ method. This is one reason why social sciences do ieap to further development. Cyberneticists applying not have a significant impact on the political pro­ for research funds who offer different interpretations cess, as shown by R. A. Scott and A. R. Shore in of their field can hardly expect to establish credi­ Why Sociology Does Not Apply: A Study of the Use bility for themselves or their projects. Cybernetics of Sociology in Public Policy. would acquire its own identity if we could show that The scientific and the cybernetic method Iead to it represents a unique field of knowledge with rather different kinds of models. After inductive application elear boundaries which separate it from other related of the scientific method results in accumulation of fields. data, the scientist suggests a hypothesis explaining All fields of knowledge can be distinguished by newly discovered and previously known relationships the method of inquiry used and by the subject between the studied parameters and predicting matter studied. The scientific method in pure new relationships in the form in which they can seiences is used for understanding the properties of be verified. A hypothesis is a scientific model of a particular systems, while in applied or engineering system. Narrowly specialized sciences dealing with a seiences this method is used for providing informa­ specific, relatively small parts of the natural (geo­ tion about the design and construction of systems sphere and biosphere) or artificial (societal, of desired specifications. Although this method is technological, and other man-made systems) environ­ common to all sciences, the way it is applied de­ ment may use a great deal of mathematics in their pends on the type of the system under study: a models. Every complex system is, however, an "in­ ehernist analyzing a compound by acting upon it, sufficiently described system." Therefore a cyber­ and a political scientist using public op inion poll to neticist, trying to build a model of a brain, learner, pred ict the outcome of an election are differing in or world system, cannot succeed without indicating applications of the same scientific method. "degree of insufficiency," ignored in his or her The most characteristic feature of Cybernetics and model and the way to control it. Frequently, cyber­ the source of its strength lies in the application of neticists do not realize the limitations inherent in a its method-the only method of purposeful activity mathematical model of even less complex systems. which is used in any transformation of a controlled The complex reality cannot fit into the straight­ system from its initial into its goal state. The deci­ jacket of a few differential equations. An analysis of sion of the controlling subsystem, selected from its the cybernetic method by which a societal system repertory of decisions and executed by the executive actually operates may give more practical informa­ Subsystem corresponding to its repertory of possible tion than any model of the system. For example, a executions, transforms the initial state of the system war or a revolution can be best explained as a into a changed state, which, after being compared specific breakdown in the application of the cyber­ with the goal state, may require another decision, netic method by the controlling and executive Sub­ etc., until the goal state is reached or modified. systems of a given social system. Control is the heart of the cybernetic method, and lt is not difficult to see that the more abstract the information is the blood flowing between the information about a system is, the less effective is organs-the parts of the controlled system through the application of the cybernetic method to the its arteries-the channels of communication. system. For example, the goal of education at a philosophical, i.e. abstract, Ievei was stated by Plato ability of articles published in cybernetic and as the "education of a good man." Different inter­ engineering journals strongly points towards a Iack pretations of "good" at the Ievei of policy making of cybernetic identity. We should also free our led to five national educational systems which vocabulary from engineering jargon, for example, stressed different goals: Education for citizenship substituting "feedback information about a system" (the U.S.A.); Education of elite (France); Education with "information about changed state of a system;" for systematic knowledge (Germany); Education of "negative or positive feedback systems" as character (England); and Education of a professional "systems exercising constraining or promoting con­ (the U.S.S.R.). Of course, no country totally ignored trol," etc. Especially disturbing is an obvious loss of the other aspects of education. At the practical identity of Cybernetics in its relation to General Ievei, the goals were restated in a concrete form Systems Theory, where it serves only as one of determined by a particular school, curriculum, several subordinate parts. The G.S.T. stress on teacher, students, school budget, etc. Thus, the common, general properties of systems and not on cyberneticists, operating at a Ievei which is too ab­ specific ones needed for effective application of the stract, run the risk of not finding a path towards an cybernetic method is a serious obstacle to the application of their findings to concrete systems. development of Cybernetics as an independent This does not mean that we should not try to un­ science. derstand cybernetic processes at a higher Ievei of A pet peeve which, although of relatively minor abstraction since it may provide us with new in­ importance, contributes to the problern of identity is sights, but that we should not forget that practice the current disagreement on the term describing our is the supreme judge of the correctness of any profession: is it cybernetician or cyberneticist? theory, and that we as a profession are judged by "Cybernetician" seems to be the term preferred in our verifiable achievements. England, and shares a common ending with such Let us turn attention to subject matter of Cyber­ words as physician, politician, musician, magician, netics, i.e. to the domain of systems operating by mathematician, etc. Except for the last one, such the cybernetic method. The first living cell which words are not associated with a science, and appeared on our planet about three billion years ago mathematics is certainly not an experimental sei· was the first object to which the concepts of infor­ ence. "Cyberneticist" derives its ending from terms mation, communication, and control could be ap­ such as physicist or biologist, which imply more plied. A system of such immense complexity (for empirical disciplines. ln the more logical languages example, a single E. coli cell consists of about 100 of German, French, and Russian the terms for billion atoms) naturally requires a special controlling professionals in physics, mathematics, and cyber· subsystem. No geospherical system needs control netics have uniform endings: -er, -cien, and -ik from outside, because there is no goal state dif­ respectively. Perhaps we should decide into which ferent from the actual one. Planets in their motion category Cybernetics should fall in English. around the sun do not deviate from the path I have intentionally exaggerated certain points in specified by the Law of Universal Gravitation. There­ this editorial to make them better objects for fore, we should reject attempts to "reduce" com­ discussion. I invite you-our readers-to respond to plex, cybernetically controlled biological, the articles in this issue questioning and defending psychological, and societal processes to much sim­ the achievements of Cybernetics, as weil as to the pler physical ones. Thus, Cybernetics can easily problern of identity of Cybernetics briefly outlined in establish its identity vis-a-vis all those sciences the editorial. I hope that with your help we can con­ whose systems do not operate by the cybernetic tinue this discussion in the following issues of the method. Cybernetics should also emancipate itself Cybernetics Forum. from the engineering sciences, which deal with relatively simple, deterministic systems. lnterchange- V. G. Drozin Notes on the Crisis of Cybernetics

Robert Lilienfeld The City College New York, N.Y. 10031

"One begins to appreciate the value of modern ed­ thought out and elegantly written. But his book re­ ucation only when one hears an expert speaking on mains subject to an important objection: like the a topic outside the area of his specialization." systems men he opposed, his book contained no -Karl Kraus empirical materials, nor did his criticism even sug­ gest the possibility of any test of the claims of the I suspect that we are witnessing not so much a systems theorists. • lt remained for others to ap­ "crisis" of cybernetics as rather simply a collapse of proach the matter empirically. the inflated claims that have been made for it. Just lda R. Hoos, in Systems Analysis and Public as the American world position in general is rapidly Policy (1962), was the first to address the problem; collapsing, and reality emerges through the fog of systems ideas are all very weil (or rather, not very rhetoric, something similar happens in specific areas weil; they are a potpourri of ill-digested materials of endeavor such as cybernetics. The two develop­ gathered from a variety of fields), but what happens ments are even related: the American belief in tech­ when they are put into practice? Focusing for the nical, i.e., a-historical, devices and solutions for what most part on the State of California, the government are inescapably historical dilemmas, has operated of which seems to have been especially prone to throughout the era of American ascendancy; these funding social programs based on the systems/cyber­ devices, whether they be aid programs or servo­ netic approach, Hoos thoroughly documented the mechanisms, do not deliver what is promised. "Cri­ failure of the systems men to deliver according to sis" may be another term for the hangever accom­ their promises. • • Hoos' formidable critique seemed panying a cold grey dawn. unanswerable and unanswered; about all that could Cybernetics did represent, after all, a genuine be said by way of comment was that Hoos focused though modest advance, and automated machines almest exclusively on California, and on the opera­ and factories do operate, more or less effectively. tions research sector of the systems movement, The inflated claims that were made, the sterile pom­ giving only minimal attention to its other branches. posities affered by Norbert Wiener and the other My own approach, in The Rise of Systems Theory founding fathers, their promises to solve the moral, (1975), was to survey the origins of systems theory social, and political problems of the world-all of within a variety of separate disciplines: economics, these claims more than a little self-serving-had to political science, sociology, psychology, communica­ Iead ultimately to disillusionment. But for a time tions theory, and the like. Within each of these they rode a very high horse indeed, producing a vast fields, scholars and thinkers found, or claimed to utopian literature. find, that the systems approach was a fruitful ad­ vance over earlier theories. They then proceeded to Countercurrents extend the claims of systems thinking beyend the Iimits of tightly defined technical problems, affering Even during the rosy dawn of the entire systems them as a solution to social, political, and philo­ movement beginning in the 1950s, there were dis­ sophical problems. senting voices. Robert Boguslaw, in his prescient I then traced the migration of the systems vocabu­ book, The New Utopians, was possibly the first to lary out of these isolated subject-matter fields into recognize the authoritarian and utopian temper of the new attempted syntheses, of which the most the new systems men, and his critique was weil prominent was the GST of Ludwig von Bertalanffy and Ervin Laszlo. From there, the movement • Not a rare phenomenon, but an interesting one: the op­ emerged as a popular ideology in such forms as ponents of a dominant stream of thought, the "antis", The Limits to Growth, the Club of Rome document, often have more in common with their opponents than and the urban and world models of Jay Forrester. they fully realize, or would wish. **But then a State which elects Jerry Brown to prominent My entire criticism of the systems/cybernetics move­ office and fosters other cultural manifestations, might be ment can be summarized as follows: expected to show unusual forms of gullibility. I. The claims of GST to be a source of new knowl- 4 Notes on the Crisis of Cybernetics

edge and insights not available from concrete work tion of increasing the dead hand of bureaucratic on clearly defined problems have not been validated. centralization on previously functioning business Aside from declarations of the interrelatedness of all organizations, government agencies, and economic things, nothing new has emerged. and urban systems. I would only add that the crisis II. The systems approach offers nothing for the of confidence in the cyberneticslsystems approach advancement of work on philosophical problems. does not stop its practitioners (who may no Ionger 111. The attempts at the creation of artificial intelli­ even be believers) from pressing its authoritarian gence, or of models of the human brain via com­ potantials wherever they can. lts emphasis on the puter Simulations or network systems have yielded inter-relatedness of all things within the framework no anthropological insights into the nature of man. of a whole directed and controlled from a brain Nor have the psychiatric-psychological-clinical ther­ center is, of course, the very old metaphor that apy branches of the human sciences made any ad­ likens societies, organizations, and the like, to or­ vances on the basis of systemslcybernetic procla­ ganisms, of which we are the directing brain center, mations. while others are the bones and muselas obeying our IV. The claims of the systemslcybernetics men to commands and sending us appropriate feedback. offer the definitive solution to political and social VIII. I concluded, then, that the cyberneticslsys­ problems such as war, crime, pollution, urban decay tems approach functioned only on the ideological and irrational political strife, all of which would now Ievei, as the mystique of a rising class of tech­ yield to the new approach, have revealed much of nocrats who saw themselves as philosopher-kings, or the narrow, unhistorical mentality of these techni­ perhaps as scientist-kings, who would usher in a cians, especially their astonishing ignorance of the new era if only the present holders of power would genuine achievements of many figures in social turn over the reins to them and their methods. Thus, science. But more than that, their achievements have I saw the cyberneticslsystems approach as a glorifi­ been as meager (actually, nonexistent) as their cation by themselves of the work habits and modes claims have been inflated. Scientists from every field of thought of a narrowly rational, more-or-less edu­ are unwilling to admit that human society remains cated, group of technicians. curiously and blessedly intractable to approaches borrowed from the sciences. * • * * • * * • * * V. The very few instances known to me of prac­ tical applications of cyberneticslsystems theoryloper­ David Berlinski's On Systems Analysis (1976) took ations research to economic and social problems a somewhat different approach. His caustic and have simply not worked. This is too mild a state­ sometimes witty critique of systems theory focuses ment: they have been expensive-and even frivo­ mostly on the internal logic of the various attempts lous-disasters. The systems men for the most part to formalize a systems approach in, perhaps, a set favor general proclamations and programmatic decla­ of differential equations, or in some other mathe­ rations while avoiding specific, concrete problems. matical or pseudo-mathematical notation that decor­ The "promising youth" of these fields threatens to ates so many pages of the systems literature. become their oldest-and only-tradition. He is merciless towards the major documents of VI. These approaches exhibit an unhealthy fascina­ the cyberneticslsystems literature; his survey con­ tion for exercises in vocabulary construction as a tains descriptions like the following: " ... a flabby substitute for either thought about the world, or and pretentious book ..." (Ashby's lntroduction to observation of it. This, I suggested, was a hallmark Cybernetics), " ... pages and pages of soggy philo­ of the 20th century, a mostly stagnant interregnum sophy" (Norbert Wiener's Cybernetics), " .. . a collec­ during which technologism has operated as increas­ tion of coughs in the night" (Walter Buckley's ingly diffident regent. lts regency was made possible Modern Systems Research for the Behavioral Seien­ only by the exhaustion and discrediting of previously fist); " ... there is much in his style of prose com­ regnant political ideas: the Divine Right of Kings; position that calls cotton wool to mind . . . he is the Social Contract; the Invisible Hand of the Mar­ not an author of passionate scrupulousness" (David ketplace; the Dictatorship of the Proletariat; Govern- Easton's writings on political science); " ... pile up ment of'ß the People, by the People, for the People. an imposingly complex system of equations and All such images have lost much of their ability to then subject them to an analysis of ineffable inno­ inspire either belief or action; they survive, but most­ cence. lt is a natural prescription for theorists igno­ ly in the rhetoric of public speechmaking. ln such a rant of differential theory" (Jay Forrester's World vacuum, the cyberneticlsystems approach might ap­ Dynamics, and The Limits to Growth); " ... there is pear as an answer: an approach based on statistical, no suggestion whatsoever that beyond the loose, in­ actuarial, or algorithmic procedures might seem pro­ formal, and disorganized bavardage there stands a mising for a time. set of precise mathematical concepts" (Morton Kap­ VII. The stifling authoritarian potential of the lan's System and Process in International Politics). cyberneticlsystems approach was noted by many I hasten to add here that I have quoted only a few writers; its general effect could only be in the direc- of Professor Berlinski's judgments, most of which Cybernetics Forum 5 have been preceded by very detailed and scrupulous almost belongs in the same category: "A social analyses of a number of systems formalisms, which scientist studies social systems to get at their laws" he groups under three major headings: General Sys­ (David Berlinski). tems Theory, Dynamical Systems, and Mathematical This statement is possible only for someone who Systems Theory. But the general dritt is clear: the has never heard of such thinkers as Wilhelm Dilthey, mathematics is often ceremonial or downright erro­ Max Weber, or R.G. Collingwood, to name just a neous; where the models make sense, their appli· few, or who conceives of social science on a posi­ cability is doubtful. tivist model drawn-1 was almost going to say, from Professor Berlinski concludes his survey with the Auguste Comte-not so much from Auguste Comte following rueful paragraph: as from J.S. Mill's simplistic version of Comte's philosophy. Both the cyberneticists and, it would There may be a moral in this record of grand appear, some of their critics, would do weil to learn efforts brought low by insufficient means, some something of the philosophical struggles which, cagey bit of philosophical wisdom. But aside from the obvious counsel that in great things great since the late 19th century, have attempted to come ambitions without great theories are insufficient, to terms with historicity and the peculiar problems I do not know what it is. it generates, which have taught many of us "para· plegics" that the notion of "laws" of society analo· gous to "laws of nature" is a will·of-the-wisp long since abandoned. We remain aware that the prob· I should like now to offer a few hesitant words lems of the social sciences are bound up with the that might suggest, for Professor Berlinski and the unresolved problems of dealing with society in the cyberneticists, just what the moral is. But before full amplitude of its concreteness. doing so, there is a small bone to pick. First, Professor Berlinski aims a barbed shaft at .. . Only certain truths, which do in fact appear the social sciences; speaking of cybernetics and in· unique and exclusive, possess 'absoluteness': formation theory, he says: these are abstract truths. But there is no difficulty here: abstract truths da not refer to reality, but to diagrammatic renderings of reality: i.e. to certain Both subjects are shaped about real theories, but elements or aspects of it (idea means aspect) ob­ their applications have Iaken place in the para· tained from a fixed point of view and with delib· plegic disciplines-sociology, psychology, political erate elimination of other points of view-that is science, and management science-a sure sign of precisely what abstraction consists in-and in debility . . . Certainly the positive results have not such a case truth must naturally be univocal and been impressive . .. exclusive . . . What has happened is that . . . logic and the theory of knowledge have been ac­ Speaking as a sociologist who is often irritated customed for centuries to work, with surprising with the state of the social sciences, I would sug­ pertinacity, on abstract truths, and have hardly gest to Professor Berlinski that the true paraplegics ever descended to thinking about the concrete, are the mathematicians and engineers who think to which presents much greater difficulties. And the become dilettantes of the social sciences on the result has been to consider as a mark of truth basis of some new technical or formal advances in something which only applies to a very special and secondary class of truths, and which only has their fields; along with them go the social scientists meaning for the consideration of that " de-real­ who think to become dilettantes of mathematics and ized " reality which appears when man devotes logic, and who take Science (with a capital S) more hirnself to a task which is highly specialized and seriously than it deserves. not at all basic in his life . . . disciplines like Second, Professor Berlinski often has his bit of logic and the theory of knowledge would assume tun by quoting sentences from the systems theorists a very different appearance if they were to state their problems in all their radicality and amplitude, which might better have not been written; thus, he so that the present forms of these disciplines quotes such gems of wisdom as the following: would be reduced to mere chapters corresponding "Ours is a complex world" (Ervin Laszlo.) to certain particular cases. • "Carthage was destroyed by the Romans. This is called system destruction or dissolution." (Morton A. Our systems thinkers, and unfortunately many Kaplan.) philosophers as weil, are unable to confront the "An automobile which runs ou t of fuel will not fundamental error in social science involved in re­ function" (Oran R. Young). treating from full concreteness in favor of abstract "The analogies between a living organism and a schemes. The abstract schemes may be any or all of machine hold true to a remarkable extent at all the many "systems," or may be some simpler theo­ Ieveis at which it is investigated" (Jean Pierre ries of social causation relating society to "under­ Changeux). lying" causes such as technological or economic Now, I am grateful to Professor Berlinski for dis­ developments, but all such attempts do involve a playing these howlers drawn from the systems litera­ retreat from reality in favor of logical or pseudo· ture, but I fear I have found a sentence of his which logical schemes which gain manipulability at the 6 Notes on the Crisis of Cybernetics

