Home Articles Reader Opinion Editorial Book Reviews Discussion Writers Guide About TCRecord
transparent 13

Alternative Schools: A Behavioral Analysis

by Chris Argyris - 1974

The thesis of this paper may be outlined as follows: 1. The basic thrust of alternative schools (public or private) is freedom—opposition to the authoritarian aspects of traditional public and private schools; 2. Political factors, economic factors, social factors, legal factors, and bureaucratic factors are all plausible and important considerations for the effectiveness or demise of alternative schools; 3. There is, however, an additional factor, namely, the way the participants actually behave toward each other in the schools; 4. Recent research suggests that people may be "programmed" by our culture to behave in ways that cancel out their uniqueness and reduce their effectiveness in groups.

Dr. Chris Argyris is James Bryant Conant Professor of Education and Organizational Behavior at Harvard University.

The thesis of this paper may be outlined as follows:

1. The basic thrust of alternative schools (public or private) is freedom—opposition to the authoritarian aspects of traditional public and private schools. The assumption is that children are naturally curious and motivated to learn by their own interests. Coercion, regimentation, teachers with absolute power, emphasis on obedience and discipline, all combine to inhibit learning.1

2. Political factors (recalcitrant school boards, hard-hat philosophies), economic factors (inadequate funds), social factors (teachers and administrators with middle-class values), legal factors (outdated and unfriendly legislation and laws), and bureaucratic factors (the educational establishment) are all plausible and important considerations for the effectiveness or demise of alternative schools.

3. There is, however, an additional factor, namely, the way the participants actually behave toward each other in the schools. This factor is too often neglected in the literature.

4. Recent research suggests that people may be "programmed" by our culture to behave in ways that cancel out their uniqueness and reduce their effectiveness in groups. People inculcated with such behavioral incompetence will tend to design organizations that protect their incompetence. Consequently, the internal activities of social educational innovations, such as, alternative schools, will necessarily become defensive; the defensiveness, in turn, will tend to create conditions of organizational entropy, where organizations will tend to produce valid information for the unimportant problems and invalid information for the important issues; and therefore, these educational innovations will tend to deteriorate, even if the political, economic, social, and legal forces are in their favor.2

The position is not that the creators and participants of educational innovations are a special breed of incompetent people. The position is that probably all people in our society are so "programmed." These "programs" are second nature to us, and therefore, are rarely clearly seen until someone tries to create innovations or until others fight the innovations.

To make an analysis of how people in alternative schools actually behave toward each other, data were required that described the actual behavior within schools. Unfortunately, most of the literature on alternative schools tended to be journalistic or to be based upon statements about behavior (questionnaire or interview data). Three cases, however, recently became available where some systematic observational records containing relatively directly observable data were included.3

Admittedly, three cases do not constitute proof of a thesis. But the purpose of this limited analysis is not to prove a thesis. The hope is that the analysis may serve to establish that the thesis is worthy of consideration and further study. If this objective is accomplished, then perhaps the paper will have helped to increase the probabilities that educational innovations will be studied less as black boxes with inputs and outputs and more in terms of their internal processes that illustrate how the inputs are transformed into throughputs and are ejected as outputs.

Before the findings may be described, it is necessary to present, albeit in summary form, some of the major concepts and empirical findings of the research that underlies the analysis.


The foundation of this analysis is research that has been conducted on the nature of organizational illness and health. The first phase of this research concludes that modern organizations are designed with structures (pyramidal), control systems (such as budgets), and management information systems that will tend to lead to organizational decay and entropy. Fortunately for any given generation of .occupants, the rate of entropy is so slow that any given generation cannot be held responsible for the decay. Over the years, though, the dry rot may become so thick that today's organizations are becoming costly to administer, are resistant to change, and are producing inferior products or services. For the first time citizens of all ages and backgrounds are showing increasing mistrust of and decreasing confidence in the effectiveness of our institutions as organizations.4

In an attempt to redesign organizations and control systems, work was undertaken with top people in organizations who had power to introduce new designs. It soon became evident, however, that the top groups were not sufficiently effective for their members to design and implement genuine change. We found that groups at all levels of management were not very effective in problem-solving, decision-making, and implementation. Indeed, the groups not only mirrored the dry rot of organizations, they reinforced it; and the organization, in turn, coerced the groups to become more ineffective.5 Apparently there existed a self-reinforcing relationship between organizations and the groups within them that perpetuated and enhanced the ineffectiveness of both.


