John L Jerz Website II Copyright (c) 2015

Economic and Social Development (Dunn, 1971)

Home
Current Interest
Page Title

A Process of Social Learning

Edgar S. Dunn, Jr.

"evolutionary experimentation is the fundamental character of the process of social learning."

"social problem solving engages in experimental tinkering with the system by subjecting it to a series of behavioral changes each of which is evaluated in terms of whether or not it improves the system's overall performance with respect to some desired goal or criterion of behavior."

"In short, prediction as hypothesis and planning as experimental design are the fundamental core of social learning. It is this rudimentary and fundamental form of prediction and planning out of which social knowledge and environmental control have evolved."

"Conditions of great uncertainty call for imaginative and flexible probings, not vacillation between inaction and commitment."

"In a basic sense, then, the means for fulfilling the aims of development is the ability to generate new ideas in action. This creative capacity is the ultimate resource in meeting human objectives"

"life... engages in a continuous, restless, opportunistic search for new forms and new functions. Each new experiment in self-transformation is tested in life's crucible to determine its relevance to the reality of the natural environment experienced by the organism. The maladaptive innovations are weeded out... The adaptive innovations are maintained in the species... They strengthen its members for the next stage of the contest."

JLJ - A citation by Erich Jantsch in The Self-Organizing Universe caused me to gamble a few Amazon dollars on this work by Edgar Dunn. Curious is his concept of "evolutionary experimentation," which in my opinion holds much value for game theory. Dunn cites a work by Harold Sackman (1967) as his source. On a curiosity I ordered Sackman's work - stay tuned for a future review. A citation of a work by Etzioni also will result in a future review.

Very specifically, a machine can execute a "scheme" developed by a human programmer that performs the concept of evolutionary experimentation.

Essentially, the programmer establishes conditions in which the machine will change its behavior - cues or diagnostic tests perhaps, to direct or guide exploration efforts down certain lines of play rather than others, or to initially select certain moves as promising, rather than others. Ideally, the intelligently constructed, programmed scheme covers all practical situational aspects of how the machine manages its attention. If performance tested thoroughly, indirect goal-directed behavior can be achieved, perhaps even leading some to conclude that that machine is "playing" a complex game of strategy. Hmmm...

From a high level, we see once again that the social sciences have previously grappled with advanced concepts which are useful for game theory. If a social program is deemed to produce useful results, it is likely continued, else it is abandoned. What if we proceed in the same fashion, using the results from an "idea" generating module on how to proceed, followed by a practical "How much should I care about that?" attention "focusing" heuristic? We might even have an internal conversation or the equivalent, in deciding how to proceed.

In my opinion, this work by Dunn is clearly foundational and can on its own form a meaningful approach to computer game-playing. In fact, as of 11 June 2016, my thinking is beginning to crystallize on a paper which will seek to introduce Dunn's concepts to a wider audience. My notes presented here are the rough outline of such a paper. The intermediate step I think will be to read some of Dunn's other works, to see if I am missing anything developed earlier or later in his own thinking.

p.3 the author is an economist concerned with the problems of regional and national economic development.

p.11 We have concluded that an explicit consideration of prevailing concept models of structure and change will provide the clearest picture of the move to generalize the practice of prediction and planning... They may be combined to form more complex models.

p.21 A learning system contrasts with a machine system because it has the capacity (a) to be reprogrammed in its behavior, or (b) to reprogram itself through the action of internal sources of new behavioral ideas, transformation moves, and transformation behavior.

p.21 These composite social behavioral systems, since they are designed and activated by man, are also learning systems - their behavior is subject to continuous reprogramming.

p.25 In order to increase the efficiency of programmed learning we wish to improve the speed and accuracy with which we can search for and identify relevant... or new behavioral options available... Prediction and planning take on a different aspect under the control of this concept. Prediction takes the form of the identification of new behavioral forms in the exogenous world. Planning takes the form of implemented programmed learning directed to forming the physical and human capital and the organizational forms appropriate to the new behavioral modes.

p.27 There seems to be a great deal of reluctance to incorporate non-deterministic processes in the concept models of structure and change of social systems. This may stem in part from the philosophical presuppositions of some social scientists. Undoubtedly it also stems from uncertainty about how to construct a meaningful nondeterministic model and a reluctance to deal with the complexity such a process might be presumed to embrace. Consequently, every effort seems to be exercised to stay within the deterministic frame.