cost of reality. tions away from the concrete. They have not had All such schemes are involved with "causal ex­ much help from either their allies or their critics, planations;" but causal explanation has to be sche­ many of whom appear to share the same errors. matic and universal, leaving out the concrete indi· But it appears to me that the fundamental error viduality which is "the most real part of the thing" of the systems approach is this basic philosophical explained;* • we are faced then with a whole new confusion of attempting either to retreat from the region of problems. ls it possible, then, that "ex­ concrete in favor of manipulability, or to generalize planation" is a form of knowledge that is both specific configurations. The quantifiers have for a secondary and defective, and that another, and long time assaulted the social and historical neglected, form of knowledge must precede it? This sciences; the latest attempt. that of the systems/ neglected form of knowledge would be something cyberneticists, has ended in the same sort of failure. called description, something which phenomenolo­ The failure will not be reversed by bigger and better gists, historians, and even social scientists, have computers, by more variables, by better and newer been attempting. mathematics, but by a somewhat chastened aware­ ln their own groping way., I think some cyberneti· ness of the historicity of all forms of knowledge, cists have stumbled across these problems; some of and of the need to confront social reality in its full their abstract schemes appear to be attempts to concreteness. When this awareness emerges, the capture in formal terms the parameters of (compara­ cyberneticists may even discover that some disci­ tively) specific configurations of factors in social plines have been there ahead of them: these are the Situations, but a basic confusion of thought Ieads humanities. them to generalize these in more schematic direc-

New York City; February, 1980

* Julian Marias: Reason and Life, translated by K.S. Reid and E. Sarmiento, (1956, Yale University Press, New Haven, reprinted by Greenwood Press, Westport, Connec­ ticut, 1975), pp. 126-7). **lbid., p. 145. Don't Bite My Finger, Look At Where lt Points

Some Notes on Lilienfeld's ldeological Analysis

Arie Ariely COMTEN, lnc. 3 Choke Cherry Rd. Rockville, Maryland 20850

Introductory Notes ambiguous term) make philosophical and societal claims on the basis of their technical work. ln an Theorizing is an important part of the cognitive arbitrary manner, I will consider this to be the prob­ contents of science. Every scientist is an observer, lern that Lilienfeld addresses. (Though, it is possible and as an observer he is continuously busy attempt­ that Lilienfeld considers his theory to be part of the ing to organize and order his observations, regard­ error-reduction mechanism, the societal claims to be less of the nature of the phenomenon observed, be the tentative theories, and system theory to be the it parts, facts, numbers, data, etc. Our theories con­ problern requiring explanation.) By assigning to vey the interpretations formed in our minds to other Lilienfeld's theory the rote of a problern (instead of a members of our society. We try from time to time to refutation technique) we are approaching it from the publish our theories in articles, essays and text­ point-of-view that is habitually called "second-order­ books partly to boost our egoes, and partly because cybernetics": our interest is not in the observed sys­ by reporting our Observations we invite support, tems (system theory and its refuted claims) but in review, and criticism. the observing system (Lilienfeld and his interpreta­ lt is with this framewerk that my evaluation of tions). Thus, my intention in these notes is to con­ Lilienfeld's theory deals. Dr. Lilienfeld(1) observed a tribute to the growth of knowledge, as defined by phenomenon; organizing his findings he published a Popper, and to transform, by a review, Lilienfeld's theory: The rise of system theory-an ideological theory from a valid verification tool to a more proper analysis. To attempt to enumerate the value and the and appropriate position of problem. damage that this theory contains, we must analyse it in terms of contents, explanatory value, predictive power and consistency. Model p, ~ TT, ~EE.~ p, The weil known philosopher of science Karl Lilienfeld's System Societal Lilienfeld's Systems Popper(2) habitually attempts to explain the growth perception theory claims Jl'theory theory of knowledge, by the use of a tetradie 'schema': at fault My Societal Lilienfeld's My notes Lilienfeld's Perception claims Theory on p, -+ TT, -+ EE, -+ p, Lilienfeld's theory at Jl theory. fault. The existence of a research-problem (P,) will cause The reader's Lilienfeld's My Reader's -?- the emergence of tentative theories (TT •), attempting Perception theory. notes on agreement to provide an explanation to the problem. The scien­ Lilienfeld's or theory. disagreement. tific process of review by error elimination (EE,) may eliminate most of the theories, or may fail to refute As stated; Truth, like contact lenses-is in the eye of the them, but will contribute to the emergence of a new beholder. research problern (P,), which may be the solution to the original problem, but could be also a reformula­ Lilienfeld's theory (and I shall refer to it from now tion of the problem. And of course the cycle starts on as the theory) appears to be constructed on the anew ... basis of the sociological model, described by Piet This is the model that I will use to evaluate Lilien­ Thoenes as "the role of the elite in the welfare­ feld's theory. state." Since I am not familiar with the original ln his preface (starting in page 2) Lilienfeld states model, I will summarize it based on Lilienfeld's des­ that his theory is based on his interpretation of cription (p. 271) as a three stage evolutionary pro­ books and articles, in which system technicians (an cess: "The announcement of the message, the car- 8 Don't Bite My Finger rying out of the plan set forth in the message, and becomes very dangeraus in the other parts, since it the conservation of order achieved by carrying out establishes a sense of belief in the author, without the plan." differentiating between his summary-ability and his The similarity between Thoenes's model and original arguments. Lilienfeld's theory is presented below in a table Unfortunately, the components of the theory of form: general-systems that were selected by the author pre­ sent a very restrictive picture of the origins of the field. The concentration is on the earlier works of Thoenes's model stages Lilienfeld's theory Von-Bertallanfy, Laszlo, Klir and their contempo­ stage # name part # name raries. lt seems that all development stops abruptly 1. The announcement I The origins of and that more current contributers to the field were of the message systems theory omitted. I found the omission of Warren McCulluch 2. The carrying out of II Societal claims of worrisome, since it implied Iack of recognition of his the plan set forth system thinkers. in the message ideas in the development of the field, and ignorance 3. The conservation 111 System theory as an about the influence of McCulluch on more current of order achieved by ideology. development, such as Experimental Biology (Matu­ carrying out the rana and Varella), the Biological Computer Lab plan . (Heinz Von-Foerester) and development abroad (Gordon Pask and lllya Prigogine). The rationale for Table 1: lsomorphism between Thoenes and Lilienfeld such omission is that the author concentrated on textbooks and other published works directed to the general public, while avoiding the more technical (As a personal note, I found in the construction sources, based on their Iack of missionary-content. of the table above that it (the description) affered The selection of works to be criticized based on me an aesthetic delight that is its own justification, the ratio between their technical contents and their just as Lilienfeld said (p. 191).) missionary-contents, with greater appreciation to The existence of a correspondence between missionary works, is biased. lt is biased because of Thoenes and Lilienfeld may provide us with a specu­ Iack of criteria to separate technical work made­ lative clue, that the strength of the author's attack simple, to increase the number of people that de­ on systems theory can be explained by the threat velop appreciation to the problem, from work made that the emergence of a technocratic elite poses to complex due to its author's desire to achieve a posi­ Lilienfeld's world. The selection of isomorphism as a tion in areas other than those of his competence, model for the theory will be discussed later on, by masquerading his personal opinions as his expert when I shall attempt to parry Lilienfeld's attack on opinion. A scientiflc article in the Scientific Ameri­ model selectioh within the discipline. can is an appropriate example of the first kind; an The main thrust of my attack on Lilienfeld's work attempt to write an article that will call for the elec­ should be directed towards the concept of "carrying tion of a cybernetician as a president, due to the out the plan, set forth in the message." lf we can capabilities implied by cybernetics, will be con­ refute an intent to use, or rather abuse knowingly sidered an appropriate example of the latter, if writ­ systems concepts to advance the discipline, while ten by a politically ambitious cybernetician, using contributing to the detriment of society, we then can technical terms to obscure his frustrations or Iack disconnect the link between systems theory, and its of knowledge in political affairs. Since the author perverted usage by a technocratic elite, to pursue elected to selectively define theoretical "corner­ its own aims. stones" he consciously decided to bypass the cri­ This approach will also enable us to avoid the cir­ terion of meaning, and instead of filtering the "werk­ cular arguments about elites, their aims and the con­ sequences ' to society, a topic that lies beyond the made-simple," from "work-made-complex" he grouped them together. This Iack of Interpretation of his sub­ horizon and the scope of tbis paper and could be ject matter, or his biased Interpretation ·may have regarded as highly speculative, to say the least. contributed to the selection of his model. The first part of the theory, organizes the work By clearing my impressions about the first part of that was done in the early stages of the discipline. The author is reviewing the "announcement of the the theory out of the way, I am ready now to ap­ message" by presenting concise summaries of early proach the subject of societal claims that were ori­ works in General System Theory and the other dis­ ginated by system thinkers, and were found to be ciplines that emerged from it or converged to it: worthless by the author. Cybernetics, Information Theory, and Artificial lntelli­ gence to mention only a few. These summaries en­ able a layman as weil as practitioner to grasp the main themes and models, while avoiding the rhetoric ls system theory a theory? of most of the seminal works. The clarity expressed ln summing up the societal claims of system in the summarization of these historical documents thinkers (another ambiguous term), Lilienfeld states Cybernetics Forum 9 that: poor assessment of any field is inevitable. A theory is never a complete and faithful descrip­ "System thinkers exhibit a fascination for defini­ tion of "reality;" it is little more than a model or an tions, conceptualizations, and programmatic State­ analogy. Theories imply that out there, there are rela­ ments of a vaguely benevolent, vaguely moralizing tions, there are logical operations that are similar nature without concrete, or specific references to historical, social, or even scientific substance __ ." with what the theory purports to explain, but the (p. 191) "similarity" is of such vagueness that theory can not be fully identified with the problern it explains. According to this argument, the definitions, concep­ A theorist does not attempt to prove, nor assert tualizations and the like which should be considered the verisimilitude of his theory, nor does he attempt as the "product" of system-thinkers cannot be an­ to anchor it in referential or concrete substance, this chored into any concrete substance. My counter ar­ will be determined, later, by further research and ex­ gument is: why should they? Scientific development, perimentation, and most likely by others. emergence, and evolution (pick your own term) is a lt is within this framewerk that "system thinkers" dynamic and therefore unpredictable process. The and their product should be validated, as M. foundations of what will be discovered tomorrow are Polyani(3) states: not anchored in any concrete substance today, as a matter of fact they cannot be anchored into a sub­ "These powers enable us to evoke our con­ stance a priori, otherwise what is there to be dis­ ception of a complex inaffable subject matter covered? Statterd Beer's adage: "absolutum-obsole­ with which we are familiar by even the rough­ tum" (if it works it is obsolete) can be verified by est sketch of any of its specifiable features. A scientist can accept therefore, the most in­ viewing all the famous cases of "concrete anchor­ adequate and misleading formulation of his ing," like the position of Gallileo's detractors, in the own scientific principles without ever realizing question of his telescope: their objections were an­ what is being said, because he automatically chored into substantive evidence, but were they supplements it by his tacit knowledge of right? Does the physics textbook, which is anchored what science really is, and this makes the in "reality" present the frontier of physics? I don't formulation ring true." think so. The determination of "substantive" or "lacking substance" is subjective judgment, an­ chored in an individual'$ perception of his world, ls the search for isomorphic models and nothing more. reliable? Do these models contribute to We can view any discipline as containing two non­ intersecting domains: that of what was proved al­ scientific development? ready and verified by repetitive experiments, and that A secend argument is being pursued by the which contains unproven hypotheses, unsubstantiated theory, namely: claims, conceptualizations, and often conflicting theories. We will view the former as the enclosure "They (system thinkers) collect analogies be­ of the domesticated problems, and the latter as the tween phenomena of one field and those of enclosure of the "wild" problems. Furthermore, un­ another (preferring to call them isomorphism, though the difference is not discernible), the less our interest is in the history of the discipline, description of which seems to offer them an we will strive to become associated with the set of aesthetic delight that is its own justificatioh." unsolved problems, and cohceptualize some hypo­ (4) theses that may reduce the uncertainty in a particu­ lar field. But can we criticize the preparatory work lt is amusing to discover, in retrospect, that Lilien­ that precedes discovery as lacking in substance? feld's theory seems to be similar (my way of avoiding What is unsubstantive and Iacks references today, "isomorphisms") to another theory (see table I); may be verified tomorrow, and once accepted by the could it be that the author, in formulating his own majority of the practitioners, will become the core of theory, conveys a message: "do what I say, not what some future textbook, where it will be considered I do?" both well-anchored, and removed from the repertoire My counter argument to the theory's suspicion of current research problems. about the intent and the usage of analogies in sys­ System theory is relatively a very young discipline, tem theory is: Why not? and as such it Iacks consensus on definitions and Let me present side by side two approaches conceptualizations. lts immaturity, is the prime con­ towards the solution of a hypothetical problem; one tributing factor to the fact that most of the work will call for a selection of an analogy in different done until now, is within the domain of the "wild" tields and its usage to formulate a solution, or at problems, while very little can be found that can be least tentative theories that will explain the problern considered as "verified." But if early Statements, at hand; the other will call for tormalization of the which in this case are the basis for the argument, problern by axiomatization (which is nothing more are mistakenly considered as an ultimate result, a than borrowing a mathematical analogy). What can 10 Don't Bite My Finger