Before work could begin on redesigning organizations and groups, it was important to assess the contributions individuals made to this problem. Did individuals represent an additional set of causal factors for systemic ineffectiveness? If individuals were released from organizational and group constraints, would they be capable of behaving in more effective ways?

Several modes for collecting such data were developed. This article focuses on the data generated from one of these methods; namely, case studies that respondents generated. In effect, the case studies described how individuals, if free to do so, would tend to design their life and their environment. Slightly over 300 case studies have been obtained to date, from individuals with a wide range of ages (18-66 years), male and female, and representing black, Chicano, and white.6

Some of the results of these studies are as follows:

1. Human beings use micro-theories of action to inform their behavior. The theories, if made explicit, can be stated in formal terms. There are two different kinds of theories that human beings tend to use. There are the interpersonal espoused theories people hold about how they behave toward self and others. These theories are stated by respondents with a relatively high degree of reliability.

2. Unfortunately, the espoused theories people report are poor predictors of their actual behavior, though most respondents tend to be unaware of this fact. More importantly, they also tend to be unaware of the theory that does predict their behavior, a theory that may be called their interpersonal theory-in-use.7

3. A model was constructed of the interpersonal theories-in-use that accounted for most (all but four cases) of the micro-theories generated by the population studied (see Table I). The model contains an internal logic. It suggests that all theories-in-use have (a) governing variables (value) that people attempt to satisfy, or more accurately, satisfice;8(b) actions that flow predictably from the governing variables; and (c) consequences of that action upon the environment, upon learning, and upon the degree of effectiveness of behavior.

4. The four governing variables of Model I are (a) achieve the purposes as I (the actor) perceive them ("Effective action means figuring out what you want to get done and then getting it done"); (b) maximize winning and minimize losing ("I strive to win; you can get an E for effort, but the winners of this world are those who hate losing"); (c) minimize eliciting negative feelings ("How to get A to do what we want him to do without upsetting him"); and (d) be rational and minimize emotionality ("Stay close to the facts and don't let people get into personalities").

5. People holding the four governing variables will tend to behave according to the following strategies. They will tend (a) to design and manage unilaterally the environment and tasks, (b) to behave in ways that, unilaterally protect themselves and make them invulnerable to losing, while trying (c) to decide unilaterally who in the environment needs protection, and protecting that person in order to minimize being accused of violating the governing variables (e.g., minimize eliciting negative feelings and maximize rationality).

6. The action strategies will tend to have predictable consequences: (a) the actor will tend to be experienced by others as defensive (e.g., inconsistent, competitive, controlling, fearful of being vulnerable); (b) the interpersonal relationships and group dynamics will tend to be defensive (e.g., there will be little additivity in the discussion, members will seek to capture air time, issues will not be worked through, conflicts will be ignored); (c) the norms that will develop will tend to be defensive (e.g., mistrust, lack of risk-taking, conformity, competitive win-lose dynamics); and (d) there will be low freedom of choice and internal commitment.

7. One consequence of the above on the learning processes is that there will tend to be little public testing of hypotheses about important issues (this could risk violating the governing variables); testing will tend to be private and covert (e.g., A attributes motivations to B and "tests" the attributions privately). Since private testing is subject to self-distortions, and since B may also be behaving in the same way toward A, there is a high probability that much self-sealing activity will occur. Finally, little learning will take place beyond what is permissible by the system. Learning will remain within the bounds of what is presently acceptable and already programmed (single-loop learning). Challenges to the steady state of the system, to the governing variables, to the behavioral strategies, etc., that could lead to significantly different learning (double-loop learning) will tend to be minimized.