p.27-28 Ashby (1962)... points out (pp. 266-69)... "...the appearance of being 'self-organizing' can be given by the machine... being coupled to another machine.... Then the part... can be 'self-organizing' within the whole.... Only in this partial and strictly qualified sense can it be understood that the system is 'self-organizing' without being self-contradictory."

p.28 machine systems can display quite complex adaptive behavior and take on the appearance of learning systems. However, such a complex machine system can act like a learning system only to the extent that it can be preprogrammed with selective criteria and well-defined learning processes. The control machine has to be given this capacity by another machine or human being... if the learning capacity of such a complex machine system is to be further generalized it will have to be governed by a hierarchical series of control units embodying successively more general criteria and process instructions.

p.30 an important part of social change - indeed the most important part - is the product of an open process of creative learning... Such a system has the capacity to reprogram its own behavior.

p.40 life... engages in a continuous, restless, opportunistic search for new forms and new functions. Each new experiment in self-transformation is tested in life's crucible to determine its relevance to the reality of the natural environment experienced by the organism. The maladaptive innovations are weeded out... The adaptive innovations are maintained in the species... They strengthen its members for the next stage of the contest.

p.40-41 each mutation or genetic recombination must survive two tests: (1) The "new idea" must be consistent with the continued viability of the organic system in which it arises... (2) The nonlethal mutation must, in turn, meet the test of relevance. If the new form or function aids the developing organism in seeking and utilizing energy in its host environment, it strengthens its survival characteristics... The favorable genetic combinations spread and the unfavorable ones disappear. In the process, phylogenetic change takes place.

p.45 Adaptive specialization characteristically generates incremental adaptations in form and function to special conditions of the environment (the fine-scale differentiating characteristics of its niche). Adaptive generalization characteristically generates new forms or functions often designated as evolutionary novelties. They result in new structural types, such as the evolution of the wing in the reptilian lineage, and they are more apt to involve harmonious changes in whole systems of organs.

p.52 As expressed by Mayr (1960, p. 355), "Most mutations appear to have only a slight ... effect upon the phenotype. More penetrant mutations are usually disruptive and produce disharmonious phenotypes ... and will therefore be selected against."

p.53 Rensch (1959, p. 289) points out that "mutations causing positive allometric growth [JLJ - allometry: The disproportionate growth of a part or parts of an organism as the organism changes in size, American Heritage Dictionary] of certain tissues will be especially important... such an increase of tissue may not serve any special function at first. Later on ... the superfluous tissue may be 'employed' by a new function in the course of subsequent evolution."

p.53 a genetic modification need convey to the organism only a very small margin of advantage for the character to spread through the population in succeeding generations.

p.56 The stimulus to adaptation and the motivation of the adapting species... is an incremental change that jeopardizes the biotype by creating a deficit in the resources or some other condition essential to its existence... The adaptations are all in the direction of offsetting the deficit and reestablishing species viability.

p.116-117 man was discovering certain environmental relationships that were sufficiently regular and reliable to permit them to be used to amplify the power of human behavior... the laws of relationship are such that man can use the knowledge of them as the basis for actual modification and control of some set of environmental events.

p.118 We observed earlier that as the power of generalization gave to man a cumulating power to predict and control aspects of his physical environment, the implementation of this knowledge through behavior required the development and organization of even more complex forms of social behavior.

p.123 Lowry (1967, pp. 197-198)... "...a wise designer should be prepared to sacrifice a good deal of visible efficiency to future flexibility..."

p.123 social systems behave as learning systems. A capacity for creative learning is essential for the transformation of social systems.

p.125 Creative adaptation and open-ended social learning inevitably invade the process.

p.132 If the prediction serves to anticipate a problem not yet manifest, it may prompt a planning action addressed to forestalling the problem and consequently will modify the prediction.

p.132-133 Game theory in its present stage of development is primarily absorbed with the determination of the most advantageous strategy in play, given the objective of the game and the nature of the rules of play. But this is an extremely narrow interpretation of the vital nature of the social process.

p.133 Social learning... is manifest as an iterative exploratory series of experiments in social action... We characterize this social learning process as evolutionary experimentation - a phrase first encountered in a work of Sackman's (1967).