we assert about these approaches? tribute, according to Ziman(6), to the task of keeping When one is to "borrow" an existing model and all scientific disciplines in touch with "reality." transfer it to another discipline, one must always Models should not be equated to mathematical remember that the model, being drawn from another techniques, though in some cases similarity or even field of knowledge than that to which it is to be identity relationships may exist. Models are the applied, carries a certain amount of pre-existing basis for any system research programme. We use understanding of its properties. An attempt to evade them to formulate tentative theories, and as our the consequences of such understanding could be critical analysis tools. We even incorporate into fatal. lf this ignoring of pre-existing understanding them any knowledge gained and any increment in is Lilienfeld's main complaint, we have no argument understanding that was identified during our Obser­ at this point. But this is a case of "bad" science, vations, hypothesis formulation, and vertification. and we must remember that system theory (like any As far as the aesthetic delight associated with other discipline) includes both good and bad analogies, I always consider it as the reward of "dis­ science. lf this is not the major reason for Lilien­ covery" to its discoverer, or the sense of elation feld's objection, we will proceed with our initial ex­ that usually succeeds the tension and frustration of ample. lf the trap of incommensurability between the research. The rewards of a discovery are common to model and the problern that it was brought to solve all human activities, to science as we!l as art. There was avoided, then I found that the knowledge con­ is no reason to expect system thinkers to be differ­ veyed in this transfer goes further than the assembly ent from society, or to behave differently. lt is the of techniques used to solve mathematical problems, tortune of system theory that the implicit use of because the model contains elements of both intui­ models enables its practitioners to enjoy the aesthe­ tive and experimental understanding. To exclude tic delight in their discovery with greater frequency such understanding, by axiomatization of the model than in other areas of human activity. Unfortunately prior to its application, would eliminate some in­ Lilienfeld's theory misses completely in this point. sights about the subject for which the hypothesis was originally proposed. Analogies enable us to gain insight without sacri­ The universality and abstractness ficing details and variety, and the act of collecting analogies is generally accepted not only by system of system theory thinkers, but among scientists in other disciplines Another of the theory's anti-system arguments is too. that: All of our knowledge is stored in an archive, where all information pertaining to a particular field "System theory achieves its self-encompass­ of knowledge can be found in a specific niche with­ ing 'Universality' only by its very abstractness: in this archive; there is a niche for every specific All things are systems by virtue of ignoring discipline. All members within this discipline share the specific, the concrete and substantive"(4) this knowledge and cooperate in its verification and validation as a prerequisite for increase of the My counter-argument is: so is Topology, group knowledge content of the field. Transfer across dis­ theory, and the physical theory of 'black-holes.' cipline boundary is not common-neither in our A theory can convey a significant amount of infor­ model of knowledge nor in the "real-world." Thus, mation, without being metrically accurate. This is disciplines adopt characteristics similar to those of known as Qualitative knowledge. ls such knowledge "closed systems;" minimizing. interactions and unscientific because it is not quantitative? communicatron with adjacent · dTsciplines. With the Of more significance is the question: Does a passage of time, this closure results in a state theory correctly represent the significant relation­ where the experts in a field become so indoctrinated ships between identifiable entities in the subject in the current set of paradigms that their critical and matter? Relationships often can not be resolved nor imaginative progress comes to a halt. lt happened to described by 'counting' or 'measuring.' Consider the physics prior to the discovery of relativfty, it hap­ following argument(7): pened ·to geology before the advancement of the continental-drift theory. No discipline is immune to To the scientist there is no confusion in such disease. This static state of affairs will be re­ treating an atom as a. tiny sphere, in the versed when intellectual intruders appear through kinetic theory of gases, yet knowing that its the interdisciplinary boundaries and Iook at the field spectroscopic properlies require that it should without pre-conceptions. The facility that separates be thought of as a horribly complex cloud of these intruders from ordinary scientists is not innate interpenetrating electron Orbitals buzzing around a central nucleus. The real question nor is it genetic mutation. lt is the result of being for the scientist . . . is whether the Ievei of trained to Iook for models in other niches, and the generality is appropriate to the problern at skills acquired while transforming solutions across hand, and whether the details are then suf­ boundaries. The skills acquired and the training con- ficently accurate to solve it." Cybernetics Forum 11

'Systems description' is nothing more than the ab­ System philosophy: Subservient to straction of the 'real' problem, the intention is to describe some phenomena (as components) and the science? Passive? Should it be? relations among the components. This is not a de­ According to Lilienfeld's argument: liberate attempt to avoid the specific, the concrete and the substantive, but an approach that guaran­ "System philosophy remains philosophically tees that the specific will not Iead us "to lose the passive and subservient to the sciences . . . forest on account of the trees." (while) ... the sciences themselves prove to The universality of the term "system" should be Iabor under philosophical difficulties that evaluated within the context of science per-se. Sys­ remain unsolved"(5) tem theory was derived from what is generally acknowledged as modern, or western science. lf we review the history of philosophy we must Modern science is monolithic and monopolistic, conclude that philosophy was subservient to religion having successfully eliminated all competition in the during most of the epoches. During the middle ages, 17th century. lt is in this frame that we consider the Catholic church labored under philosophical science as universal. The universality of science can (theological) difficulties that remained unsolved for a be demonstrated by the facts that all physicists long period of time, while philosophy remained pas­ share corrimon paradigms and communicate among sive and subservient to the church. themselves by the use of their consensual language. Did the theological difficulties impair scientific de­ lt will be correct to conclude that scientific univer­ velopment, or did they serve as what Kuhn will refer sality was forced on us by the structure of science, to as "the essential tension" between innovativeness and the system theory only adhered to what was and secretarianism that opposed Innovation, using expected from it to pass from parascience into dogma as its most powerful and restricting tool? scientific status. Did any attempt by a religious or other bureaucratic Let me elaborate on the Statement that universal­ organization ever succeed in stifling radical changes ity is forced by the structure of science. When we in the way that scientists perceived 'reality'? ls it become involved in scientific activity, be it learning not true to state that the scientific progress that we the rules of a specific scientific discipline, or exper­ enjoy today, either directly as scientists or indirectly imenting and verifying the current paradigms, we as members of society, is the result of this "sub­ must use a consensual language to communicate. A servience"? communication of knowledge contains both ambigu­ Using the same analogy, to present the case for ity and variety, since it is being handled as a subset system theory, we can state that due to the chang­ of a natural language. An investigation into the ing picture of society at, or around the time that existing cultural differences in the structure of systems-theory was conceived, the themata selected natural languages will Iead us to view what was by the pioneers as starting point was not religion referred to before as 'universality' as nothing more but the physical sciences, at that time the most than the result of a 'schooling' (to differentiate from successful enterprise of western science. learning) process that forces all written languages to System theory remains subservient to epistomo­ conform to universal Standards of contents and logical problems posited upon science, by virtue of structure. lt is this form of language-writing, deliber­ its dependency on science as its themata. To define ately taught in our schools, that develops 'abstract" this dependency as "passive" becomes therefore a thought as characteristic of scientific standards. matter of style, not content. Abstraction played a significant role in the develop­ Using the Popperian model I can present the argu­ ment of mathematics, and the significance is the ment as follows: fact that abstractness is related to the more ad­ vanced stages of evolution of mathematical con­ cepts: first there were the quantities to be measured (EE,a) and counted; in its evolution, the discipline dis­ Life,cosmology -+philosophy ._. physical sciences ~ covered and introduced numbers-a more advanced religion ~ social sciences ~ ? (and abstract) form of the basic quantities; only at (P•) (TT,) (EE,b) (P,) maturation did the concept of "number" emerge­ more abstract and more universal. I maintain that "universality" and "abstractness" Can it be that we consider the unknown solution to are complementary relations, through which complex the problems as "system theory," where the solution problems can be analyzed and, hopefully, solved. is shift in the original problern based on changes The inabllity to appreciate the power of abstractnass in the scientific thinking that is 'subservient' to is unfortunately a problematic deficiency of those philosophy? And even if system philosophy could who did not acquire the mathematical foundation transform itself trom , 'passive' to 'active' role, how necessary to operate on an abstract Ievei. could it free itself from 'subservience' to the sciences? 12 Don't Bite My Finger

Failure of system theory or failure of the restrictions mentioned above. The implicit usage of metaphors within the frame theory applicators? of system theory provided possibilities to social Repeatedly through the theory, the author builds a scientists to formulate problems and research pro­ case that is based on what can be considered as a grammes by transforming models that were proven major 'weakness' of system theory or the minimal in other disciplines. success that the theory achieved in the solution of Therefore the lack-of-success that the theory point­ problems in science, as weil as in social arena: edly emphasizes, must be considered within the possibilities of mismatch between the model and the "Where it failed (system theory) in applica­ problem, or mismatch between the model and the tion, we are told that the theory is fine but current paradigms of the field where the experiment the application was poorly carried out. This of took place. course draws the rebuke that if a technique Any theory that is general, this is to say abstract is not judged by its application how can it be judged? .. . " (p. 192) in the sense that the more ·you have to agree on it, the less content there is, can not and should not Let's start the response to this argument by con­ attempt to verify itself (see my early comments structing an iconic model of the relation between about the role of theory). lt should be used as a theory and its application; let's consider the theory model, map, or better album of maps to be studied (T) as the model of a particular problem, let's con­ not . ~s an ultimate truth but as way-points that the sider the application (APP) as the programme car­ applicator can use, provided that he verified the ried, based on the theory, to an end in the 'real mudel by his intuition, imagination and the condi­ world.' lt is clear that between the theory and its tions of the problern at hand. application we must introduce a critical com­ The relative youth of system theory affected the ponent-the applicator (A) as the person, group or process by which a practitioner became a full­ machine that uses the theory as a map to "cook" fledged member of the discipline: there is no institu­ an application (technique). Graphically our iconic tional standard of education, and no agreed-upon model will Iook like: learning processes by which paradigms can be veri­ fied and val!dated via communal consensus. This freedom to declare oneself as a system thinker has T------~A---t~e-c~hn~i~q-ue--~)~.APP contributed to the field by drawing numerous bril­ liant individuals who gained formal training in differ­ The interlocking role played by the applicator in ent disciplines, as weil as charlatans that feit that the model can point us to the starting point of an the Iack of formal procedures and consensus will analysis, where the application is considered either enable them to establish themselves as members of as a 'failure' or as a 'success.' Therefore we can the new discipline. All that is required from an indi- . divide the model into two components: The relation vidual to declare hirnself as a system thinker is his between theory and the applicator, and the relation ability to use system terms, in proper or improper between the applicator and the application. Situations. ln general, Lilienfeld has some value to his criti­ lt is the interpretations of the theory into opara­ cism; it is not the ultimate truth of the system pic­ tional terms that is wholly applicator-dependent and ture that matters, but the answer to particular ques­ can not be checked, nor verified by comrriittee of his tions posited by 'system thinkers' that count. But I peers. The interdisciplinary nature of the models will try to be more specific in discussing both the used leaves the verification of a model to the appli­ questions and the answers. cator; if the applicator abuses this opportunity to When we have a theory, the most accepted man­ farther his own ends and aims-it is the applicator's ner by which we can either test it or use it is failure and not the theory's. through experimentation, verification and validation The relation between the applicator and his appli­ of the results. ln the world of the physical sciences cation is based again on the applicator's ethical Ievei this can be performed in a laboratory, using elab­ and his education as to what is considered 'bad' orat~ instrumentation, in a controlled environment. science and what is considered 'good' science. lf Since the subject matter slowed considerably its one camouflages his own narrow opinions by at­ evolution, experiments can be carried out with rela­ tempting to force a model on a problem-it is a tive ease, and with the ability to predict (because of case of 'bad' science, whether in genetic research or the cessation of evolution). ln the social sciences, in system theory; however, if one attempts to study on the other hand, test and experimentation are a model in order to better understand the magnitude restricted due to non-controlled environment, Iack of of the problern (like Stafford Beer did in Chile and instrumentation and constant dynamics of evolution: Canada) this is 'good' science, though it does not "You can not step twice into the same river." The guarantee the results in terms of desirability or appeal of system theory to social scientists was in validity. We can learn from our mistakes more than the ease by which its usage can bypass some of from our successes. Cybernetics Forum 13

The theory is confusing the term 'system analysis' outside of context, coupled to speculation that sys­ with system theory repeatedly. This categorical mis­ tem theory may be put to work to the detriment of take of assuming that any application of system mankind, by a technocratic elite. This speculative theory in the social arena must adhere to the usage part evades the thought that no sicentific theory is of large scale digital-simulation, contributes more immune to misuse of inappropriate interpretation. than any other factor to Lilienfeld's confusion. The According to this vision it is the fault of the medi­ attempt to utilize system models in social problems um, if it is used for antisocial purposes. is relatively young and maybe could be considered a in both versions of the value of system theory, it posteriori-immature, but at least it provided us with is considered by the theory as dangeraus to organ­ important lessons about the limitations of system ized knowledge and free society. Social scientists theory, and with evidence that more basic research are singled out as prime candidates for coercion, is needed in system theory before it could consider since by adopting system terminology they lower itself as capable of solving social problems. their skeptical guard and neglect the consequences lt is these instances where theory fails to provide of the theory. explanation, that signify the beginning of "paradig­ lt seems to me that para- and superscience are matic. shift," which may or may not provide better two non-intersecting domains, therefore system theories and explanations of 'reality.' lf this is the theory can not reside in both at the same time. Any case, then in spite of Iack of understanding by appli­ attempt to create an intersection of those two cators, system theory may prevail, and only time will separate domains will result in contradictions. judge whether these 'failures' are indicative of the in­ The scenario that seems more feasible to me is applicability of the · system theory (Lilienfeld's posi­ that system theory started as a parascience, like tion) or just a single episode in the area of scientific any other new, immature and dynamic discipline, but development (the position that I take on this issue). gained recognition during the last decades and was Can we judge mathematics as a failing theory, incorporated into the archives of science knowledge and geologists as poor applicators, only because of (maybe to Lilienfeld's distress). I doubt whether the fact (now, only historical episode) that during the system theory, because of 'newness,' could provide a debate about the theory of continental-drift the sound ideology to any power-hungry elite group, theory was poorly used? Was this a case of poor whether it is scientific, social, political, or bureau­ theory or poor application? Of course we can not cratic. This is precisely because of its dynamic evo­ categorically maintain that it is the fault of the lutionary nature. Assurne for example that an elite theory. Why, then, should we single out system group adopted system theory as an ideology during theory? the peak of the closed-system paradigm, will not the shift from closed- to open-system paradigm affect the political ramifications of such ideology faster than the ability of the elite group to adjust its posi­ ln Summary: System Theory­ tion to reflect the change? parascience or superscience? Any attempt to associate ideology to scientific theory raises some questions about the motive that The consistency of the theory is grossly impaired lies unoerneath. Lilienfeld appears to play the role by the author's attempt to oscillate between two of proxy for mankind, and guardian of scientific reli­ contradicting opinions about system theory. One, ability and social freedom. This is quite a demanding that prevails in the first two parts of the theory, is commitment on his side. One may suspect that such that system theory is nothing more than a para­ commitment is limited to the influx of system ideas science. The other is that by its emergence as ideol­ as long as they will conform to his paradigms of ogy system theory (in the last part) is superscience verification, validation, and prediction. This, carrying (scientist-king, due to the system ideology). 8oth the "Gallileo mantle," is a dangeraus position for opinions are the result of two-valued logic, I main­ any scientist to assume. tain that system theory is somewhere in between. lf his intention was to make rational assessment Parascience can be viewed as a fringe of ideas, of our ignorance on the particular application of sys­ surrounding the boundaries of institutional science, tem theory in the social sciences, by identifying waiting and seeking formal recognition. lts propaga­ enigmas and raising consensual questions, his tors in many cases are asking for more than being model could initiate a healthy discussion and dis­ heard, they want to be accepted without passing sent in the field. But the methodology used in the through any verification attempt. According to the construction of his theory affected its explanatory theory, this is the position of the system thinkers, powers, its prediction potential and its consistency. who attempt to predict and promise social solutions I have some criticism about the current state of due to their systemic insights, but fail to deliver. affairs in the discipline: the disappearance of real The second position, that of superscience, is not system thinkers, our recurring attempts to concen­ grounded in any serious system theory work, it is trate on professional societies instead of society a combination of interpretation of early seminal work, and its problems, and the universal misuse of terms 14 Don't Bite My Finger and models by applicators, all cause me discomfort. But I view these deficiencies as applicator-oriented phenomena, and not as weaknesses of the theory. We are, after all, human. Mv last Observation is: I have used in this paper definition, conceptualiza­ tion, analysis, isomorphic models, abstract, universal and some philosphical terms, in short all the argu­ ments that Lilienfeld raised against system theory and the people associated with it. lf all is wrang and misleading the fault is mine. lf all is right and appropriate, all credit is mine also. Systems theory did not create these notes, this writer did.