8. The result of these activities will diminish problem-solving effectiveness in the present as well as in the long-run, especially for highly potent and threatening issues. The decreased effectiveness, if sensed by individuals, will not generally be discussed (that would require public discussion of double-loop issues). The members may redouble their efforts to protect themselves by adhering even more rigidly to the governing variables which would set off and reinforce the processes again.


To conclude, individuals, programmed by their society according to Model I, are in a symbiotic, mutually reinforcing relationship with the organizations and groups in which they exist, all creating and reinforcing a Model I world.

What does all this have to do with alternative schools? I shall hypothesize seven connections and present indirect evidence that make these hypotheses worthy of further consideration.

1. The traditional or established schools are designed in accordance with Model I concepts.

2. The teachers within these schools learn to teach and manage learning environments that are consonant with Model I requirements.

3. Administrators, teachers, and students are programmed with Model I theories-in-use.

4. The creators of alternative schools were correct in diagnosing schools as representing Model I institutions and concluding that they did not facilitate student learning and growth.

5. The creators' espoused theories for alternative schools were never matched by an effective set of theories-in-use for their schools.

6. The vacuum of effective theories-in-use was filled by the Model I interpersonal theories-in-use with which the creators were programmed.

7. The creators were able to remain blind to these deficiencies perhaps by maintaining a high degree of anger toward the established schools.


Alternative schools were created in most cases by concerned educators, students, and parents who were bound together by a consensus that the present schools were ineffective. Most children, they believed, stayed away from schools because the schools tended to be dull, rigid, and bureaucratic, focusing on outdated ideas, with outdated learning processes, managed by equally rigid and outdated teachers and administrators.9 One major theme in all the writings was an anger toward the established schools.10

A second major theme has been described as "organic growth": students, freed from the oppressive restrictions of traditional schools, would naturally evolve a new learning community through open, honest, and trusting relationships.11


There are at least two assumptions implicit in these views. The major cause of student lack of involvement in their own education is related directly to the failure of the schools as human learning environments. Students and teachers brought together would naturally deal openly and honestly with each other to create the new, effective learning environments.

Two strategies flow from these assumptions. First, select teachers and students who are interested in being involved in designing their learning environments. The schools described in the three studies largely accomplished this objective.

Second, the new learning environments should be minimally oppressive; that is, they should strive:

1. To allow children the intrinsic satisfaction and delight of successful employment of their own intellectual and aesthetic energies at whatever level those energies are or can become capable of operating;

2. To encourage children to become increasingly self-motivated and increasingly responsible for their own learning and education—to make education as self-directed as possible;

3. To insure that children are able to cope with the world and competent to manage themselves;

4. To help children learn to be more skillful observers, classifiers, categorizers, perceivers, and interpreters of problems and to build more valid models;

5. To make teaching more productive.12

Such learning environments would require and encourage:

1. A close relationship based on mutual trust and understanding between student and staff.

2. Community decision-making actively shared by students, parents, and staff.

3. That human and physical resources of environment become a major resource for learning.

4. Reconsideration of traditional curricula and fundamental revision or elimination of grading procedures and age divisions.

5. The assumption by students of a major role in determining the nature and direction of their own learning.

6. The cooperation and respect of fellow students from diverse cultural backgrounds.13

Examining the assumptions, objectives, and characteristics of these new learning environments, several additional assumptions may be uncovered. They are:

1. Children have more growing and learning capacity than they are presently able to develop and to express.

2. Children, with help from competent professionals, can identify and develop this untapped capacity for growth and learning.