p.133-134 Prediction takes the form of a developmental hypothesis... "We hypothecate [JLJ - hypothesize] that if we undertake a certain course of action, the performance of the social system will be modified in such a way as to improve its efficiency in satisfying the goals or objectives of the system." This is not known to be true on the basis of established deterministic laws... It is an experimental design. Planning takes the form of conducting an experiment... It can be viewed as testing the developmental hypothesis in action. If the developmental hypothesis is not falsified by the results of the experiment (i.e., if the performance of the system with reference to its goals is improved), the novel mode of behavior will be reinforced and persist. If the results call the hypothesis into question, a new or modified one will take its place and a new experiment in social action carried out.

p.135 evolutionary experimentation is the fundamental character of the process of social learning.

p.135-136 the object of the planning experiment is to change the performance of the system... To achieve this, social problem solving engages in experimental tinkering with the system by subjecting it to a series of behavioral changes each of which is evaluated in terms of whether or not it improves the system's overall performance with respect to some desired goal or criterion of behavior. The fact that this mode of experiment does not yield universal laws makes it no less scientific if it is consciously and objectively carried out and subjected to the test of performance.

p.136 In short, prediction as hypothesis and planning as experimental design are the fundamental core of social learning. It is this rudimentary and fundamental form of prediction and planning out of which social knowledge and environmental control have evolved.

p.140 A social system or man-machine system acting as a behavioral system is characterized by the organization of component behaviors in the pursuit of a set of objectives. If this were not so it would have no entity as a system. Since system objectives are often not adequately fulfilled by system behavior, the resulting anomalies give rise to the application of evolutionary experimentation. New ideas are developed and implemented that modify the behavior of the system in a way, if successful, that narrows the gap between objectives and performance.

p.145 projections are no longer viewed as templates for system engineering, but as instruments in probing realistic options. They are employed to develop some sense of the direction and perhaps the potential rates of change in certain dimensions of the system. They are not taken as an anchor for the future system design but as a rough indicator of emergent problems and as a finger of light point out some aspects of the future terrain - but becoming dimmer and less reliable as it approaches the time horizon.

p.146 in the formulation of a developmental hypothesis one may gain some insight into the comparative merit of behavioral options if one utilizes deterministic or game behavioristic models to perform limited pretests of the options... In short, the less general models of social change may survive in the service of the more general process of evolutionary experimentation.

p.149 The more complex a social system and the more variegated its behavior, the more likely it is to generate social reorganization and revised objectives. It has the behavioral variety that facilitates the formulation of new developmental hypotheses and their test through social action.

p.150 The principal problems and opportunities with which we are faced in both the developed and underdeveloped worlds do require social reorganization and paradigm shifts to be successfully negotiated. But the foregoing suggests that the social action that will make the greatest contribution to this end will be that directed to improving the capacity for social learning at both the individual and social levels.

p.152 Modern science rejects the notion that the terminal state of any physical or social system can be read from the course of history... However, it is sometimes assumed that planning may still take its goal from history if it is seen as an intermediate goal determined by more objective and justifiable scientific methods. This view sees history as an arrow that projects into the future. The limitations on the range of vision prevent perceiving its terminus, but over some reasonable range of time its direction and orientation can be ascertained. It is assumed, therefore, that we can preadapt by designing the social system that some projected intermediate future point will require, and undertaking to assist history in generating the transformation. Through prediction and planning man contributes to the efficiency of the historical process and facilitates its ends.

p.154 the thing we learn from Kuhn (not from Popper) is that the normal practice of evolutionary experimentation is the source from which the major changes in social form and function arise, just as adaptive generalization is another manifestation of the process of biological adaptation.

p.156-157 A statement by Nelson et al. (1967, pp. 173-174) illustrates the haphazard and inefficient way that social learning takes place in the formulation and initiation of public policy...

Under such circumstances the most fruitful way to proceed is sequentially and experimentally; neither by doing nothing because knowledge is less than perfect nor leaping farther than necessary in a pre-judged direction... The knowledge myth... seriously impedes the development of public policy. The channeling of large sums of money into programs predetermined on the basis of sketchy information narrows the range of alternatives that can be tried, and thus reduces the range of policy instruments that have to be tested. Further, it deters useful experimentation... It places a high premium on actions likely to yield simple-minded qualitative indexes of immediate success... Conditions of great uncertainty call for imaginative and flexible probings, not vacillation between inaction and commitment.