REFERENCES 1. Lilienfeld, Robert, The Rise of System Theory­ An ldeological Analysis, Wiley lnterscience, p. 287, 1978. 2. Popper, Karl, Objective Knowledge, Oxford Press, p. 243, 1979. 3. Polyani, Michael, Personal Knowledge, Routledge, Kegan & Paul, p. 169. 4. Lilienfeld, p. 191. 5. Op. cit. p. 192. 6. Ziman, John, Reliable Knowledge, Cambridge, p. 134, 1978. 7. Op. cit, p. 84. Review* of Berl i nski 's ''On System Analysis''**

Joseph P. Martino University of Dayton Dayton, Ohio 45469

The dustjacket describes this book as follows: issues of what are systems, what is GST, and why "Here is a book of uncompromising negativism. Sys­ should GST be studied as a separate discipline. ln tems analysi·S, David Berlinski observes, is largely a most of Berlinski's quotations, it is clear the writers sham; such content as it has involves nothing more are trying to generalize from physical systems to than the purely ornamental uses of mathematics." biological, social and political systems. Consciously (Emphasis in original.) I regard this as a fair assess­ or unconsciously, they draw analogies with physical ment of the book. lt is indeed a vigorous attack on systems and hope to emulate the success Systems Systems Analysis in its many manifestations (such Analysis has had in dealing with such systems. as General Systems Theory and World Dynamics), Berlinski first takes on von Bertalanffy. He pre­ prosecuted by an author who is both a Professor of sents several quotations from von Bertalanffy's Philosophy and a competent mathematician. writings to demonstrate that von Bertalanffy's ob­ To say that Systems Analysis is largely a sham is jective was to derive in a formal way laws true of all strong language. Many would reject the book at the systems, whether these systems be physical, biolog­ outset on that basis alone. However, the author's ical or social. Berlinski's attack first no1es that only qualifications, as evidenced by both his credentials the laws of logic are true in all systems, and these and by his writing in the book itself, are such that laws are all form and no content. lf any other laws his arguments cannot be rejected lightly. lt is there­ are to be true, the set of systems to which they fore worthwhile to see what he says, why he says apply must be restricted to something less than the it, and how much truth there might be in what he set of a/1 systems. That is, if a law is to have any says. substance at all, it can be true only for some sys­ ln this review I will first synopsize the book, since tems, and there will be some systems for which it is it is impossible to comment on Berlinski's argu­ not true. ments without first summarizing those arguments. Just in passing, Berlinski quotes von Bertalanffy Since comments on one portion of the book might that "(a certain equation illustrates) a point of in­ also be relevant to other portions of the book as terest for the present consideration, namely the fact weil, I will save almost all my comments until after that certain laws of nature can be arrived at not the synopsis. ln that way the book can be treated only on the basis of experience, but also in a purely as a whole. formal way." Berlinski first demonstrates that the The book contains three chapters, representing purported law of nature is not true of all systems; three basically different approaches to the demoli­ that in order to derive the equation von Bertalanffy tion of Systems Analysis. Each could stand on its made certain assumptions which restricted the range own as an independent essay, but the three together of systems for which the law is valid. Then Berlin­ provide mutual reinforcement for Berlinski's basis ski closes in for the kill, using the viewpoint of a thesis that Systems Analysis is a fraud and is in a philosopher. Of the sentence quoted above, Berlinski state of intellectual bankruptcy. states that it "suggests somehow that being based The approach in the first chapter is to quote pas­ in experience and being formally derivable are alter­ sages from several authors who have written on native but symmetrical procedures whereby a sen­ General Systems Theory (GST), then to demonstrate tence may be counted as a law of nature. This col­ deficiencies in the quoted work. ln each case, lapses the distinction between inductive and deduc­ Berlinski selects quotations which deal with the tive experience." (Emphasis in original.) Had Berlinski been a historian of science rather than a philo­ sopher, he might also have noted that major pro­ • This review is reprinted from Technological Forecasting gress in science came only after "natural philo­ & Social Change., Vol. 14, No. 3, August, 1979, by permission of the editor of the journal. sophers" gave up the idea that pure reason could .. David Berlinski, On Systems Analysis, The MIT Press, deduce laws which simply must be · true in nature, Cambridge, Mass., 1976, $15.00. and instead concentrated on experimental discovery. 16 Review of Berlinski

Berlinski next turns to Ervin Laszlo. He cites a proper form is surely a choice to be made on empiri­ number of quotations dealing with the idea that sys­ cal, not theoretical, grounds. Second, certain variables tems are more than the sum of their parts, that are included in the equation. However, there are systems somehow have the property that they re­ other variables in Forrester's model which could, main more or less unchanged despite changes in with equal plausibility, be included in the equation. their components or in their environment. Laszlo is No rationale is given as to why some are included quoted as describing these properties with such and others excluded. Again, the choice of variables terms as adaptive self-stabilization, cybernetic sta­ to be included is an empirical question. At no point bility, adaptive self-organization, and hierarchies. The does Forrester provide any empirical justification for specific quotations with Berlinski uses are shown to either the form of the equation or the variables in­ be vague and indefinite, and to have little if any cluded in it. Moreover, in Berlinski's view, Urban meaningful content. Regarding a particular explana­ Dynamics is dominated by "assumptions that belong tion by Laszlo of the meaning of hierarchy, Berlinski quintessentially to General Systems Theory." That is, states, "This is not quite a definition of what it is to all the important properlies of a city can be ex­ be a hierarchy, since the definiendum seems to plained by describing the processes and structures figure prominently in the definiens. Still, one sees within the city itself. The environment becomes sole· what Laszlo has in mind and only the purist, I sup­ ly a source of men, or a sink for men leaving the pose, will observe that the formalism introduced (in city. Berlinski is saying that there is more to urban the quotation) is strictly speaking meaningless." dynamics than is captured by Urban Dynamics, and Here again, Berlinski focusses on the purported goal all of GST shares the same faults. of finding general laws true for all systems, and also ln Chapter 2, Berlinski returns to the attack on attacks the use of mathematics for ornamental Forrester. The chapter opens with a brief digression purposes. on systems of differential equations. Berlinski rein· Rapoport is Berlinski's next victim. ln this case forces a point made earlier about the importance of the attack dwells particularly on Rapoport's misuse distinguishing between a model and the theory of of the concept of isomorphism, particularly the that model. He quotes Andrei Monin to the effect attempt to show that if two systems are described that a physico-mathematical model of a system is by equations of the same form, somehow the laws composed not only of a set of dynamic equations of the two systems must be analogous. Berlinski describing the behavior of the system, but also the concludes that Rapoport's attempt to obtain general boundary conditions for the system, and the algo­ laws applicable to all systems is no more success­ rithm for numerical solution of the equations. ful than those of von Bertalanffy and Laszlo. Berlinski next turns to an examination of Systems Berlinski turns next not to a specific individual Dynamics, as presented in a number of works by but to what he refers to as "affable disciplines." Forrester, Meadows and others. While his attack is By this he means the collection of disciplines and on System Dynamics generally, his emphasis is on pseudo-oisciplines which General Systems Theorists the World 111 model and most of his examples are claim either are related to GST or are part of GST. drawn from World Dynamics. Berlinski traces one These affable disciplines include cybernetics (upon particular set of relations in World Dynamics, those which Berlinski heaps a great deal of scorn), mathe­ governing population. Population is of course a matical systems theory, graph theory, set theory, "Ievei" variable in Forrester's terminology, which automata theory, recursive function theory, and infor­ means it can not change value instantaneously, but mation theory. His basic argument here is that those can change value only through the accumulation of affable disciplines which, like information theory, "rates" of growth or decline. "Rate" variables, in have genuine content, are too narrow to support the Forrester's terminology, may change value instanta· broad uses to which GST would put them. On the neously. The form of a rate variable in the World other hand, those "disciplines" which have some 111 model is a normal rate multiplied by one or more breadth, like cybernetics, haYe little if any genuine "multipliers" (whose values may be greater. or lesser content. than 1.0) which represent the effects of various The final section of the first chapter presents factors on the rate of accumulation or decline of a Berlinski's attack on Urban Dynamics. He examines Ievei variable. This particular form for the rate equa· in detail the equations describing arrival of under­ tions i.s the same as that which Berlinski criticized employed males in an urban area. Berlinski's attack in the preceding chapter as being arbitrary and un· can be reduced to two points. First, the form of the tested empirically. Here he makes another point equation is purely arbitrary. lt consists of the pro­ against this form for the equations. The form implies duct of a set of variables which represent degrees that the relationship between the variables is inde· of preference between the city and the external envi­ pendent of time. ln fact, Berlinski states, the multi· ronment from which the underemployed come. How­ pliers are functions of time and should be treated as ever, no rationale is given as to why the product such. The Birth Rate from Material Multiplier and the form is chosen. Other forms of the relationship Material Standard of Living, for instance, are fixed might be equally plausible, and in any case the to have a particular relationship on the basis of the Cybernetics Forum 17

way they are related in 1970. But, Berlinski points However, there is no justification for projecting this out, they might weil have had a different relationship behavior to the industry as an aggregate, and as­ in 1870. To use the 1970 relationship as though it suming that the industry attempts to maximize were true for all time is incorrect. industry profits. in fact it probably doesn't. The Berlinski next turns to the issue of non-linearity. actions of individual firms, in maximizing their indi­ No one disputes that the real world is nonlinear in vidual profits, may even cause industry profits as a its behavior. Mathematically, this means that the whole to suffer. There exist weil known conditions relationship between almost any pair of variables is for the validity of aggregation. World 111 shows no dependent upon the values of other variables. As attempt to demonstrate they are met. Once more, these other variables change, the relationship be­ Berlinski complains about the Iack of necessary em­ tween any two given variables will change. But the pirical justification for the model. very form of System Dynamics models, in which in Chapter 3, Berlinski addresses Mathematical rates are affected by multipliers which are allowed System Theory (MST) (his term). in one sense, this to deviate from a "normal" value of unity, is not very chapter is Berlinski's most general attack on Sys­ nonlinear. Berlinski shows that the general form of tems Analysis in general and GST in particular. Ber­ the System Dynamics equations can be reduced to linski argues that MST grew out of roots in engi­ the form of differential equations with time-varying neering and applied physics. He questions, then, the coefficients. Moreover, given the assumptions stated suitability of engineering models for social and bio­ by Meadows as being behind the World Dynamics logical sciences. Specifically, he argues that sys­ model, Berlinski next shows that it is possible to tems, broadly conceived, do not share any common place bounds on the coefficients for the differential properties, and engineering systems are not like equations. Once these bounds have been imposed, social systems. There are two specific reasons for the behavior of the equations is completely defined. this difference. Most engineering systems are linear The Ievei variables, such as population, must behave or only midly nonlinear (because they are designed in a "grow, overshoot, collapse and repeat" manner. to be that way); political and social systems are Thus the detailed computer simulations are com­ highly nonlinear. Engineering systems in general are pletely unnecessary. The system behavior can be of low dimensionality, in that only a few variables determined by a qualitative analysis of the equations are needed to describe them; social and political describing it. This Ieads him to observe that the systems involve a high degree of dimensionality. distinction between the theory's assumptions and its Both these objections bear on the suitability of MST, cardinal conclusions is very hard to grasp. w~ich deals with Ordinary Differential Equations as Berl inski next addresses the many problems which classically applied to engineering systems, to the beset the passage from theory to model. A great problems of political and social systems. in short, deal of work has been done in the past two decades MST is simply not suited to handle highly nonlinear on the procedures which must be used to estimate systems, nor is it suited to handle systems of many the coefficients of equations from available data, dimensions. lt is therefore inappropriate to try to and the precautions the model-builder must take. use MST for political and social systems. Berlinski's Much of this work is exemplified by the procedures strongest scorn, however, is reserved for those (e .g., of econometrics, although it appears also in other Easton, Kaplan) who use the language of MST (e.g., fields. Berlinski's complaint is that System Dynamics input, output, feedback) in situations where these seems totally innocent of this work, and completely concepts are empty of content. in short, he objects ignores the hard-learned lessons about statistical to those who seriously attempt to apply MST to treatment of data which are today put into practice social and political systems, but he objects even by econometricians, biometricians, etc. more strongly to those who simply discuss the prob­ Berlinski's next major point against World Dyna­ lems of political and social systems in jargon drawn mics, in particular, and one made frequently by other from MST. critics as weil, is that it is a very highly aggre­ in addition to the dimensionality and linearity gated model. The variables in World 111 are not only issues, Berlinski summarizes his objection that MST aggregated, they do not correspond to Observable is not suitable as a model for political and social state variables in the real world (with the exception systems under the following aspects. of population). None, not even population, act as do macroscopic variables such as pressure in physics. 1. Feedback is a meaningful concept in control lnteractions among the variables actually occur at theory. lt refers to measuring the difference between the microscopic Ievei of the entities which make up an actual and a desired state and using this "error the aggregates. For instance, the effect of pollution signal" to move the system toward the desired on population actually occurs as effects of specific state. Many systems theorists describe systems as pollutants on specific individuals. Treating the aggre­ having goal-seeking behavior. They then characterize gate variables may not be justified. Berlinski offers goal-seeking in purely behavioristic terms (often in an example from economics. Individual firms in an terms only of what an outside observer could see), industry act to maximize their individual profits. then equate this behavioristic goal-seeking with feed- 18 Review of Berlinski back. This, according to Berlinski, is a completely ing characteristics 'not possessed by (their) parts inappropriate use of an engineering concept. singly.' But so do heaps since only the whole of a heap of systems theorists, for example, is the whole 2. Much discussion of political and social systems of that heap." (P. 13) involves use of the terms "input" and "output." "Equation (1.4) does have modest usefulness in These terms arose in engineering systems, and are the description of uniform growth-a continuously meaningful concepts in such devices as filters, am­ copulating clutch of rabbits, for example." (P. 7) plifiers or petroleum refineries. Their use in discus­ "The world model itself is expressed as a num­ sion of political, social or economic systems is bered string of finite-difference equations, the nota­ strictly by analogy, and in many cases the analogy tion largely in the cumbersome DYNAMO style, a may not be valid. Moreover, even where the ana!ogy Stalinist mass of tightly bunched capital letters." is valid, use of such terms may not be the most Nevertheless, it must be recognized that this book effective way to analyze or describe a system. ln­ is a one-sided debate. The Systems Analysis Berlin· homogenous differential equations not only exhibit ski attacks have no opportunity for rebuttal. For in­ the outputs of a system (the solutions of the equa­ stance, the debate between Farrester and the econo­ tions), but the inputs to the system are contained metricians is not as one-sided as Berlinski implies. in the inhomogenaus portion of the equations. Thus Farrester has hirnself landed some telling blows on Systems Analysts often discuss inputs and outputs the econometricians. lf the book is viewed simply in Situations where discussion of inhomogenaus dif­ as entertainment, one has to agree it is entertain­ ferential equations (even if many-dimensioned and ing. However, one also has to recognize that if Ber· highly nonlinear) would be more appropriate. I i nski 's targets were present to defend themselves, the book would likely be even more entertaining. 3. Control theory is often used an an analogy by lt is necessary to go beyond looking at the book those trying to apply MST to political and social simply as entertainment. A basic issue is whether systems. But control theory arose in circumstances Berlinski's attack can be considered fair in the demanding regulation of one or more variables with­ sense that he has attacked key points of the writers in a system. There is no proof offered by the Sys­ he has discussed, or that he has analyzed a repre­ tems Analysts that social and political systems are sentative sample of what they wrote. lt is clear that at all similar to control systems. Berlinski has overstated his case in at least a few points, possibly in overzealous pursuit of "uncom­ To summarize the third chapter, Berlinski argues promising negativism." For instance, he objects to that MST, and its full mathematical elegance, applies the behavior of the population variable in World 111, to linear, time-independent systems. Engineering sys­ arguing that the behavior is not "lawlike." One pre· tems are of this type, which explains the power of sumes that he would have been more satisfied had MST in analyzing them. Social, economic and bio­ it shown behavior such as logistic curve. However, logical systems are not of this type, and MST one has to ask, why should it be lawlike? lt could shouldn't be used on them. This is not to say, he behave in that fashion only if it is purely autono­ asserts, that appropriate mathematics cannot be de­ mous, or governed by variables endogenous to itself. veloped for these kinds of systems. lt is to say that lf it is affected by other variables, one would expect the appropriate mathematics hasn't been developed, it to reflect that dependence and deviate from a and that those political scientists, biologists, and "lawlike" pattern. Berlinski also goes too far in ob· economists who talk about their subjects in the jecting to the use of the concept of "feedback" in jargon of MST, without developing the necessary social systems. While he rightly condemns those mathematics, are simply usirig MST as window who use the term in at most a figurative sense dressing. while pretending they are using it in a mathematical Having said all this, in an attempt to present Ber­ sense, he manages to throw out the baby with the linski's arguments in a manner which condenses bathwater. Analysis of economic phenomena, in them while not doing them an injustice, what is one which demand and supply interact to establish Ieveis then to think of the book? How can a practicing sys­ of production, can be analyzed very fruitfully in I tems analyst respond to Berlinski's onslaught? terms of feedback, with no compromise of mathe­ The book can be considered on two Ieveis. The matical rigor. Even Adam Smith, in the Wealth of first Ievei is as pure entertainment. Berlinski's wit is Nations, likened the behavior of economic systems dazzling. With rapier-like thrusts he punctures the to a regulator on a steam engine. He didn't have overinflated prose of the System Analysts. He care­ the word "feedback" in his vocabulary, but at least fully dissects the corpus of Systems Analysis and he had the idea. Mathematical economists since shows that its true mathematical content is near then have made effective use of the concept. zero. A few examples will suffice to illustrate the Despite these errors of over-statement, however, point. Berlinski has been right on target with a great deal " . .. Laszlo argues only that the natural systems, of his criticisms. Clearly, von Bertalanffy did take as opposed to heaps, are 'nonsummative,' possess- some ideas ranging from the trivial to the down- Cybernetics Forum 19