3. Children want to identify their learning capacities and actualize them.

These assumptions may be characterized as optimistic in that they suggest that most students are or want to be growth-oriented rather than deficiency-oriented. None of the schools developed ways to assess the degree to which the students were, in fact, growth-oriented. Three reasons for the failure may be inferred from the available literature. First, there was a strong belief, as already pointed out, that the major cause of student deficiency orientation was the school as an organization and learning environment, and not the personal characteristics of the students. Consequently, deficiency orientation could be decreased and growth orientation increased by designing schools with (a) minimal bureaucratic rules, (b) power shared equally between students and teachers, (c) minimal unilateral grading and rewarding procedures, and (d) a wide range of learning opportunities and paths for achieving those opportunities, hopefully designed jointly with the students.

The second reason for the optimistic view of students was that the teachers and students who first started free schools were indeed growth-oriented. They were intent on designing schools that fit their orientations. A third reason for optimism was based upon a defensive posture found in the early adherents to alternative schools included in this analysis. The sense of messianic commitment to change and the undifferentiated anger toward established schools may have had the effect of protecting these individuals from examining their own interpersonal incompetence and unresolved problems. To put it in terms of our theoretical framework, the people who created the alternative schools probably held espoused theories of growth, yet their theories-in-use approximated Model I. The anger toward the schools and the messianic ambience served to maintain their blindness to the fact that their espoused theories and theories-in-use were incongruent.


Examples abound of the contradictions and incompleteness in the theories used to create alternative schools. From the espoused theories described herein, it is possible to infer some of the governing variables of the theories-in-use of alternative schools. They are all related to maximizing students' and faculties' control and involvement over their learning environment. But such control and involvement, to be effective, must occur in a setting under real time conditions. A realistic theory would have to specify how to maximize student and faculty control under conditions of finite time and finite financial and other resources. But the theories used to design the alternative schools had little to say about facing the realities of managing systems under real time conditions. Indeed, the majority of the rhetoric was antiorganizational and antisystemic. There was an assumption that most, if not all, structures were bad and oppressive and that the elimination of such structures would enhance learning. The theories-in-use lacked, as we shall see, propositions about healthy organizations and propositions about what students would have to give up in order to create effective learning environments.

Turning to examples of the contradictions, how does a learning environment that ignores the legitimate requirements of systemic health teach children to cope with the world in which they are immersed? Isn't the world of the alternative schools, in which the child is to be immersed, a type of socially germ-free world because it is designed primarily around the (presumed) capacities of the students? How does living in a child-centered world help the child learn to alter effectively someday a world that is not child-centered? How will a student be helped to be an effective problem-solver if the world in which he lives has little relationship to the world in which he is to participate as an adult?

One response to these questions has been that it is the children who learn to live in genuinely free schools who learn to deal more effectively with the constrictions of real life institutions. Such a response may be challenged on two grounds. First, there are no data, known to the writer, that show that students who do well in alternative schools are more effective in the larger suppressive environment. There are some data to show that children who have been deprived (minorities) and frustrated (minorities and upper-class disenchanted students) and who have developed genuine anger toward society can be helped to reduce the anger and become more productive.14 One interpretation of these data is that, for these students, the society may have found a way to defuse them.

The second challenge relates to the fact, as will be shown below, that students and teachers, when given an opportunity to design genuinely free environments, created conditions that were minimally free and not particularly effective as measured by their own criteria. How may one expect students to be effective in changing the constraining "real" world when they create the same kind of world within their schools?

But why do teachers and students who espoused freedom, growth, honesty, truth, and cooperation create worlds that apparently rarely exhibit such qualities? Let us examine the actual behavior of the participants in order to understand their theories-in-use.


Unfortunately, very little direct or indirect data are available about the theories-in-use of the creators and participants of alternative schools. The literature is full of personal and technical espoused theories, but, as we have pointed out above, these are not the theories that will create the actual world of the alternative school.

