Instituto Mexicano de Relaciones Grupales y Organizacionales
Mexican Institute of Group and Organizational Relations

Organizational Learning as Cognitive Re-definition: Coercive Persuasion Revisited

Edgar H. Schein

MIT Sloan School of Management

1997, The Society for Organizational Learning.  All rights reserved.

Generative Learning and Culture Change as Coercive Persuasion

The purpose of this essay is to link the concept of coercive persuasion, popularly known as "brainwashing" to the concept of cognitive re-definition or reframing which is an essential element of what has come to be called generative learning. Adaptive learning is applying the same old concepts or skills in new ways. Generative learning or what Argyris and Schon (1974, 1996) call "double loop learning," what Bateson (1972) called "deutero-learning," and what Michael (1973) called "learning to learn" requires the learner to reframe, to develop new concepts and points of view, to cognitively re-define old categories and to change standards of judgment. Such changes increase the learner's capacity to deal with situations in new ways and lay the basis for developing radically new skills (Senge, 1990). 

When we speak of "culture change" in organizations we are typically referring to this level of learning (Schein, 1992). The magnitude of changes required can be appreciated when we observe that the kinds of culture changes being advocated and touted involve "building trust and openness," "empowering employees," asking employees to "commit" to organizational tasks, asking managers to work in "flat and lean" organizations, asking previously competitive units to become "teams," and so on. To make changes at this level requires more than behavioral change. It requires the learner to reframe the situation, to learn new concepts and to develop new attitudes or the behavior changes will not last once the immediate incentives are removed. If we are to understand the full implication of such generative learning and culture change, it is essential to understand how cognitive redefinition comes about, and, to this end, we must understand coercive persuasion. 

Many contemporary proponents of organizational learning, notably Peter Senge, argue that generative learning is not only necessary for organizational survival and growth but that it is, in the last analysis, the only consistent advantage that organizations will have over their competitors. Implicit in this point of view is the idea that if one develops the right set of "capacities" for learning, generative learning becomes a voluntary, even pleasurable process. But, as I will argue, in order to develop those capacities one must undergo a learning process that is functionally equivalent to what POWs underwent in the communist prison camps and that involves at the early stages periods of sufficient anxiety to motivate learners to reject the learning situation unless they are coerced either by physical restraint, positive incentives, or the threat of loss of desired rewards to remain in the learning situation. 

Most generative learning involves questioning one's basic assumptions, and this is an inherently anxiety provoking process that will be resisted. At the extreme this resistance takes the form of simply not grasping what the new concepts are and dismissing them as irrelevant. The coercive element of coercive persuasion comes into play in that the easiest way for the learner to avoid the anxiety of examining his or her own tacit assumptions is to walk away from the situation. For the learning process to begin, therefore, requires either some incentives and/or some constraints that keep the learner in the learning situation. If the incentive is to learn and the learner is inwardly motivated to go through the pain of learning, so much the better. But the key is that the learner must remain in the situation even though it becomes painful at times. Before exploring the analogy to coercive persuasion further it is necessary to describe what the POW experience consisted of. 

Coercive Persuasion Described

Coercive persuasion as a concept was first developed in trying to understand the seeming conversions and collaborative behavior of prisoners of war who were subjected to interrogation and indoctrination during World War II and particularly during the Korean conflict (Schein, 1956, 1961). Whether they were military POWs or civilians arrested suddenly in their homes, their interrogators routinely treated them as guilty and accused them of crimes, of espionage, and of holding values that were inimical to "the people." 

Most prisoners reported that they were convinced of their innocence, they did not have a clue what the interrogators were talking about, and if pressures got severe enough they were willing to sign false confessions, to engage in collaborative behavior, and to allow themselves to be used in propaganda activities such as posing for pictures, but they never accepted their guilt. They were coerced but not persuaded. On the other hand, there was a substantial number of civilians including students, businessmen, missionaries and members of various local religious orders who had lived in China for decades who came away from several years of imprisonment admitting their guilt, saying that they had been spies and criminals, and expressing gratitude to the Chinese Communist captors for being treated leniently given the magnitude of their crimes (Lifton, 1956; Schein, 1961). For all intents and purposes they had undergone a generative learning process, though, in this case, the outcome was viewed as undesirable from our point of view. What made it generative rather than adaptive is that the repatriates really came to believe in their guilt and many of them worked on behalf of their communist captors to bring the message to others after they were released and "free" to think whatever they liked. 

I described the process these prisoners went through as "coercive persuasion" to indicate that if a prisoner was physically restrained from leaving a situation in which learning was the only alternative, they would eventually learn through a process of cognitive redefinition. They would eventually come to understand the point of view of the captor and reframe their own thinking so that the judgment of having been guilty became logical and acceptable. In effect they had undergone what might be called a "conversion" experience except it did not happen in the sudden way that religious conversions are often described. 