In short, where the learning process is not understood and consciously applied, adaptation to new problem situations that are complex tends to vacillate between the paralysis of inaction and the temptation of solutions visualized... as a holistic panacea. We can recognize these oscillating behavioral responses in individuals as well as social systems of all levels.

p.157 We must come to understand the process of evolutionary experimentation in the progressive redesign of social systems so that we can make a more conscious and purposive application of it to match our powers of social learning in the domain of physical systems.  The hazards of evolutionary experimentation will be diminished if we can make it less haphazard. To attain this result we must find ways to (a) improve the efficiency of evolutionary experimentation and (b) improve its directed character.

p.160-161 Of necessity, behavior directed to changing behavior embraces behavior directed to evaluating behavior... if a social science is to concentrate on evolutionary experimentation or social learning and establish principles for its sound practice, it must become involved with the role of goals and values.

p.176 the process of evolutionary experimentation is engaged in problem solving... The behavior of social systems is progressively modified by testing developmental hypotheses designed to improve the system's performance in relation to its target goals.

p.177 through the remarkable process of phylogenesis, man has come to embody certain basic characteristics of the evolutionary process itself in his behavioral motives... The irreducible absolute value of the evolutionary process is survival... The goal of survival stands behind all the other genetically inscribed instincts. They serve survival.

p.178 Activity directed to mere existence or self-maintenance, while necessary, is not sufficient to assure survival... Over the long run survival is better served by adaptability than by adaptation.

p.180 motives are of two orders, though in a given instance the orders may fuse. … there are deficit and growth motives. Deficit motives do, in fact, call for the reduction of tension and the restoration of equilibrium. Growth motives, on the other hand, maintain tension in the interest of distant and often unattainable goals.

p.180-181 Becker (1968, pp. 171-72)... Man is the only animal who is not “built into” his world instinctually... Man alone among animals gradually develops his own perceptual response world by means of imaginative guiding concepts. He is actually, in this way, continually creating his own reality... the cycle of eat, fight, procreate, and sleep - that absorbs the adult members of other species - has only the barest meaning for man. He... dwells in dreams and concepts, in past-present-future, a space-time largely of his own creation... the everyday reality that galvanizes all other animals tends to lack conviction for him. In order to render reality meaningful, in order to stimulate his own productive energies, man must bring his own meanings to the world, impart his own sense of conviction. This is the tragic burden as well as the unrivaled creative opportunity... Man creates his own meaning... Evolution has thus left man with the greatest burden and challenges: he is born, not into a world, but into a “backdrop,” that contains the raw materials for his manipulation and for the creation of his own world.

p.187-188 an excellent article by Polanyi (1968), appearing in Science.... makes the argument that life's irreducible structures are the boundary conditions or the system designs that control or harness the laws of inanimate nature, but which are themselves irreducible to those laws... The structure of machines and the working of the structure are thus shaped by man, even while their material and the forces that operate them obey the laws of inanimate nature. … we harness the laws of nature at work in its material and in its driving force to make them serve our purpose.

This harness is not unbreakable; the structure of the machine and thus its working can break down. But this will not affect the forces of inanimate nature on which the operation of the machine relied; it merely releases them from the restriction the machine imposed on them …

So the machine as a whole works under the control of two distinct principles. The higher one is the principle of the machine's design, and this harnesses the lower one, which consists in the physical-chemical processes on which the machine relies.

… we may borrow a term from physics and describe these ...restrictions of nature as the imposing of boundary conditions on the laws of physics and chemistry. …

p.189 behavior and structure, the act and its framework, are both system attributes. They are two sides of the social system “coin.” ...Polanyi says that a system is defined by its boundary... the boundary is conceived as a behavioral design. It is a pattern that constrains the behavior of subsystem components to action modes consistent with the design... It is design for behavior that is not reducible to the behavior of its components or subsystems.

p.190 The boundaries defining systems are hierarchical and emergent... These more general boundaries emerge through a process of evolution because they yield solutions to life problems. Emergent boundaries are equivalent to system reorganizations, which we have already identified in Chapter IV as equivalent to the notion of the paradigm shift.