right wrong and dress them up in a Iot ot unneces­ capability ot their theories to produce important and sary mathematics. Clearly, Laszlo's writings include practical results. lt must be said, however, that they some indigestible polysyllabic gibberish. Clearly, did this in all sincerity. They believed that the body Rapoport has written some nonsense about the simi­ ot mathematics contained in such fields as informa­ larities among systems, however detined. Clearly, tion theory, control theory, etc., would soon prove systems analysts, particularly those involved in Gen­ capable ot handling the problems they proposed to eral Systems Theory, have tried to make a great deal deal with, if it were not already capable. They be­ more out ot a mish-mash ot cybernatics, intormation lieved they were on the verge of having mathematics theory, etc., than these disciplines justity. Clearly, it adequate to the task of describing political and is not obvious that System Dynamics is suited to social systems. lf looked at objectively in retrospect, t describing the kinds ot systems present in the their writings seem filled with hubris. ln comparison, social, political and economic world, and its suit­ one thinks ot Newton, who had the presumption to ~ ability is a matter ot empirical veritication, not think he could describe the motions ot the entire theoretical argument. But is it also clear that none solar system. He believed that the calculus was ade­ of these things have any merit or content? Even quate to the task, and moreover he believed it suffi­ granting that Berlinski has pointed out serious and cient to approximate the entire solar system as a possibly important errors, is it the case that he has collection of point masses in an inverse square demolished Systems Analysis completely? This, I be­ force field. The degree ot presumption involved in lieve, is the really crucial issue which must be ad­ such heroic assumptions borders on the rash. And dressed in examining this book. yet, Newton got away with it. Changes in the des­ One thing is certain. This crucial issue cannot be cription of the solar system, tollowing Newton, are addressed satistactorily in even a book the size ot but minor refinements. Even the changes introduced Berlinski's, Iet alone in this review. Nevertheless, I by the Theory of Relativity represent only higher­ will offer some thoughts on the matter. First, it arder corrections. Newton's presumption turned out must be said that it will not be sufficient tor sys­ to be justified. ln the twenty-five years since I first tems analysts to reject any criticism which doesn't studied control theory and read Cybernetics (with offer alternatives. One needn't be a cook to teil that enthusiasm and awe), it has become clear that the the soup is spoiled. One needn't be a tailor to ob­ systems analysts are not going to share Newton's serve that the emperor has no clothes. Thus pointing luck. The mathematics available to them was not out errors is important even when one has no better adequate to the task ot describing social and politi­ alternatives to offer. ln doing this, Berlinski has cer­ cal systems. lt still is not adequate, and shows· no tainly performed a signal service, tor which systems signs ot being adequate in the foreseeable future. analysts should be gratetul, even though the medi­ However, they are still doggedly pursuing the same cine may be hard to swallow. course, heading down the same deadend trail, re­ Secondly, although Berlinski has had a great deal tusing to go back to the beginning and start over of tun pointing out the absurdities and the dense in a new direction. fog present in many attempts to detine the concept Fourth, someone has remarked that the greatest "system," there is some meaning in the term. We do obstacle to solving a problern is the illusion that you experience things which seem to remain themselves already have a solution. Recognition of the true despite changes in their surrounding or even state ot Systems Analysis is the necessary first step changes in their constituents. This reality ot sys­ toward removing the illusion that we have a body tems in the world can't be made to go away simply of tools capable ot analyzing and describing political by pointing out the pompous nonsense in the and social systems. Removal ot that illusion will writings ot systems theorists. Thus there is a need then allow us to start searching for the solutions to be able to address systematic and system-like be­ that, twenty-tive years ago, appeared to be already havior, including both analysis and design. Even it in hand. Since no one within the Systems Analysis there are no general laws ot systems, that doesn't field has taken on the task ot dispelling the illusion mean there can be no tormal means ot dealing with that we had the solutions we needed, we may be systems. thanktul to Berlinski tor undertaking the task. ' Third, we have to face up to the Ievei ot presump­ Fifth, there are problems abroad in the land. tion, bordering on arrogance, which permeated and These problems aren't going to wait until we de­ still permeates much ot Systems Analysis. The early velop the full array of tools we need to handle them. writers on General Systems Theory, tor instance, To pick just one example, the problern of simultan­ produced some trivial and even incorrect work re­ eaus inflation and unemployment isn't going to go garding the similarities among systems considered away or Iet us alone until (at some date in the far as systems, and dressed this work up in the inap­ future) we manage to develop an understanding ot a propriate mathematics with which Berlinski had so complex economy, and develop the mathematics much tun. They made reterences to intormation necessary to express our understanding. ln the theory, to cybernetics and feedback, to automata meantime, it is necessary to search for palliatives theory, as though these proved something about the and partial solutions. One thing which Systems Ana- 20 Review of Berlinski lysts can contribute to that search is the notion that "everything is connected to everything eise," and that therefore "you can't do only one thing." With some humility about how much we really know about the interconnectedness of the world, and some more humility about our ability to state laws governing the behavior of specific systems in the world, we may then begin to give tentative but use­ ful advice regarding actions to be taken in the short term, not to optimize things, but to keep them from getting any worse. And with a view to the long term, we can begin developing mathematics that is suited to the kinds of systems we find in the political, social and economic world. ln summary, then, I believe that Berlinski's attack, although overstated in places, has by and large hit the mark. Taking it seriously may be unpleasant, but doing so will prove to be of significant benefit to Systems Analysis and to Systems Analysts. Some Thoughts On Systems Theory and lts Critics: A Letter To The Editor

Ervin Laszlo United Nations 801 United Nations Plaza '#. New York, N.Y. 10017

Dear Sir, seeing even the most fundamental tenets of systems thinking (when he says, for example, that "the basic I am honered that you asked me to respond to forms of systems theory remain classical positivism criticisms of my writings on systems philosophy and behaviorism" and asserts that "the system, if it contained in Berlinski's On Systems Analysis and is to be a system, must not only have boundaries; Lilienfeld's The Rise of Systems Theory.(1) However, it must also be a closed system ... ").(2) for reasons I shall detail below, I shall not provide lt may-and probably will-be argued that cri­ a systematic response to these authors, but will tiques of one conceptual paradigm can and should offer a few thoughts of my own, which you may be made from the standpoint of another. This dis­ share with your readers if you wish. regards the fact that any paradigm, if it claims to First, the above named books do not as yet merit refer to events 'out there' in reality, represents some a systematic response from the viewpoint of my mode of organization of ideas, sensations, and prob­ work in systems philosophy; they oparate out of an ably intuitive and emotional elements as weil which entirely different intellectual tradition and their con· cannot be tested by reference to a simple inspec­ ceptual spectacles distort the very nature and intent tion of the reality it refers to-since we only 'see' of systems philosophy beyond the point where rea­ such reality through one or another of our concep­ soned argument could be of help. Berlinski-like lda tual constructions. There is no 'final' test for the Hoos who seems to have started this particular tack verity of any basic conceptual mode. Relative tests of systems-criticism-conceives of systems philoso­ do apply, however. These oparate on the meta-para­ phy as some kind of systems analysis and makes digmatic Ievei, where the basic conceptual modes (or his critique on the basis of the subtitle of his book paradigms) are compared for such things as power " ... Concerning the Limitations of Some Mathe­ of explanation, practical guidance-value, elegance and matical Methods in the Social, Political, and Bio­ simplicity, testability, internal consistency, etc. This logical Sciences." But systems philosophy is not a is an interesting and valuable exercise, but it is not method, mathematical or otherwise, nor is it a the one that Lilienfeld performs, although at times science. As its name indicates, it is a philosophy, he appears to be on the verge of it. For in such although one that uses the concepts and theories of exercise the investigator must remain on the meta­ some new trends in contemporary sciences to paradigmatic plane and use conceptual tools which answer perennial philosophical questions. are as neutral as possible. He must not, as Lilien­ Lilienfeld's book is much more aptly conceived. feld, take the concepts and theories of one paradigm As his main title notes, systems theory is indeed on to 'disprove' those of the other. That makes the at­ the rise. lt also has aspects, or at least implica­ tempt biased and reintroduces all the problems of a tions, for a socially and politically active worldview, cross-paradigm dialogue. and may in consequence be analyzed, in part, as an My second reason for not affering a systematic ideology. The attempt in itself merits attention and response to these books is a simpler one. lt is that would merit a systematic response as weil. Unfor­ they, and others of their ilk, perform a most useful tunately, he too, operates out of an entirely different function-which is not diminished by the fact that paradigm, which makes dialogue next to impossible their authors have never intended to perform it. lt is and probably' fruitless (I use 'paradigm' here in the an educational function different from the one they sense of Kuhn, and of Hansen in Patterns of Scien­ envisage: they 'teach' the theories they criticize even tific Discovery, although Hansen did not use that if they intend to 'bury' them. The beneficiaries are term himself). While giving fine overviews of the those readers who have not been thoroughly encul­ ideas of systems thinkers, as soon as he voices his tured into either the 'home' or the 'target' paradigms own views he appears to apply some entirely exter­ of the writers (there are some such people, though nal conceptual scheme which prevents him from they seldom occur beyond the undergraduate Ievei). 22 Thoughts on Systems Theory

Young and as yet relatively virginal intellects are At first, these are somewhat condescending-after suddenly given a choice: you may think this way or all, one does not want to commit the mistake of that. This is particularly significant when the new taking the new theory too seriously, lest one be mis­ way is really new for the reader, and when he or she understood by one's colleagues. The purpose is to has been accustomed, as most high school students expose the sins and crudities of the challenger, and college undergraduates, to hear one kind of making tun of it in an erudite manner. Whether the theory or mode of thinking presented as the gospel critic has truly understood and properly represented truth. the criticized theory is of small moment, since most The educational function is brilliantly performed of his colleagues are certain neither to know nor to by Lilienfeld, who gives detailed overviews of the care. systems theories before he gets down to criticizing lf the challenger continues to gain ground in the them. The same function is of lesser caliber in Ber­ intellectual community (or in society as a whole), linski (as also in Hoos) who keep hammering at their the critics tend to become more expert. Now they targets even if they are largely made of straw. Such read up a little more on the Challenger and try to practices seldom, if ever,· truly convince anyone of understand it before shooting at it. The shots, how­ the falseness or insignificance of a theory who is ever, still come (for a time) from the home base of not already so convinced; the history of philosophy, another mode of thought which, for the critic, grasps and even that of science, testifies to that fact. But the real truth and uses the proper logic. those who are not already convinced of the truth of This is how far we appear to have come today, their pet theory but have a still open mind and some with Berlinski-Hoos representing the penultimate, measure of intellectual curiosity, will note with in­ and Lilienfeld the (sofar) ultimate phase. That we did terest that there exists a kind of theory, or mode of come this far is a remarkable achievement of sys­ thinking, which-though practically without merit in tems thinking. lt has become an innovation that is the eyes of its critics-seems nevertheless to de­ legitimate to criticize, and indeed good business to serve writing entire books about. Few people with do so. Books on it sell, and are used even by one's more than a modicum of intelligence would fail to establishment colleagues. note that there is more to such exercise than target There could-and in this case I believe will-be practice. Just what there is to them they will not more advanced points along the trajectory aptly des­ see very clearly from books such as Berlinski's (and cribed as the 'rise of systems theory.' A logical next Hoos') but will see rather weil from writings such as stage indicating this rise would be the appearance Lilienfeld's. of books and studies which undertake a consistent On reading Lilienfeld the reader gets a fair over­ meta-paradigmatic exploration of the merits and view of the systems field, and such overview, unless faults of the new theory vis-~-vis the older schools, one is already a thoroughly indoctrinated positivist, without using some of the latter as an axiomatic behaviorist, Weberian, existentialist, Marxist, or what­ basis for criticizing it. Subsequent to this we shall ever, cannot help but be impressive. To this writer at witness the publication of an increasing number of least, the overviews are of an entirely different critical essays which already move within the con­ quality than the evaluations. The former show ima­ ceptual universe of the new field. These constitute gination, breadth, penetration, some brilliant insights internal critiques, exploring inconsistencies, correct­ and offer much food for thought, while the latter ing biases, and suggesting further applications and often appear petty and small-minded, saying in ef­ developments. For systems theory this stage is yet fect that 'if anyone thinks any differently from me to come, but its coming is prepared by the previous he is either talking nonsense or just saying what I stages, including those that have just been attained. am saying but through a Iot of useless verbiage.' lt would be fruitless, and indeed counterproductive Publications such as those discussed here may from the viewpoint of the new theory, to engage in represent' points along a curve that marks the tra­ polemies with writers who trace out the above sug­ jectory of a theory innova.tion. When a new theory gested curve in their books. Rather, one .should wel· or conceptual mode of thought appears, it is almost come them for their educational function, and as totally ignored by the adherents of the established signposts that the field continues its rise. paradigm (or paradigms). As it wins adherents here ln conclusion I shall add two further thoughts. and there, more than eyebrows are raised in the Consider first the nature of systems theory. lt is not circles of the establishment: some definions or find­ a specialty falling within the confines of an aca­ ings attributed to the Challenger become known demic discipline, but is a multitude of loosely con· (falsely, as often as not). There are occasional refer­ nected concepts, theories and assumptions which in­ ences to it in books and journals. ln time, it be­ form a whole array of fields from the natural comes acceptable to include a criticism of it in sciences to theology. What they articulate is not a one's writings, and may even become fashionable to simple and immediately testable connection between do so. The day is not far away when it becomes a few isolable 'facts' or data, but an as yet hazy good academic politics to show one's sophistication insight into the nature of growth and evolution in all by producing a full-fledged critique of the challenger. spheres of experience. lt is easy but vacuous to Cybernetics Forum 23