The essence of this process, from the point of view of the captor, was to create a situation in which several conditions obtained simultaneously: 

The prisoner was put in jail with an indeterminate sentence, articulated by the captor as "you will never get out of here until you make a sincere confession and accept your guilt as a spy and criminal, and recognize how your bourgeois values are an inherent danger to the communist people." 

The prisoner was put into a group with others who were more advanced in their learning process. 

The group was rewarded on the basis of its total progress; only if all the members learned the new point of view would they get more privileges and fewer punishments. 

The new point of view was presented in many ways--personally by the interrogator, by lectures, by printed materials and by more advanced group members in informal discussion. 

The main vehicle for learning and assessing the degree of learning was the written confession and self-criticism which was required as a regular activity and served to stimulate the prisoner to rethink his or her past actions and begin to assess them from a new point of view. 

Any evidence that the prisoner was beginning to grasp the new point of view or new concepts was instantly rewarded and, on the other hand, any evidence of insincerity or superficiality of understanding was severely punished. 

Communications that in any way reinforced the old point of view or that reminded the prisoner of his or her links to old membership or reference groups were withheld, e.g. mail from home was delivered only if it contained bad news such as the message that a spouse was seeking a divorce or that a valued friend had died. 

Physical pressures of all sorts were constantly applied to weaken the prisoner's physical strength, with sleep deprivation being the most potent of these pressures; "torture" was only used as a punishment for insincerity or lack of motivation to learn. 

Psychological safety was produced for the prisoner by fellow prisoners who were farther along in their re-education and could be supportive of the target prisoner's effort. 

In this kind of physical, social and psychological milieu the process of learning can be thought of as occurring in several stages.  Because western prisoners came into the situation with a clear self-image and set of judgments about what crime, guilt, and espionage meant, their first potent experience of being arrested, accused of guilt, thrown into jail and threatened with dire consequences if they did not confess served as a powerful "disconfirmation." But how could they confess if they did not believe in their own guilt and did not have a clue what it was all about, except the rationalization that it was a giant mistake that would shortly be cleared up. They were certain that they could convince the interrogator of their innocence and that the arrest must have been a mistake. When the interrogator would counter with "you are guilty because you have been arrested, we do not arrest innocent people," the prisoner would simply not understand what that could mean except that it was a mistake or a miscarriage of justice. 

If the prisoner was in a group cell, he or she might discover others who were similarly convinced of their innocence and a few who said "you are guilty." The new prisoner's first reaction would be that the cellmates must being "crazy" or that they had been planted there to confuse him or her. But as days, weeks, and months went by with no mail, no outside intervention, constant further interrogation and pressures to write a confession and self-criticism, prisoners would begin a process of self-examination.  Most everyone is, after all, guilty of something so one's generalized capacity for guilt began to be felt as possibly connected to the present plight. But there was still no insight into what the captor meant by guilt since the prisoners "knew" they were not government agents, had not sent intelligence information home, and had, in fact, done nothing but lead their ordinary life in pre-communist and after the takeover communist China. 

I am suggesting that this sense of frustration and puzzlement which comes about from being heavily disconfirmed is comparable to what it feels like to an employee or manager when they are told that the way they have worked for decades is no longer adequate and that they will have to learn some completely new concepts and skills in order to retain their jobs. For someone who has spent a lifetime in individualistic competition to be told that their concepts and prior behavior are now "wrong," that they now have to be a team player, share their insights, help their peers, trust their bosses and commit completely to the welfare of their employing organization might seem just as "crazy." 

How then do either the prisoners or the employees get past this fundamental impasse? Two further psychological processes had to come into play before cognitive redefinition became possible. First, the level of survival anxiety or guilt had to be strong enough to lead the prisoner to psychologically surrender, to give up, to experience despair, or in Alcoholics Anonymous terms to "bottom out."  The essence of this state is that the person accepts that he or she is no longer in control and that "higher powers" will determine his or her fate. The person is now willing to put him or herself into the hands of others and what this amounts to is accepting the possibility that the captor may have some knowledge or power that one must begin to pay attention to. The defense of denial, the sense that this is unreal and one will be released at any moment, that justice will prevail, the sense that the interrogator captor is just playing a game has to be given up. But even this is not enough for new learning to occur. It is possible to continue in this state of despair. 

The second process that had to come into play was that the prisoner had to begin to feel psychologically safe in this state of openness and vulnerability. If there was insufficient psychological safety and despair was high enough, the prisoner would experience a mental break and deny reality in a psychotic way. To become open to new information, the prisoner had to feel some support for the new learning process that was now going to begin. 