p.192 According to [Herbert] Simon... He states (p.73) that “most things are only weakly connected with most other things; for a tolerable description of reality only a tiny fraction of all possible interactions needs to be taken into account.”

p.194 The problem becomes too complex to extend the diagrammatics, but these lines of connection can form complex feedback loops that vastly complicate system hierarchy.

p.195-196 In short, the multiple-purpose nature of many transformation and transfer processes results in a state where the activities performing these functions may simultaneously perform as components of two or more higher-order control systems. This leads to overlapping boundaries and reduces the adequacy of the familiar concept of the formal nested control hierarchy.

p.199-200 In the biological realm we recognize ecological systems as well. These are systems whose components are not subjected to direct or formal management control. The systematic behavior arises out of the coadaptations of multiple species that constitute a living interdependent complex. The elements of a set of biological species... define for each other the major aspects of their operating environment. They form feeding elements and the protective, supportive, or reproductive elements necessary for self-maintenance. Any change in either the scale or the behavior of one requires changes in the scale or behavior of others.

p.200 Many social systems are ecological in character...The components of the ecosystem... These systems are linked in an interdependent input-output relationship so that each enterprise... lives in an operating environment formed by the rest. They coadapt.

They tend to form a system of the whole because: (1) at some level of complex combination a significant degree of functional closure is evident; (2) the coadaptation tends to be equilibrating; and (3) there is an implicit objective of maximizing the material welfare of the system components. These attributes establish the boundary conditions of the system.

p.200-201 it is not likely that any social system exists that is an example of a pure managed system or pure ecosystem. In reality most social systems would be more properly characterized as managed ecosystems.

p.201 The stability conditions that help define the boundary condition of the system are imposed by explicit and conscious management of the system.

p.208 social behavior may be absorbed with [JLJ - formatting for readability]

  1. the task of collecting and examining the evidence that reveals the states of the operating environment at input-output termini and that maintains the organization of endogenous component behaviors,
  2. interpreting the meaning of these signals in the light of system objectives, and
  3. managing the component behaviors so as to reconcile the system objectives in the most efficient way with the operational environment.

Such social behavior is absorbed in the service of maintaining a steady state system [JLJ - not necessarily, if the system objectives are growth and signals are interpreted in light of a practical potential for "becoming", then successful system behavior is growth, not maintenance of a "steady state".]

p.208 The human cognitive process has the capacity to be active and anticipatory in nature, and to promote its goals through behavioral innovations. It has the capacity to discover new knowledge and new modes of behavior through play - that is, through behavior that seeks understanding and experience not immediately relevant to self-maintenance.

p.209 Behavior-organizing behavior may be pressed into the service of designing and implementing new behavioral modes. [JLJ - yes, but what do we call behavior that organizes behavior-organizing behavior?]

p.209 social reorganization is an essential element of evolutionary experimentation

p.210-211 social system control centers may learn in the sense that some units conduct analysis and acquire new knowledge. However, until such knowledge becomes embodied in a developmental hypothesis, incorporated into control system design and manifest in experimental social controls organizing component activities, social reorganization or social system learning does not take place.

p.211-212 In many instances... Change is brought about through component tinkering... This kind of normal problem solving is relatively low risk in character. Although the novel modes of behavior and controls may be new to the system, they have often been pretested in exogenous systems. It can be pretty well anticipated that their introduction will narrow the gap between total system goals and performance. Even where the subsystem reorganization is more experimental... The... modifications are subjected to careful and continuous reality testing in terms of the desired total system goal convergence. [JLJ - keep in mind that in certain strategic positions our "goal" can be to effectively determine - then aim to improve - the score we award ourselves in a diagnostic test of our practical ability to grow and adapt, to favorably reorganize, or even to hold our own in a difficult position.]

p.218 It is useful to note... that any change in social behavior generated by social learning is effected through a change in either a transformation process, a transfer process, or both. There are no exceptions. This is just as true of symbolic components of social learning as it is of the processes of transformation and transfer associated with material and energy transformations.

p.219 It is possible for a change in a transformation process to have either a direct or indirect effect upon a transfer process, and vice versa.