criticize such a mind-boggling paradigm for being low and Project Director at the United Nations Insti­ fuzzy or inconsistent. Think only what it would be tute for Training and Research, is resulting in a were it not fuzzy and inconsistent! Systems theory series of 17 volumes in collaboration with some 98 would eclipse Einstein's wildest dreams for a unified research institutes and teams in all parts of the field theory, for while a unified field theory would world.(4) ln creating, organizing and bringing to frui­ explain the phenomena of the physical universe in tion this international research effort, I have made reference to a single basic field or depth-structure, constant recourse to systems thinking. lndeed, it is systems theory would explain the emergence and not something one can merely leave behind, like a the behavior of physical, chemical, biological, social worn vest. lt is a mode of thinking, a way of look­ as weil as cultural entities and their interaction with­ ing at the world, and even a set of values and ~ in a shared time and space matrix. No wonder we priorities which become deeply embedded in one's are not there yet. BUt great wonder, indeed, that we mind and personality. could already get going in this direction. The critics of systems theory accuse it of being Lastly, Iet me merely register a polite protest to "nothing but" a fashionable jargon. ln reality, sys­ repeated claims that 'systems theories are of no tems terminology has the same function as any use.' Lilienfeld asks, "What Substantive theoretical other intellectual or scientific language: to state or even "applied" sociological, political, economic, ideas, concepts and relationships as concretely, pre­ problems have been resolved by the elaboration of cisely and economically as possible. The ideas, con­ systems theory," and adds, "We await the answer." cepts and relationships are "systemic" in character (ln another place he repeats: "To what new insights even in the absence of the systems terminology. and substantive results does [systems theory] Iead? The reader will find hardly any such terminology in So far there appears to be none.") lf Lilienfeld is my writings after 1977, when I became more con­ still awaiting the answer, could it be that he is cerned with communicating the results of my re­ awaiting it in the wrong place? Could he perhaps search then fighting for systems theory in the spirit wait for it in a bookstore or a library? Or in labora­ of paradigm chauvinism. Yet, as systems theory be­ tories, think-tanks, in corporate planning offices, in comes more known and accepted, use of its governmental policy argans or international organiza­ terminology will not place excessive burden on the tions? He would not have to wait any langer. Take communication of its message. I plan to begin using merely my own experience as example. After finish­ the more concise and precise systems terms in ing lntroduction to Systems Philosophy (1972, re­ coming years, expecting that more people will un­ vised edition 1973) and its popularizing overview The derstand it, and fewer will need convincing that it is Systems View of the World (1972)-books that nothing but an empty shell. ln the meanwhile I, and Lilienfeld has read (at least in part, as he says)-1 my fellow systems thinkers, could ask for few things went on to write A Strategy for the Future (1974), a better than to be read at length-and to be criti­ book that is subtitled, "The systems approach to cized extensively, even by scholars coming from and world order." This, however, Lilienfeld has not read subscribing to different traditions. or even heard of, it seems. Yet it is an application of my systems philosophy to problems of world order, coming out of a series of seminars at Prince­ REFERENCES ton's Center of International Affairs. After this came a Report to the Club of Rome, called Goals for Man­ 1. Berlinski, David, On Systems Analysis. An Essay kind (1977), a short summary of the main findings Concerning the Limitations of Some Mathematical in The Inner Limits of Mankind (1978), and several Methods in the Social, Political and Biological dozen studies and articles in a wide variety of jeur­ Sciences. Cambridge, Mass. MIT Press, 1976; and ... nals and newspapers. Almost without exception, they Robert Lilienfeld, The Rise of Systems Theory. An apply theories and concepts which are elaborated on ldeological Analysis. New York: John Wiley, 1978 . in my work on systems philosophy to international 2. Lilienfeld, pp. 249 and 248. and world affairs. They are not unread (except by the 3. Lilienfeld, pp. 172 and 264. critics of systerris theory): Goals for Mankind, for 4. The UNITAR-CEESTEM Library on the New Inter­ example, has been published in New York (by national Economic Order. Published by Pergarnon Outton and later by the New American Library), Press, New York and Oxford, in conjunction with London, Tokyo, Milano, Amsterdam, Mexico, and the United Nations Institute for Training and Re­ Helsinki, with condensed versions in Moscow, War­ search, New York, and the Centro de Estudios saw and Budapest. My current work, as Special Fel- Economicos y Sociales del Tercer Munda, Mexico. Linear and Elitist: How Our Critics See Us

Stuart A. Umpleby George Washington University Washington, D.C.

A review of David Berlinski, On Systems Analysis, For theory development empirical research is there­ MIT Press, 1978, and Robert Lilienfeld, The Rise of fore directed at the nervous system. Of course Systems Theory: An ldeological Analysis, John Wiley models can be constructed of any reference system. & Sons, 1978. But in cybernetics this work is usually considered to lie in the area of application rather than theory Cybernetics is a curious discipline. For those who development. practice it, it is in danger of dying out. For at least Anatol Rapoport's emphasis on searching for some people outside the field, it threatens to take mathematical isomorphisms guided the articles in­ over the world. lf nothing eise, the recent books by cluded in the General Systems Yearbook for twenty David Berlinski and Robert Lilienfeld prove that years. Although these articles contain many thought­ someone out there is following our work. But if provoking analogies, I know of no widely referred to these books are an accurate reflection, we are not scientific success that has resulted from this strate­ making ourselves very clear, and indeed we may gy. lndeed the strategy has led to one important have something to learn. Berlinski's book has re­ misunderstanding-the confusion over the relation­ ceived more attention, but Lilienfeld's is the more ship between information and entropy and the notion useful. 8oth books fall short of the ideal-a read­ that as thermodynamic entropy increases, informa­ able, carefully analytic, constructive contribution to tion or organization must necessarily increase as the field. On Systems Analysis is notable for sarcasm weil. Numerous people, including Laszlo, have been more strident than I have ever seen before in an misled by this notion.(1) However, the relationship academic treatise. The Rise of Systems Theory uses does not hold. lf organization on Earth increases as so many long quotations set in type so similar to the sun expends itself, then why do species become the main text that one spends a fair amount of time extinct and civilizations rise and fall? lndustrialized figuring out whether the ideas being expressed are countries consume more energy per capita than de­ Lilienfeld's or someone else's. veloping countries, but some nations at the same Ievei of development are more wasteful than others. A corporation cannot guarantee high profits by con­ Shortcomings: Under the Aspect of suming more energy. Information and thermodynamic Berlinski entropy are different concepts. Only the equations are similar. 1 do agree with some of Berlinski's criticisms. Berlinski's statements that general systems theory There are fundamental difficulties with Von is lacking in content deserve some reply. The view Bertalanffy's explanation of his empirical approach. that a general theory of systems would be trivial or Berlinski notes that if one Iooks at a variety of sys­ would Iack a foundation has been expressed more tems to discover their common properties, what is than once.(2) Several replies have been made to this common to all systems are the logical truths. ls criticism. Bateson has noted that cybernetics is a systems theory then a branch of mathematics? I do science of form and pattern rather than a science of not believe that this question can be resolved, as substance.(3) Ashby deals with the issue in the first the Binghamton group is inclined to do, by saying chapter of An lntroduction to Cybernetics. that general systems theory lies between philosophy and mathematics on the one hand and political Cybernetics treats, not things but ways of science, sociology, and engineering on the other behaving. lt does not ask, "what is this hand. Such a formulation is not an explanation but thing?" but "what does it do?" Thus it is very a compromise. lt may calm a dispute but has little interested in such a statement as "this vari­ persuasive power. ble is undergoing a simple harmonic oscil­ lation," and is · much less concerned with Cybernetics has dealt with this issue by focusing whether the variable is the position of a point on cognition. Since all systems are described by on a wheel or a potential in an electric cir­ observers, an understanding of the observer will teil cuit. lt is thus essentially functional and be­ us something about all systems we shall encounter. haviouristic. Cybernetics Forum 25

Cybernetics started by being closely associ­ should be included. But merely calling attention to ated in many ways with physics, but it de­ dimensional complexity will not help someone who pends in no essential way on the laws of must make a decision. physics or on the properties of matter. Cyber­ Furthermore, Conant and Ashby have constructed netics deals with all forms of behaviour in so far as they are regular, or determinate, or a proof that every regulator of a system must con­ reproducible. The materiality is irrelevant . .. tain a model of that system.(B) Before people tried The truths of cybernetics are not conditional to construct mathematical models of systems, they on their being derived from some other were using mental models. For example the domino branch of science. Cybernetics has its own theory was a conception used to justify United foundations.(4) States intervention in Viet Nam. Few people worried about whether the relationships in the domino theory Berlinski apparently found Ashby's arguments unper­ were "linear" or not. The theory was accepted or suasive. lt would be useful to know why. rejected on other grounds. Consider a secend ex­ Another way of dealing with the criticism that sys­ ample. The Club of Rome models used as key vari­ tems theory is too general is to go back to first ables population, natural resources, food, and pollu­ principles. lt is frequently said that scientific con­ tion. The principal mental model used prior to that cepts are neither true nor false, only more or less time dealt with nations, alliances, and the military useful. Theories are judged by their usefulness in balance of power. Berlinski does not examine the explaining and predicting phenomena. Furthermore, "mathematical properties" of the earlier model, nor an objective of science is to explain the largest does he argue that the natural resources model is number of phenomena with the smallest number of less appropriate than the nation-state model. lndeed propositions. From this line of argument we can he seems to be unaware that two models are com­ conclude that if a proposition is useful, it cannot be peting for public acceptance. Berlinski brings to too general. lf it explains a !arger number of mind a ball player who has mastered the locker phenomena, it is more useful even though it is also room repartee but has forgotten, if he ever knew, more general. Consider the concept of mass. An in­ what happens out on the playing field. sect has mass, and an elephant has mass. An atom has mass, and a star has mass. Mass is therefore a very general concept, encompassing both living and nonliving things of · vastly different size. Does ldeology in the Eyes of Lilienfeld this great generality render the concept meaning­ Whereas Berlinski's arguments are purely aca­ less? Quite the contrary. And likewise the concepts demic, Lilienfeld is principally concerned with the of variety and the tendency to equilibrium which lie social consequences of cybernetics. Lilienfeld views at the heart of systems theory are useful even systems theory as the ideology of the welfare state. though they are very general. By "welfare state" he means a society run by a When dealing with mathematical models, Berlinski centralized bureaucracy, a society in which the key does not ask why the model was constructed in the issues are decided by a technical elite rather than first place, or what model was being used prior to through a democratic process. He maintains that the the mathematical model. Nor does he suggest how models constructed by systems theorists assume a better model might be constructed. Consequently, and help to create such a decision-making although the reader can agree with many of Berlin­ apparatus. ski's technical points, the book provides little that is One can make many small criticisms of Lilienfeld's helpful. case. The summaries of the theories of Von Berlinski's criticisms of mathematical models re­ Bertalanffy, Ashby, Wiener, Shannon, Turing and flect a Iack of understanding of several ideas basic others shed no new light on their work. Frequently to cybernetics. He notes that social systems are he selectes quotations which help to make his case many dimensional and implies that models are not regardless of whether they are a good reflection of useful if they include only a Iew variables.(5) Berlin­ a person's overall point of view. This is standard ski does not seem to understand that any act of debating practice, but I am disturbed by his tend­ regulation requires selecting a set of variables to pay ency to attribute anti-democratic motives to people attention to.(6) Even when dealing with inanimate ob­ whose goals were quite the reverse. Cyberneticians jects, the number of variables one could Iook at is may be naive about social systems, and some are virtually unlimited. The trick lies in deciding which ambitious. But I know of none who would advocate are the important variables to take note of. As "the welfare state" that Lilienfeld describes. How­ Ashby put it, "Man adapts by conquering the reduci­ ever, these are merely qualifying remarks. There is ble, the irreducible is impregnable."(7) ln any particu­ something important in what Lilienfeld is saying, and lar situation a useful criticism would be to suggest I hope that we shall not be so defensive or proud a different set of variables and argue why they are that we miss it. more important than the set that was chosen. Or I would make the case as follows. ln a social sys­ one could argue that a Iew additional variables tem there are basically two ways to make a deci- 26 Linear and Elitist sion. One can examine the system of interest, study cision-making has advantages in certain circum­ its past, construct a model, project the future, stances, can an expanded cybernetics teils us when decide what variable or variables one wants to try to the second method is preferable or what combina­ maximize, and then select the best alternative. Or tion of the two methods to use in a particular Situ­ one can consult the people concerned, Iet everyone ation? Or is this not a theoretical question but have their say, and then make the most political rather a decision for the group? decision-the one that pleases the most people or Second order cybernetics, as it was developed by the most powerful people. ln a democratic society people associated with the Biological Computer the second alternative has an important advantage­ Labaratory (BCL) during the 1970's, has laid a firm people learn best by making their own mistakes. foundation for this new arena of cybernetics in­ Societies learn in much the same way that individu­ quiry.(9) By investigating the nature of the observer, als learn. lf an elite makes decisions that the the BCL group developed a theory of autonomaus general population does not understand, political and systems.(10) A next step will be a theory of how ethical learning will not take place. groups of autonomaus systems make joint decisions Cybernetics has tended to emphasize the first and act together. decision-making method. This may have been a necessary conceptual stage for the field to go through. Most of the theoretical work in the first REFERENCES few decades dealt with cognition in a single indi­ vidual. The widespread assumption was that the 1. David Berlinski, On Systems Analysis, MIT Press, same principles would hold at the societal Ievei. 1978, p. 17. Some people, such as Ausseil Ackoff, have argued 2. V.N. Sadovsky, "General Systems Theory: lts that there is an important difference between a brain Tasks and Methods of Construction." General and a society. lndividuals have their own goals but Systems, 17, 1972, p. 173. neurons do not. Others, such as Stafford Beer, ac­ 3. Gregory Bateson, Steps to an Ecology of Mind, knowledge the difference but claim that it is not Ballantine, 1974. theoretically significant. Lilienfeld's book has helped 4. W. Ross Ashby, An lntroduction to Cybernetics, me to arrive at the conclusion that the difference is Chapman and Hall, 1956, pp. 1-6. quite important, not only theoretically but even more 5. Berlinski, Op. Cit., pp. 129-131. in how we act. The emphasis in cybernetics on the 6. See the discussion in Chapter 2 of W. Ross first style of decision-making-building a model and Ashby, Design for a Brain, Chapman and Hall, then using it to select an alternative-is at the root 1960. of the feeling that cybernetics is at least potentially 7. W. Ross Ashby, Class Handout, 1964. anti-democratic. 8. Roger C. Conant and W. Ross Ashby, "Every A science of the second decision-making method Good Regulator of a System Must Be a Model would be a science of how groups make decisions­ of That System," International Journal of Sys­ not an individual acting for a group but a group of tems Science, Vol. 1, No. 2, 1970, pp. 89-97. autonomous, opinionated people. Such a science 9. Stuart A. Umpleby, "Second Order Cybernetics would have ~o deal with how a consensus is pro­ and the Design of Large Scale Social Experi­ duced, how motivation is increased or maintained, ments," Proceedings of the Annual Meeting of and why people work tagether despite disagree­ the Society for General Systems Research, ments. As cyberneticians move into this new domain Boston, February 1976, pp. 69-75. of inquiry, there is much that we can learn from 10. Francisco Varela, Principles of Biological Auto­ social scientists. Since the second method of de- nomy, North Holland, 1979. ----