There are, however, several sources of data available to help us to develop a tentative picture of the interpersonal theories-in-use that prevailed among the early pioneers of the movement. They are:

1. The Smith and Keith study of a new school included some actual verbatim transcripts of meetings held at the school. The scenarios represented meetings of teachers attempting to design their new school and meetings where they were attempting to implement their design.15 It is possible to study these scenarios and infer the theory-in-use of the participants. A scoring procedure exists that has been shown to be reliable and valid.16

The observational categories used in this research are graphically summarized in Table II. The categories above the zero line are hypothesized to facilitate interpersonal relationships, those below the line to inhibit interpersonal relationships. Each category has an idea (i) and a feeling (f) component. The categories positioned closest to the zero line are the easiest to perform; those farthest away are the most difficult.


For example, it is easier to own up to one's ideas or feelings (to express one's view and feelings) than it is to experiment with ideas or feelings (to discuss those ideas or feelings that, if wrong, would jeopardize one's self-acceptance). There are two levels of analyses. Level I represents the individual and interpersonal. Level II represents norms of the group. Every unit of behavior is scored on both levels.

For example:

Sample statement

1. I believe that we should reject the idea, even though we are told not to.

2. I feel tense.

3. Would you please tell me more about your theory?

4. This is not easy for me to talk about. I feel as if my whole life has been a shambles. I feel frightened and bewildered.

Would be scored as

own i

own f

open i

experimenting f

individuality i

individuality f

concern i

trust f

The scores represent a clustering of behavior that has been called Pattern A. Pattern A is almost identical to Model I, and it is a highly common behavioral pattern found in most problem-solving, decision-making, and implementation settings.17

own i

open i

conf i

conc i






n = 54 (percent)





These scores mean that people tended to own up to their ideas in such a way that their behavior contributed toward group norms of conformity and concern. Openness to ideas was relatively low. There were no feeling scores, no scores indicating risk-taking, helping each other to own up, be open, and take risks, nor were there any scores indicating the existence of norms such as individuality and trust.

Such empirical patterns of behavior have been shown to lead to little public testing of views, little openness to being confronted, little double-loop learning, and little problem-solving effectiveness.18

2. Protocols of difficult and easy interventions have been collected from 120 school administrators and twenty-six graduate students representing seven schools of education (many of whom had participated in alternative schools) and fifty graduate students in education from Harvard. All the cases were analyzed, and all were of the Model I variety. Again, this does not constitute direct proof that any of the participants in the alternative schools described in this paper held Model I theories-in-use. It does indicate, we believe, that along with the previous data that there is some probability that this is the case.

3. The Gross et al. and the Smith and Keith studies provided detailed evidence that directors were a primary force in the respective schools. In both cases the teachers began to realize relatively early that (a) the directors did not have a clear idea of the school nor how to achieve it in reality and (b) that whatever the teachers and students were doing, it was not working. Yet, and these data speak to the issue of Model I, the teachers did not feel free to raise the issue and publicly test their doubts about the director's clarity and vision. "It's his baby and you can't tell a person that his baby is no good."19 "This was (the director's) philosophy; this is what he believed. . . . He knew where he wanted to go; he had a diagram of the room, but he didn't know how to get there."20 Thus we have the director unilaterally designing the environment; the others attempting to behave according to his design, even though they did not understand it and did not think that the director understood it; and finally, the others feeling little freedom to surface their questions, to test publicly their doubts with the respective directors (all Model I characteristics).

It may be worthwhile to ask why the teachers, in these two cases, permitted themselves to live in a behavioral world vis-a-vis their respective superiors that was not consonant with the world that they espoused and was antithetical to effective learning and growth. Why did they choose to live partially in a world that is like the one they wish to overthrow?

Unfortunately, no direct evidence is presented in either case that can answer the question. All we may do is guess that since the teachers' interpersonal theories-in-use are Model I, they may tolerate this inconsistency because (a) to confront it openly would violate the governing variables of Model I and (b) they probably have not developed the capacities or skills to confront power people effectively. Nor would they tend to feel that they were in a world in which the systemic norms encouraged such confrontation of a superior. An additional possible explanation may be that the teachers attracted to these experimental programs (during the early days) could live with the tension involved in creating new schools if they "knew" that they could ultimately blame others for any failures, should they arise.