In the political prison this support was provided by the "good interrogator," or, more typically, by the cell mates who now became supportive because the new prisoner was displaying signs of willingness to learn. In effect the cell mates became mentors to the prisoner and showed him or her how to think in new ways, how to cognitively re-define certain critical concepts. But they could not do this until the prisoner was "ready," until enough disconfirmation and anxiety had built up to allow the prisoner to let go of his or her prior assumptions of injustice and invulnerability. 

At this point prisoners began to identify with one or more of their cellmates who were more advanced in their learning and through them learn some of the concepts that underlay the structure of communist thinking. For example, they learned that a "crime" was not, as westerners thought, an act that could be proven to be harmful and against a law; a crime was any action that could at any time in the future become harmful to "the people." In a collective groupist society, self-seeking was a crime because it harmed or could harm others whether or not one consciously intended it. Prisoners learned that middle class "bourgeois" attitudes led to behavior that was automatically harmful, such as a Jesuit Mission employing lower class Chinese houseboys or gardeners and thereby exploiting them to do the menial work. Writing postcards home about the beautiful ricefields in the country was automatically espionage because that information could at some point be of value to an enemy. At the extreme there were examples that seemed ridiculous such as defining rolling over in one's sleep into someone else's space in a crowded cell as "imperialistic expansionism." But however ridiculous it seemed, the conceptual system hung together and gradually the prisoner
came to recognize how all kinds of innocent acts from a western point of view were crimes from the communist point of view. 

"Cognitive redefinition" involved two different processes. First, concepts like crime and espionage had to be semantically redefined. Crime is an abstraction that can mean different things in different conceptual systems when one makes it concrete.  Second, standards of judgment had to be altered. Even within the western concept of crime, what was previously regarded as trivial was now seen to be serious. The anchors by which judgments are made are shifted and the point of neutrality is moved.  Behavior that was previously judged to be neutral or of no consequence became criminal, once the anchor of what was a minimum crime was shifted. These two processes, semantic re-definition and changing one's anchors for what is good or bad, acceptable or unacceptable, are the essence of cognitive re-definition. It is through these two processes that "reframing" occurs.  But it is only possible for these processes to occur once the learner has developed the openness that comes from despair and found the psychological safety to begin to learn. 

It should be noticed that both semantic shift and shift in anchor is necessary for genuine reframing to occur. One can engage in semantic shifts alone and "understand" how someone else might define crime differently and thereby enlarge one's intellectual horizons. But that alone does not produce voluntary behavior change. It is when one recognizes that one's prior behavior is from the new point of view "bad," that one has truly reframed the concept and launched into a learning process of how to avoid such bad behavior in the future. By identifying with their cellmates prisoners came to see that what they had done was indeed harmful and could, then, make a sincere confession. 

Once a sincere confession had been made, prisoners were usually released fairly quickly, leading to the assertion by the repatriates that they had been leniently treated given the magnitude of their crimes. The western reaction that this was bizarre behavior resulting from "brainwashing," reflected the western semantics and standards of judgment, leading to the irony that the repatriates now felt just as the interrogator had felt in regard to them--"you just don't understand." 

What does all of this have to do with organizational learning? 

Generative Organizational Learning as Coercive Persuasion

I would suggest that generative organizational learning puts most managers and employees into a situation comparable to the prisoner in a political prison. It is not a spontaneous joyful process to give up one's beliefs, values and concepts in favor of untested and inimical new concepts and anchors for judgment. It is not a particularly comfortable situation to be subjected to re-engineering or culture change programs with the clear threat that unless one participates wholeheartedly one might lose one's job. Particularly at a time when downsizing and massive layoffs are the order of the day, change or learning programs are likely to be viewed as highly coercive. 

The executive who launches these programs is not likely to appreciate the degree to which they become coercive, nor the degree to which they challenge the assumptions on which the organization has previously been built. For the learning process to begin, some heavy disconfirmation is likely to be required, leading to high levels of both survival and learning anxiety, and ultimately to the creation of despair. Only when enough psychological safety has been provided will the learner even hear the new message, much less accept it and internalize it. 

It may seem absurd to the reader to draw an analogy between the coercive persuasion in political prisons and a new leader announcing that he or she is going "to change the culture." However, if the leader really means it, if the change will really affect fundamental assumptions and values, one can anticipate levels of anxiety and resistance quite comparable to those one would see in prisons. The coercive element is not as strong. More people will simply leave before they change their cognitive structures, but if they have a financial stake or a career investment in the organization, they face the same pressure to "convert" that the prisoner did. 

To the extent that the analogy holds, one can now see what the problem is. The new cultures that are usually called for involve concepts, attitudes, and skills that are typically not understood in the first place, nor accepted even if partially understood.  Consider, for example, what it means to impose a "culture of teamwork" based on "openness and mutual trust" in an individualistic society that has operated by competition and survival of the fittest and has created by this means one of the most powerful economic systems in the world. So either the person calling for the new culture does not understand what he or she is really asking for, or the targeted managers and employees simply will not understand or accept it and the leader will find him or herself in the same position the interrogator was in with prisoners who kept insisting that they were innocent. 