Cybernetics: Search For A Paradigm

N. A. Coulter, Jr. Biomedical Engineering and Mathematics Curriculum Department of Surgery-University of North Carolina ... Chapel Hili, North Carolina 27514

"The concern for man and his destiny must ed as a goal signal representing a future state, and always be the chief interest of all technical whose activity, guided by feedback, proceeds unti I effort. Never forget it among your diagrams the output matches the input-the goal is achieved. and equations."-Aibert Einstein On this basis the apparent purposiveness of physi­ ls there a cns1s in cybernetics? lf so, I am sur­ ological phenomena can in principle be scientifically prised to learn it, for it has seemed to me that the explained. Even the differences between biological ideas and techniques of this field have had as and engineering servomechanisms are not beyond powerful and pervasive an influence as any develop­ the cybernetic modes of thinking to explain; thus, ment in recent scientific history. Perhaps the crisis the fact that biological systems have the capability has emerged because cybernetics has succeeded to generate their own goals requires only a relatively too weil! straighttorward generalization of cybernetic control Looking back at Wiener's book (1948), one finds theory (Coulter, 1976). three main themes depicted there as central to his The chief problems cybernetic control theory has definition of cybernetics as the science of "commu­ encountered arise from the complexity of physio­ nication and control in the animal and in the logical processes and the difficulty of isolating machine." These are: (1) control systems; (2) informa­ physiological control systems from the matrix of tion theory; and (3) the brain-computer analogy. Each interactions in which they are embedded. This prob­ of these has had a profound influence on the fields lern is formidable but it is not insurmountable. lt is affected; each has encountered difficulties; each has a challenge for future research. aroused controversy. Let us briefly survey the ad­ The application of information theory has been vances and problems each of these has had. less successful. True, it has been a source of some Prior to cybernetics, there was virtually no theory useful concepts in neurophysiology and genetics; to account for the behavior of biological control but efforts to apply the theory to biological pheno­ systems. True, the concept of homeostasis had been mena have not been very fruitful to date. Leibovic advanced to describe the relative constancy of many (1969) points out that the statistical concept of infor­ physiological quantities (like body temperature and mation is "too restrictive" from a biological Stand­ the pH of body fluids); but this was little more than point. When information is defined as "negative en­ an empirical generalization. Cybernetic control theory tropy," we are given a precise mathematical expres­ provided a powerful and rapidly evolving source of sion of limited biological utility. As a result, the ideas, techniques of analysis, and mathematical general practice has been to regard information as models which soon produced a flourishing and still synonymaus with "pattern," and in this sense it has growing literature. There is no reason to doubt that been useful, especially in focussing attention on an research in this area will continue to be fruitful. aspect of neurophysiological phenomena that is The concept of the servomechanism also effective­ sömetimes disregarded. Information, in this sense, is ly resolved a problern in biology that had previously as real as mass and energy, and as important in been difficult to explain: the apparent purposiveness understanding biological phenomena. The song of a of many physiological phenomena. Biologists gener­ radio singer exists as a pattern imposed on sound ally rejected the Aristotelian concept of a final waves, then as a pattern of electric voltages in the cause, since it was difficult to see how a future transmitter, then as a modulation of an electromag­ state could affect the present course of events. netic carrier, etc. The pattern is invariant under "Teleological thinking" was regarded as "unscien­ transformation from carrier to carrier; and it is re­ tific." Yet many physiological processes seem to be lated, by some as yet not understood coding pro­ goal-seeking in character; and purposes guide our cess, to nerve impulse patterns in the brains of the conscious lives. A servo-mechanism is basically a singer and, ultimately, the people who hear the teleological mechanism, whose input may be regard- song. Physical laws govern the transformations of 28 Search for a Paradigm

matter and energy involved in all these processes; Specialization has been and continues to be a but they do not explain the song. Yet the song can vital necessity in the advance of science. But it has be described mathematically and there is no need to the inevitable side-effect of creating more and more evoke some unknowable vital force to account for special fields, each separated from its neighbors by the coding and decoding processes of brains. walls of ignorance that constitute barriers to under­ The brain-computer analogy has probably been the standing-the modern equivalent of the biblical most controversial of the three main cybernetic Tower of Babel. This problern has long been recog­ themes. Wiener considered the computer to be "al­ nized but a fully effective solution has yet to be most an ideal model of the problems arising in the found. nervous system. The all-or-none character of the dis­ One approach is to foster the development of charge of the neurons is precisely analogaus to the interdisciplinary educational programs and research single choice made in determini.ng a digit on the institutes which focus on the interfaces between binary scale . . . The synapse is nothing but a related or relatable special fields. Such programs mechanism for determining whether a certain com­ can and do help. But they themselves encounter bination of output from other selected elements will problems, of which two in particular seem important. or will not act as an adequate stimulus for the The first is that it is difficult enough to master discharge of the next element and must have its one special field, Iet along two (or more). One runs precise analogue in the computing machine." This the risk of "falling between two stools." This can picture of the brain and its working is highly over­ result in mediocre work. lt must, I think, be acknowl­ simplified and, to neurophysiologists, naive. Never­ edgP.d that this has happened sometimes in theless, the analogy is highly suggestive, and, to cybernetics. many, far better than earlier analogies (like tele­ But it must also be noted that specialists in a phone switchboard, energy transformer, the "hydrau­ given field tend to judge work related to that field lic" model of Freud, etc.). And it has led to the in terms of their own backgrounds, knowledge, and emergence of an exciting, though itself controversial interests. Thus, a physiologist may judge a cyber­ (Dreyfuss, 1972, Weisenbaum, 1976), area of com­ netic paper purely from a physiological perspective, puter science-the field of artificial intelligence ignoring or belittling its non-physiological content. (Boden, 1977, gives an excellent overview of this On the other hand, a theoretical physicist may ig­ field.) nore or belittle its physiological relevance, and find These basis ideas, and others, of cybernetics have little new or interesting from the Standpoint of theo­ thus had a pervasive influence; so the "crisis" evi­ retical physics. As Small (1980) points out, "lt is a dently has other sources than the scientific content sad reality, but a reality all the same, that when­ involved. What are these sources, and what, if any­ ever one does something new and different, one has thing, should be done to resolve the crisis? to expect to be shot down." lt must be acknowl­ edged that this, too, has happened. A partial solution to this problern is to design The General Problem of special new courses and curricula which are selec­ tive in content and methodology. There are many lnterdisciplinary Programs details of biological science which are unnecessary Cybernetics (and General Systems Theory as weil) for a cyberneticist to know. Similarly, many aspects originated basically as an interdisciplinary endeavor. of engineering control theory have little biological Wiener was convinced that "the most fruitful areas relevance, and examples chosen from industrial and for the growth of the sciences were those which had military applications do not illustrate how these been neglected as a no-man's land between the vari­ methods can be used to elucidate biological pheno­ ous established fields." He was concerned by the mena. Meanwhile, as the cybernetic Iiterature grows, fact that "science has been increasingly the task of such courses and curricula may progressively be specialists, in fields which show a tendency to grow weaned from dependence on their parent sources. progressively narrower ... These specialized fields Such a selective approach, reinforced by a critical are c9ntinually growing and invading new territory. emphasis on quality, can filter out mediocrity. The result is ... an inextricable tangle of explora­ The second problern is socioeconomic in nature. tion, nomenclature, and laws. There are fields of Scientific research requires both institutional homes scientific work ... which have been explored from for scientists and financ-ial support for their work. the di_fferent sides of pure mathematics, statistics, The established departments of academia tend electrical engineering, and neurophysiology; in which sometimes to be jealous of their territory and zeal­ every single notion receives a separate name from ous in defense of what they perceive as their "intel­ each group, and in which important work has been lectual property rights." Where is a cyberneticist to triplicated and quadruplicated; while still other find an academic home, with reasonable freedom to important work is delayed by the unavailability in conduct research in cybernetics? How many cyber­ one field of results that may have already become neticists obtain funding without prostituting them­ classical in the next field." selves to military and commercial interests-especi- Cybernetics Forum 29

ally when more appropriate granting agencies are A third line of action is to clarify and develop a dominated by the established disciplines? The pro­ cybernetic paradigm. Kuhn (1970) characterizes scien­ gress of science in general, and in particular areas, tific paradigms as achievements that "serve for a is not determined by scientists alone; it depends time implicitly to define the legitimate problems and on the socioeconomic environment as weil. methods for succeeding generations of practitioners. Unlike the first problem, this second problern does They were able to do so because they shared two not have a clearly definable solution, or one that can essential characteristics. Their achievement was suf­ be resolved primarily by cyberneticists. lt depends ficiently unprecedented to attract an enduring group on the politics of academia and the fickle winds of adherents away from competing modes of scien­ of scientific funding. This does not mean, however, tific activity. Simultaneously, it was sufficiently that nothing can be done. There are several lines of open-ended to leave all sorts of problems for the re­ action that can be and are being pursued. defined group of practitioners to resolve." He dis­ The first is the establishment and development of cusses various examples, such as Copernican scientific organizations dedicated to cybernetics, and astronomy, Newtonian physics, Lavoisier's Chemis­ concomitantly the publication of scientific journals try, relativity, and quantum mechanics. in this field. Although lang delayed, this fortunately Does cybernetics have a paradigm? I believe that is now being done; and although the effects are not it does but that is needs clarification and develop­ spectacular, they are bound to secure the survival ment. Wiener's book was the catalyst, but it is far and flourishing of cybernetics in the lang run. too sketchy to provide a paradigm that is sufficiently A second line of action, suggested by our presi­ powerful to solve the problems it addresses or to dent, is to "provide leadership ... for the humaniza­ fulfill the expectations it arouses. But it does define tion of systems." This involves investing a major ef­ a field of research-a general approach to communi­ fort in addressing the problems of society that are cation, control, and "computation" whose elements the source of so much human misery, and whose are clearly related and whose range of application is complexity has thus far defied the best efforts of quite broad. And there has been considerable pro­ the established sciences traditionally concerned with gress since Wiener's book was published. their elucidation., One of these, for example, is the lf and when a cybernetic paradigm clearly problern of war, and the steadily increasing threat of emerges, it will inevitably attract an enduring group nuclear holocaust. lt does no good to rail at the of adherents and provide an enduring basis for de­ politicians; they are prisoners of their roles, and all fining problems that challenge young minds to use too often simply the end-products of systems which the tools it provides to solve them. Unfortunately, reward those motivated by the Iust for power. What a paradigm cannot be designed to order; it can only needs to be done is to analyze these complex sys­ emerge as a result-indeed, as a byproduct-of an tems in cybernetic terms, and to find ways to cata­ ongoing process of scientific endeavor. lyze and aceeierate a cultural evolution which will ultimately make war as unthinkable as cannibalism, human sacrifice and slavery have become, almost everywhere. Effective measures to achieve this are Some Fundamental Problems unlikely to be found by directly addressing the What, then, are the failures of substance and issues and crises that recurrently seize publ ic atten­ method that may have led to the crisis in cyber­ tion. These are symptoms of cybernetic disorders, netics and that block the full emergence of a cyber­ unstable states of extremely complex systems netic paradigm? whose parameters have shifted in subtle and un­ One such failure, I believe, resulted from the prac­ known ways. Rather, it is possible that cybernetic tice of drawing attention to analogies between bio­ .. analyses can reveal the hidden relationships among logical phenomena and their engineering Counter­ these parameters that govern the global behavior of parts. The assumption was then made that the engi­ these systems, and suggest changes we and others neering theory could fully explain the biologlcal can implement which will amelierate that behavior. phenomenon. The similarity between engineering There is no assurance that cyberneticists can do communication networks and those of the central this, and we should make no promises. But we can nervaus system is an example. lt was assumed that at least try. information theory could be directly applied to the lf we are able to make effective contributions to analysis of neural signals. The results have not been the resolution of these problems of society, then a impressive. favorable socioeconomic climate for cybernetics will Since the biologist rarely understood the mathe­ naturally emerge. Thus, efforts to facilitate the hu­ matics of information theory, he had to rely on the manization of social systems are clearly in our own engineer or physical scientist to select the appro­ self-interest. They are also very much in tune with priate mathematics. The engineer, on the other hand, the social outlook of Norbert Wiener, whose advo­ rarely had an adequate understanding of the com­ cacy of organizational changes promoting "the plexity of the biological phenomenon. As McCulloch human use of human beings" is weil known. once aptly noted, "One has to have a reasonable 30 Search for a Pardigm