If the teachers' behavior was informed by Model I theories-in-use, and if, as we suggest, it is the holding of such theories that inhibits the confrontation of anyone, we should also predict that such inhibitions ought to operate between students and teachers. Teachers should report experiencing inconsistencies, yet they should not be observed confronting these inconsistencies in their relationships with each other and with the students. There were many examples in the literature to illustrate this prediction:

1. Students joined alternative schools because they saw traditional schools as unilaterally structured "against" them, giving the power to the teachers and administrators. Yet, in the alternative schools described in these studies, students preferred to bring the problems to the attention of the teaching staff and give them the responsibility to develop solutions.

Student: The way you got to do it is to make decisions. Then, if we don't like it, we'll let you know. You do something and we'll react.21

This stance makes students reactive, pawns rather than origins, and it reduces their personal responsibility. The students are apparently designing a world that is similar to the old one (the teachers are making the decisions) and then ordering the teachers to behave according to this Model I conception.

2. Apparently students do not permit the inconsistency above to remain as naked as we have just described it. They add another aspect to the design. It is the personal ethic of "do your own thing" or "hang loose." The ideal government is none at all; only "people dealing with people."22

Student: We're going to have a beautiful anarchy. Everybody's going to do their own thing and leave everybody else alone. We decided we don't need a government.23

There are several inconsistencies in this stance. First, the students have opted for a government in which the teachers make decisions and then the students may unilaterally reject or alter them. In doing so, the students are not asking for anarchy. They are asking for a Model I world that alternatively gives the unilateral power to the teachers to make the decisions. If the students do not like the decisions, then they have the power to reject the teachers' decisions (thereby giving the power to the students). This may be described as a world full of oscillating Model I relationships.

The second, perhaps more profound, inconsistency is that no state of human affairs exists where there are "no rules, only people dealing with people." According to our framework, all people have interpersonal theories-in-use. These theories are full of rules for behavior; they create behavioral worlds; and these behavioral worlds are constrictive. If students were helped to face reality, they would be helped to see that "people dealing with people" probably means people dealing with people using Model I behaviors, and therefore creating Model I learning environments.

3. Students, we are told, have learned to mistrust student government. Their concept of government holds that someone in a power position pushes around someone in a subordinate position. The someone in power could be teachers or students on the council. Consequently, "All government is some guy going around telling you what to do. There are people here nobody is going to force them to do anything."24

Yet the same students were ordering teachers to make unilateral decisions and then were reserving the right for themselves to reject unilaterally these decisions (which is making decisions). So we have learning environments that apparently permit students to hold espoused theories of government that they violate with their interpersonal theories-in-use. If membership in new schools is itself a source of learning, then what are they teaching students?

Moreover, the data show that politically concerned students cited national government as an example of unfair and unilateral government that ought to be stopped. The faculty did not make a major point of asking students to examine the fact that they were creating unilateral government in their schools and simultaneously developing a theory-in-use to keep them blind to this fact. Apparently enough students behaved according to these views that norms existed in the learning environment that "called into question the motives of those students who wished to form a representative government. Thus students who privately admitted that they were extremely interested in participating in a student government were reluctant to come forward for fear of being branded as power-hungry by their peers."25 As Model I predicts, such worlds develop norms to prevent the public testing of views that might signal the inconsistencies and the self-sealing processes. These community norms are, in effect, rules that govern behavior. Again we see that students' views not to create rules is part of an espoused theory that is incongruent with the theory-in-use. Also, as predicted by Model I, if people committed to Model I are given an opportunity to design a world, they will create one in which the norms are defensive, where they will tend to sanction self-sealing processes and inhibit the public testing of hypotheses that raise basic issues about the effectiveness of the system.

4. The administrative staff of the board of education laughed at the plan for involving students in decision-making because they believed it was unworkable.26 The principal, staff, and students believed that this indicated that the school bureaucracy did not share Metro's vision of the capacities of adolescents.