In a similar vein, consider what it means to "empower" employees, to ask them to participate and to become committed to organizations that have been built on the "divine rights" of managers to hide essential economic information from employees on the grounds that it is none of their business, that have treated stockholders as the only constituency worth responding to seriously, that are driven by the capital markets on the one hand and technological imperatives on the other hand. Consider what employee empowerment implies for the levels of management whose whole careers have been built on supervising the people below them. 

Consider what it means to abandon hierarchy in favor of "flat organizations of inter-locking and inter-dependent project teams with shifting leadership and membership" in organizations whose very essence has been hierarchy as the prime means of coordination and control, and the major means of identifying career progress. Consider what this means to managers whose power has been based on their organizational position, whose very concept of management has been to be "over" others, to give orders, to call the shots, to be individually accountable, to be a successful rugged individualist. 

Consider what it means to shift the emphasis from caveat emptor and "you can have any color so long as it is black" to creating not only satisfied but "delighted" customers who are encouraged to want and demand anything, anytime, anywhere. Consider what it means to abandon our linear cause and effect way of thinking and substitute systems thinking. Each of these kinds of changes involve extensive semantic redefinitions of core concepts such as "managing," "coordinating," "teamwork," "interdependency," and "commitment," and drastic changes of the anchors around concepts like "quality," "customer satisfaction." 

It is one thing to advocate from an outsider or academic perspective that organizations will have to adopt such new assumptions, learn that they actually work, and, thereby, gradually build up new kinds of organizational cultures. It is quite another thing to expect that just advocating such new assumptions will bring them into being. Organizations will either have to go through painful periods of coercive persuasion, or they will have to start with new populations of employees and managers who hold such assumptions in the first place. In either case, it is likely to be a long and difficult road so one should not kid oneself that cultures can be ordered up and cooked like restaurant meals. And even if we successfully impose and/or learn new assumptions, we still do not know whether they will make organizations more effective or competitive. New cultures can be imagined, but they will only be created by experienced success over a long time. 

The more one thinks about it, the more one sees that imposed culture change and coercive persuasion are quite similar. It remains to be seen whether the level of organizational change that is implied by "generative" learning can be accomplished without imposed culture change. And if such imposed culture change is involved we must accept the reality that learning may involve some painful periods of coercive persuasion. One of the most difficult aspects of this reality is that we cannot ignore that the same methods of learning, i.e. coercive persuasion or colloquially brainwashing, can be used equally for goals that we deplore and goals that we accept. In making organizations more competitive we may well resort to methods that under other conditions we would deplore.  But when we encounter resistance from managers and employees who are the target of these change programs we should not be surprised. They may be reacting to the methods as much as the message. They may resent the feeling that they have to learn new concepts, attitudes, and skills "or else." We cannot escape the moral choices that then have to be made. The issue is similar to that faced by parents of children who have joined cults that have used coercive persuasion. Are the parents in turn justified in kidnapping their child out of the cult and using a deprogrammer to coercively persuade them back to a set of values that the
parents are more comfortable with? Are managers justified in imposing new methods of thinking on employees who have been programmed by decades of industrial experience to think in a certain way? 

These are not easy questions to answer, but it is time we looked at organizational learning realistically and accepted the fact that for most members of the organization the choice between holding on to their prior beliefs and learning new beliefs, values, concepts, and behaviors is often not a choice at all. Not to learn means loss of job or career advancement. Learning therefore is a coercive persuasion process whether we admit it or not.


Argyris, C. & Schon, D. A. Organizational Learning. Reading, MA.: Addison-Wesley, 1974. 

Argyris, C. & Schon, D. A. Organizational Learning II. Reading, MA.: Addison-Wesley, 1996. 

Bateson, G. Steps to an ecology of mind. N. Y.: Ballantine, 1972. 

Lifton, R. J. "Thought Reform" of Western Civilians in Chinese Communist Prisons. Psychiatry, 19, 1956, 173-195. 

Michael, D. N. On Learning to Plan and Planning to Learn. San Francisco, CA.: Jossey Bass, 1973. 

Schein, E. H. The Chinese Indoctrination for Prisoners of War: A Study of Attempted Brainwashing. Psychiatry19, 1956, 149-172. 

Schein, E. H. Coercive Persuasion. N.Y.: Norton, 1961. 

Schein, E. H. Organizational Culture and Leadership (2d Ed.) San Francisco: Jossey-Bass, 1992. 

Senge, P. The Fifth Discipline. N.Y: Doubleday, 1990.


ii 2016
ix 2013

vi 1999