p.221 All social problems and all social problem solving emerge out of the anomalies of social organization. Social system boundaries are often found to be ineffective activity organizers. The controls are not adequate to serve the goals. This discrepancy is a constant source of... dissatisfaction and leads to efforts to improve the relationship by means of evolutionary hypotheses that are tested in action. These evolutionary hypotheses have to do with the modification of relationships between system goals and controls. So we see that social system learning emerges out of the anomalies of social organization and the acts through an experiment in social reorganization.

p.221 How do we organize for learning?

p.222-223 In an analytical essay, Bennis and Slater (1968) call attention to the fact that we live increasingly in a "temporary society." Social systems are temporary in the sense that they undergo frequent modification and reorganization... The authors advocate a new approach to organization they would characterize as "organizational revitalization,"... "There will be adaptive, rapidly changing temporary systems.... groups will be arranged upon an organic rather than mechanical model, meaning that they will evolve in response to a problem rather than to preset, programmed expectations. People will be evaluated... flexibly according to competence..." (p.98)... "...The vocabulary for adaptive organizations requires an organic metaphor, a description of a process, not structural arrangements" (p. 120).

p.228 Changes in both endogenous and exogenous environments... require the formulation of developmental hypotheses leading to experimental system reorganization. The structure of organization needs to allow for some problem-oriented groups that can act to bring about network realignments and the readjustment of goals and controls.

We conclude that the more a social system is confronted with problem-solving situations and the need for social learning, the less it can rely on formal prescriptive hierarchical controls.

p.229 consensus through coercion or propaganda - a spurious kind of cohesion that is dynamically unstable.

p.229 Critical to the organization of social learning is the understanding and practice of communication as dialogue.

p.231 In dialogic communication it is knowledge and behavioral designs that are both transformed and transferred... It is through the... transfer and exchange of ideas that the transformation of ideas takes place - that creativity is spawned... In organizing the process of social learning we must obviously devote much more attention to the design and implementation of systems that make dialogue possible. It is through dialogue that developmental hypotheses are generated and the experiments in social system transformation are carried out... The dialogical process is the key to the agreement upon the context of the system - its goals and controls.

p.237 We have examined the limitations of the conventional growth models and have concluded that the concept of learning is a more appropriate metaphor... We concluded that only a creative learning model is adequate to cope with the reality of social and economic development.

p.238 mankind, as individuals and as groups, is capable of behavior directed to changing behavior... this behavior-changing behavior embodies two other distinctive aspects of social behavior: behavior directed to evaluating behavior and behavior directed to organizing behavior.

We observed, further, that behavior-changing behavior conformed to a process that can best be characterized as evolutionary experimentation - the process of social learning.

p.239 phylogenesis gradually evolved a program (genotype) that provided the organism with the power to reprogram itself - to act as a true learning system at the organism level.

p.241 the social system experimenter... exists as an endogenous component of the system he is attempting to understand and transform... He is immersed in the act of social system self-analysis and self-transformation. He is the agent of social learning - a purposive, self-actuating, but not fully deterministic process... the social system which engages his activity is phenomenologically unique and both its structure and function are temporary in character. He is engaged, rather, in formulating and testing developmental hypotheses... a presupposition that, if the organization and behavior of the social system were to be modified in a certain way, the goals of the system would be more adequately realized.

p.241 Problem solving - hypothesis formulation and testing - is an iterative, sequential series of adaptations of an adaptable, goal-seeking, self-actuating system. It can be characterized as evolutionary experimentation.

p.243 Because the process has not been understood and consciously applied, social change has frequently been dominated by an attempt to implement change by processes incompatible with the reality of social evolution. Acting on the basis of inadequate paradigms and metaphors, we have been inclined to practice a form of social engineering. It is presupposed that the change agents... have the knowledge and power to design a terminal state that will bring about consistent goals and controls in a deterministic fashion.

p.246 When one begins to operate under the control of a social learning paradigm the role of organization and the nature of its study are extended in important ways. Then social reorganization becomes revealed as both the target and the instrument of social change.

p.247 When we are faced with the reality of the social learning process and work under the control of its metaphor, it becomes evident that since social systems are frequently temporary systems, less emphasis needs to be given to the nature and design of optimum transfer networks and more to the design of adaptable networks - systems that can more easily be adapted to the requirements of new social goals and controls.

p.248 The social learning metaphor... sees the hypotheses of evolutionary experimentation as subject to testing and reinterpretation in the light of the record of history.