knowledge of both engineering and biology in one sulted in severe criticisms which went to the oppo­ head. and there is no use having in one room what site extreme. in my opinion, this still remains one of should be in one head." the most potentially usetul parts of the cybernetic But this Iailure need not be occasion for despair. paradigm-not so much tor the results that it may The analogy remains. A model or theory is no good ultimately achieve, as it is tor two by-products of if it is not falsifiable: and such failures can Iead to the endeavor. First, programming a computer to sim­ major advances if appropriately used. in this case, it ulate neural (or mental) activity imposes a tough dis­ led to a better appreciation of the differences be­ cipline, a requirement for clear thinking, that has all tween neural and engineering communication chan­ too often been lacking in the psychological and nels. in an engineering system, a design is required social sciences. And second, the failures and inade­ to transmit a signal that faithtully represents the quacies of artificial intelligence research have ad­ message originated at the source, and to do this in vanced our understanding ot, and appreciation tor, the most efficient manner. The entropy of the the complexity and sophistication of "neuro­ source, a quantity based on the ensemble ot possi­ computers." ble messages, is signiticant tor this task. in a neural The camparalive study of brains and computers communication system, however, what is signiticant is as scientifically Iegitimale an enterprise as the ac­ is the relevance of the patterns characterizing the cepted practice of studying the brains ot lower ani­ source to the behavior ot the organism concerned. mals for clues to the pertormance of the human The classical paper ot Lettvin, Maturana, McCulloch, nervaus system. Ultimately, brains and computers and Pitts-"What the Frog's Eye Teils the Frog's may be seen as special cases ot a still more general Brain" (1959)-illustrates the difference. A taithtul kinc ot system, the theory ot which will explain the representation ot the visual tield ot the trog is not behavior ot both. The development ot such a theory required. What is important is the detection ot small, may yet emerge from artiticial intelligence research. moving patterns like those cast by a bug on a trog retina-or the global dimming ot light intensity caused by the shadow ot a !arge predator that eats trogs. Appreciation of these tindings then enabled Conclusion Moreno-Diaz (1968) to develop a model of the "bug The toregoing is admittedly inadequate and incom­ detector network" ot the trog retina which account­ plete, and reflects the interests and biases ot the ed tor the experimental data. author. I have taken Wiener's book as a basis for Another example is the analogy between an engi­ detining cybernetics and describing its content, neering servomechanism and the neuromuscular net­ which untairly ignores the contributions ot others. work controlling movement ot a limb. A preliminary And even here, I have not considered other themes application ot this analogy would Iead one to iden­ Wiener discusses, such as cybernetics and psycho­ tity the motor born cells ot the spinal cord as the pathology. I hope the reader will pardon these tlaws comparator, the outputs trom these cells to the mus­ and remedy these deticiencies. cle as the controller, the muscle itselt as the con­ Nevertheless, the survey leaves me more confident trolled system, and the signals trom the muscle than ever that cybernetics will survive the "crisis" spindie receptors as the teedback. in actuality, the and prosper, as one of the more signiticant scien­ system is tar more complex, and the simples! tific developments ot the twentieth century. scheme to characterize it includes not just a given muscle, but its synergists and antagonists as weil, which tagether produce partial rotation about a joint. in addition, some ot the motoneurons send outputs not to the muscle itselt, but to the muscle spindie REFERENCES receptors, changing the "bias" ot this "transducer." 1. Wiener, N. Cybernetics: Or Cantrot and Commu­ The system includes nonlinear components (notably nication in the Anima/ and the Machine. New the muscle itselt), analog-ta-digital convertors, time York: John Wiley and Sons, 1948. delays, and other complexities. in this example, how­ 2. Coulter, Jr., N. A. "The Self-Determinism of Teleo­ ever, the cybernetic paradigm has stimulated a Iot ot genic Systems," J. Cybernetic 5:9-20, 1976. research and has considerably advanced our under­ 3. Leibovic, K. N. ln Leibovic, K. N., editor. Infor­ standings ot the system; and the rapid development mation Processing in the Nervaus System. New of engineering control theory has provided powertul York - Heidelberg-Berlin: Springer-Verlag, 1969, tools tor analysis. pp. 335-336. The computer-brain analogy has been intermediate 4. Dreytuss, H. L. What Computers Can't Do: A in usetulness, and has evoked the most controversy. Critique Of Artificial Reason. New York: Harper Unfortunately, some early workers in artiticial intelli­ and Row, 1972. gence were excessively optimistic in their predic­ 5. Weisenbaum, J. Computer Power and Human tions of what it could do, and overly simplistic in Reason: From Judgment to Calculation. San claims made tor some ot its results. This has re- Francisco: Freeman, 1976. Cybernetics Forum 31

6. Boden, M. A. Artificial lntelligence and Natural Man. New York: Basic Bocks, lnc. 1977. 7. Small, M. G. "What's Wrang With Systems? An lnquiry lnto The Negative Criticisms Directed at Systems Analysis and Systems Theory." Proceed­ ings of the 24th Annual North American Meeting, Society for General Systems Research, 384-392, 1980. 8. Kuhn, T. S. The Structure of Scientific Revolu­ tions, Chicago: University of Chicago Press, 1970, Secend Edition. 9. Lettvin, J . Y. , Maturana, H. R., McCulloch, W. S., and Pitts, W. H. "What the Frog's Eye Teils the Frog's Brain," Proceedings of the IRE, 47:1940- 1959, 1959. 10. Moreno-Diaz, R. "An Analytical Model of the 'Bug Detector' Ganglion Cell in the Frog's Retina," ln Ostreicher, H. L. and Moore, D. L., editor, Cyber­ netic Problems in Bionics. New York - London Paris: Gordon and Breach, 1968, pp. 481-492. About the Authors

NORMAN A. COULTER, JR. JOSEPH P. MARTINO Norman A. Coulter, Jr. is Dr. Joseph P. Martino is a Professor and Chairman of member of the Technology Biomedical Engineering and Forecasting Group at the Mathematics Curriculum at University of Dayton Research the University of North Caro­ Institute. His activities there lina. He received his B.S. include the preparation of from Virginia Polytechnic ln· forecasts of technological stitute in 1941 and his M.D. change in specific areas or from Harvard Medical School fields, and the assessment in 1950. Dr. Coulter was post­ of the consequences of doctural fellow in biophysics those changes. Recent pro­ at Johns Hopkins from 1950 to 1952. He was Assis­ jects have included forecasts of satellite communica­ tant to the Associate Professor of Physiology and tions for the National Aeronautics and Space Admin­ Biophysics at Ohio State University, 1952-1965. His istration, and the development of a model for the current interests are teleogenic system and neural diffusion of innovations in industry, for the National networks, and synergetics. Science Foundation. Prior to joining the Research Institute, Dr. Martino served 22 years as an Air Force Officer, retiring in the grade of Colonel in 1975. He received the AB in Physics from Miami University, the MS in Electrical Engineering from Purdue, and the PhD in mathematics from Ohio State. He is the author of numerous papers and re­ ports on technological forecasting. He has written ROBERT LILIENFELD one book, Technological Forecasting For Decision­ Robert lilienfeld, Associate making, and is an Associate Editor of the journal Professor, Sociology, at the Techno/ogical Forecasting & Social Change. City College of the City Uni­ versity of New York. Ph.D., Graduate Faculty of the New STUART A. UMPLEBY School for Social Research, Stuart A. Umpleby is an as­ in Sociology and Philosophy, sociate professor of Manage­ 1975. At City College since ment Science at George 1969; Previous positions in­ Washington University. He clude posts as Research Di­ teaches in the program on rector, Graduate School of General Management Sys­ Social W9rk, New York University, and at the Gradu­ tems and Organizational Cy­ ate School of Public Administration, New York Uni­ bernetics (GEMSOC). He re­ versity, and as Project Director, Population Health ceived degrees in mechani­ Survey, New York City. ln addition to The Rise of cal engineering, political sci­ Systems Theory, has published, as co-author with ence, and communications Joseph Bensman, Craft and Consciousness (Wiley, from the University of lllinois in Urbana-Champaign. 1973), a study of the influence of occupation on While at the University of lllinois he was connected world views and habits of thought, and also Be­ with the Biological Computer Laboratory and the tween Public and Private: The Lost Boundaries of Computer-based Education Research Laboratory (the the Seit (Free Press, 1979), a study of modern ambi­ PLATO system). For two years he has been the valences towards both our public selves and our pri­ moderator of a computer conference among about vate lives. Present research interests are centered on fifty cyberneticians and systems theorists in the intellectuals and intellectual movements. Has also United States, Canada, and Europe. He has recently published many articles in The Nation, Diogenes, completed a system dynamics model of national Philosophy and Phenomenological Research, Social development for the Agency for International Research, and other journals. Development. Cybernetics Forum 33

ARIE ARIELY ERVIN LASZLO Arie Ariely is currently a Dr. Ervin Laszlo is currently Doctoral candidate at George directing a world-wide re­ Wash ington University, in search network at the United general management sys­ Nations. He is also serving tems and organizational cy­ on the boards of a number bernetics, with interest in of social science and world Cybernetics, Philosophy, affairs journals, extensively Mathematics and neuro- writing and editing, and main­ sciences. taining a world-wide lecture ...., For the last twelve years ' _ _. "\ schedule. He was awarded Arie was involved in the the Doctorat es-Lettres et management, research and development, and the im­ Seiences Humanines by the Sorbonne in 1970. The plementation of computer-based solutions. Fi rst with first of his several books with a philosophical orien­ the Chemical Bank of N.Y. and later with Martin­ tation was published in 1962, and by 1966 he was Marrieta data systems and Tadiranllsrael. His pro­ offered a fellowship at Yale University. Subsequently fessional interests are in: Operating system and com­ he was professor at lndiana University, the University munication systems architecture, computer systems of Akron, and the State University of New York. performance evaluation and software reliability. During this time, Laszlo contributed to and served as editor of various philosphy journals, wrote and edited several books, and lectured widely in America and Europe. ln 1972 he spent a semester lecturing at Princeton University, where he attempted to inte­ grate international affairs and world development with biological and social evolution through systems theory. His work at Princeton led to work with the Club of Rome and a fellowship with the United Na­ tions Institute for Training and Research. Statement of Editorial Policy

The ASC CYBERNETICS FORUM is an internationally distributed quarterly publication of the American Society of Cybernetics. lt is published to promote the understanding and advancement of cybernetics. lt is recognized that cybernetics covers a very broad spectrum, ranging from formalized theory through experimental and technological development to practical applications. Thus the boundaries of acceptable subject matter are intentionally not sharply delineated. Rather it is hoped that the flexible publication policy of the ASC CYBERNETICS FORUM will foster and promote, the continuing evolution of cybernetic thought. The ASC CYBERNETICS FORUM is designed to provide not only cyberneticists, but also intelligent laymen, with an insight into cybernetics and its applicability to a wide variety of scientific, social and economic problems. Gontributions should be lively, graphic and to the point. Tedious listings of tabular material should be avoided. The Editors reserve· the right to make stylistic modifications consistent with the requirements of the ASC CYBERNETICS FORUM. No Substantive changes will be made without consultation with authors. They further reserve the right to reject manuscripts they deem unsuitable in nature, style or content. Opinions expressed in articles in the ASC CYBERNETICS FORUM do not necessarily reflect the opinion of the ASC CYBERNETICS FORUM or its editors, or the American Society for Cybernetics or its directors and officers. All material published in the ASC CYBERNETICS FORUM is Copyright by the American Society for Cybernetics who reserve a/1 rights.

lnstructions to Authors

Papers already published or in press elsewhere are not dard letter-size paper. The first page of the manuscript acceptable. For each proposed contribution, one original should carry both the first and last names of all authors and two copies (in English only) should be mailed to Dr. V.G. and their affiliations, including city, state and zip code. (Note Drozin, Physics Dept., Bucknell University, Lewisburg, PA. address to which galleys are to be sent.) All succeeding 17837. Manuscripts should be mailed flat, in a suitable pages should carry the last name of the first author in the envelope. Graphie material should be submitted with suit­ upper right-hand corner. able cardboard backing. Style: While the ASC CYBERNETICS FORUM demands Types of Manuscrlpts: Three types of contributions are a high standard of excellence in its papers, it is not a considered for publication: full-length articles, briet commu­ technical journal. Authors should avoid mathematical for­ nications of 1,000 words or less, and letters to the editor. mulae and lang lists of references or footnotes. Titles Letters and brief communications can generally be pub­ should be brief and specific, and revealing of the nature lished sooner than full-length manuscripts. Books, mono­ of the article. Acknowledgments and credits for assis­ graphs and reports are accepted for critical review. Two tance or advice should appear at the end of articles. copies should be addressed to the Editor. Subheads should be used to break up-and set off­ ideas in text. Processlng: Acknowledgment will be made of receipt of all manuscripts. The ASC CYBERNETICS FORUM employs Graphie Materials: All artwork submitted must be in fin­ a reviewing procedure in which ali manuscripts are sent to ished form suitable for reproduction (black on white) and two referees for comment. Wtlen both referees have re­ large enough so that it will be legible after reduction of as pljed, copies of their comments are sent to authors with the much as 60 percent. Photographs shouid be biack and white Editor's decision as to acceptability. Authors receive galley glossy no less than S"xl". proofs with a five-day allowance for corrections. Standard proofreading marks should be empioyed. Corrected galleys About the Authors: A briet biography (less than one should be returned to Colonlai Printing, West Market and page), aiong with a small photograph, must be sent with all 20th Street, Lewisburg, PA 17837. manuscripts. This will be included in the "About the Authors" section of each issue. Format: Manuscripts should be typewritten double spaced, on white band paper on one side only, leaving about Manuscrlpt Return: Authors who wish manuscripts re­ 3cm (or 1.25 Inches) of space araund all margins of stan- turned must include a stamped, self-addressed enveiope along with their manuscript. ASC PUBLICATIONS ORDER FORM Dr. Barry A. Clemson, American Society for Cybernetics • College of Education, Sibles Hall, University of Maine, Orono, ME 04469

Cybernetics Forum lssues (lf not entire volume) Cost v 6. issues #2-4 (1974) $51issue v 7. issues #1-4 (1975) $51issue v. 8. issues #1-4 (1976) $51issue v. 9. issues #1-4 (1979) $7.50/issue v 10. issues #1-4 (1980) $351volume

Journal of Cybernetics and Information Service v 1. issues #1·4 (1977) $451volume v 2. issues #1·4 (1979) $551volume v 3. issues #1-4 (1980) $601volume

Proceedings v 1 (1967) Purposive Systems $15 ______v 2 (1968) Cybernetics and the Management of Large Systems $15 ______v 3 (1969) Cybernetics, Simulation, and Conflict Resolution Out of Print v 4 (1970) Cybernetics, Artificial lntelligence, and Ecology $15 ______v 5 (1971) Cybernetic Technique in Brain Research and the Education Process $12 ______

Journal of Cybernetics

V #1 (1971) $10 _____ (Members of the Society are entitled to 50% discount) V 1 #2 (1971) $10 _____ V 2 #1-2 (1972) $10 _____ Total Purehase Cost V 2 #3-4 (1972) $10 _____ Handling Charge .50 V 3 #1 (1973 $5 _____ Total Cost V 3 #2 (1973) $5 _____

PREPAYMENT IS REQUIRED-Make checks to American Society for Cybernetics

Name Address ______Zip

MEMBERSHIP AND RENEWAL APPLICATION

TO: Membership Committe of ASC Piease consider my application for membershiplrenewal in the American Society for Cybernetics. (Annual dues are $25 for members and $10 for students. Dues entitle member to receive, free of additional charge, quarterly ASC CYBERNETICS FORUM, and the quarterly Journal of Cybernetics and Information Science.) Return together with your check (payable to American Society for Cybernetics) TO; Phyllis Carr Membership Chairwoman 30 Walker Ave. Gaithersburg, MD 20760

Name ______Address ______City State Zip ______TitleiOccupation ______OrganizationiAffiliation

Signature ______International Congress on Applied Systems Research and Cybernetics December 12·15, 1980 Acapulco, Mexico Sponsored by The School of Computer Science, University of Windsor Society for General Systems Research Canadian Information Processing Society Computer Science Association Societe d'lnformatique Fondamentale American Society for Cybernetics The Congress will be held at the new Acapulco center, one of the most beautiful convention facili· ties in the world. The Congress will provide a forum for presenting and discussing scientific works in the areas of applied systems research and cybernetics. The main theme of the Congress is: The Quality Of Life and How To lmprove lt The emphasis for the 1980 Acapulco Congress will include: applications of systems research in social and natural sciences, advances in development of systems research methodologies, and applica· tions of cybernetics and systems concepts to ethical management of human systems. A special Symposium will be arranged to focus on the question of how systems research and cybernetics can be practically utilized to help us improve the quality of human life in our society and what scientists from different fields can do to more effectively mobilize their resources toward this goal.

Any topic from the area of Applied Systems Research and Cybernetics can be presented at the Congress. The submitted papers may cover: system modeling and Simulation, research on social, poli· tical, economic and ecological systems, methodology for analysis of systems behavior, systems research in education, health care systems, biological systems, information-processing and communi· cation systems, ethical and philosophical aspects of human systems management, and others.

All inquiries should be sent to:

Dr. George E. Lasker, Congress President, School of Computer Science, University of Windsor, Windsor, Ontario, Canada N9B 3P4