p.250-251 the emerging experiments in the organization of social learning will have to be performance tested by an emergent history... the evolutionary process is not deterministic... What we can hope for is that... we can refine our negative predictions. We may bring the understanding and exercise of the process to the point that we will be able to rule out a larger number of developmental hypotheses before behavioral testing because they are seen not to satisfy the necessary conditions of the process.

p.270 when we are faced with a learning process: "The central requirement becomes an epistemology capable of handling expansions of knowledge, breakouts from the limits of prior wisdom, scientific discovery. ... A focus upon the growth of knowledge, or acquisition of knowledge, makes it appropriate to include learning as well as perception as a knowledge process." (Campbell, forthcoming.)

p.272 Kuhn's (1964) interpretation... "... something like a paradigm is prerequisite to perception itself. What a man sees depends both upon what he looks at and also upon what his previous visual-conceptual experience has taught him to see. In the absence of such training there can only be, in William James's phrase, 'a bloomin' buzzin' confusion.' "

p.273 Ittelson and Cantril (1954)... put forth a theory of perception... as "a transaction" taking place between concrete actors and concrete situations. Perception is viewed, in essence, as a dialogic process.

p.280 guiding decision processes in unstructured and open-ended decision contexts... She [Dr. Mack] sets as the proper objective "progress toward rather than the achievement of goals. ... proper goals are always unachievable. They are an arrow and not a point." [JLJ - such as, in my opinion, a diagnostic test of the adaptive capacity to mobilize coercion, in a complex game of strategy.]

p.281 special mention should be made of a work that combines some of the features of both decision theory and organization theory. Sackman's Computers, System Science, and Evolving Society (1967) illustrates remarkably well the emergent character of the social learning paradigm and is one of two works that comes closest in spirit to the intent of the present volume... Sackman writes... Development of the SAGE air defense system... complex control systems... are the product of a continuous process of what Sackman terms evolutionary experimentation.

p.294 Along with the work of Sackman, the recent writing of Etzioni (1968) [JLJ - The Active Society: A Theory of Societal and Political Processes] comes closest to presenting a comprehensive view of the process of social learning.

p.295 [Etzioni quoted] A societal unit has transformability if it also is able to set - in response to external challenges, in anticipation of them, or as a result of internal developments - a new self-image which includes a new kind and level of homeostasis and ultra-stability, and is able to change its parts and their combinations as well as its boundaries to create a new unit. This is not a higher-order ultra-stability but an ability to design and move toward a new system even if the old one has not become unstable. It is an ability not only to generate adaptive changes or to restore new stability to an old unit, but also to bring about a new pattern. (p. 121)

p.295 [Etzioni quoted] An active approach to societal decision-making requires two sets of mechanisms; [JLJ - formatting below added for readability]

(a) a high-order, fundamental policy-making process which sets basic directions and

(b) an incremental process which prepares for fundamental decisions and revises them after they have been reached.

[JLJ - I have previously advocated the two-step process of asking (in dialogue form): How might I proceed? followed by How much should I care about that?]

p.295-296 [Etzioni quoted] The main questions for the transformation toward an active society are whether or not societies can mobilize themselves and their member collectivities to high, crisis-like if not higher, levels in noncrisis situations, and whether or not they can generate power for internal self-transformations instead of exerting their wills on other societies. Further, can this level and kind of mobilization be attained without generating so many counter-currents and so much alienation that the consensual base of society and values related to it will be undermined as the realization of the values expressed in the goals advanced is enhanced? ...(p.399)

p.296 Lindblom has focused on the political aspects of the process of social problem solving... a process of "disjointed incrementalism." (1965). He emphasizes that social problems are not truly solved but are subject to never-ending series of attacks through the political problem-solving process. The process is incremental and deliberately exploratory.

p.307 Once adaptive behavior evolves, an important part of the "available means" consists of the ideas embodied in conscious behavior that brings... resources... to a controlled application to human needs.

p.309 If social development serving human development becomes the goal, the ultimate resource, the basic means for achieving this end, is the ability to develop new ideas that improve the relationship between social system goals and controls operating in a mixed physical and social environment... In a basic sense, then, the means for fulfilling the aims of development is the ability to generate new ideas in action. This creative capacity is the ultimate resource in meeting human objectives