Principal to staff: I showed the plan for the administrative board to the district superintendent. She couldn't stop laughing. Teachers and students have equal votes. She showed it all around the office.27

The district superintendent's disbelief was confirmed by the behavior of the students and teachers. The former refused to take on their responsibilities. They shaped administrative processes that not only were authoritarian but then clothed them with the hang-loose ethic and the espoused beliefs about beautiful anarchies. The latter, by not confronting these issues, colluded in the facade and deception. One could argue that the overt laugh of the district superintendent was more congruent and genuine (not necessarily more effective) behavior than the self-deception that had become part of the relationships in the free school.

5. The "school without walls" put students in contact with the outside world that gave limited rights to young people. These experiences discouraged the students and confirmed to them that the outside world was rigid. Apparently, no attempts were made by the faculty and students to learn that they had probably created inside the school a world that was equally rigid. For example, if we follow the teachers' theories-in-use, it tended to be, on the one hand, to strive for total community participation, yet, on the other hand, not to confront the students with their refusal to meet their responsibilities. The students espoused that society was sick because it placed some people (students) in dependent, submissive settings where they were manipulated by the more powerful (staff). Yet the students created a world where the teachers became dependent and submissive and there to be manipulated by the students.28


To date, our argument has been that the technical espoused theories of alternative schools, although understandably primitive, were congruent with the personal espoused theories of the early participants. As has been true in the case of all other populations studied so far, each participant tended (a) to manifest an interpersonal theory-in-use that approximated Model I, (b) to be blind to the incongruity between his or her espoused theory and theory-in-use, (c) to be more aware of the discrepancy in others' espoused theories and theories-in-use, but given the requirements of their theory-in-use, (d) they would not tend to help others see the incongruity, and consequently (e) others would not help them to see their own incongruity.

Such a self-sealing, increasingly closed living system should lead to an increase in the frustration of the participants, an increase in the feeling of failure, and an increase in feelings of withdrawal from the involved management of the system. These reactions, in turn, would simultaneously create an increase in the feelings of failure, frustration, and withdrawal as well as an increase, by each teacher, to his or her classroom experience, and an increase by each student of feelings of worrying primarily about himself or herself.

Again, we are unable to present data from research designed to test directly these predictions. All that is available are the anthropological reports of the several studies that we have cited above. These four illustrations are almost directly reproduced from the studies cited below.

1. Meetings manifested the dynamics, such as, unresolved personality clashes, that sometimes led to obfuscation of issues, aimless group discussion, confusion about what, if any, decisions were reached, responsibility neither clearly understood nor effectively carried out, and hesitation to confront people who violated their own agreements.29

2. The committed teachers spent considerable time and energy in listening to students, in attempting to fulfill their needs, in satisfying their complaints, and in pleading continuously for participation. Since the commitment for genuine, mutual involvement on the part of the students did not exist, the teachers became harried; indeed, the researcher reported that teacher exhaustion was pervasive. Eventually, the energy expended and the exhaustion led to self-protection, and teachers began to withdraw. Some still pleaded, "Last year, I put up signs and signs and no one showed. I guess they're not interested."30 Finally, this teacher, as well as others, gave up, withdrew, and became alienated from their own system. They worked hard to do their own thing in their classrooms,31 and some even began to hang loose. The teachers had traveled from deep commitments, enthusiastic work, high energy consumptions, to withdrawal, low energy commitment, doing their thing, and becoming as un-involved as the students were in the first place.

3. The concept of organic development, when subject to these harsh realities, has generated a similar pattern of events in one alternative school after another. The school is started in an atmosphere of high energy and good will. The general commitment to a more humane way of operating, the high level of personal dedication, and the good feeling that permeates any new enterprise carry it through a honeymoon period of six months to a year. The positive experience of the honeymoon period sustains the belief that just about any problem—student involvement in decision-making, race relations, moderately severe mental disturbance, development of a relevant, curriculum—can be solved in a free and open atmosphere with a strongly articulated commitment to interpersonal honesty.

As the honeymoon draws to a close, evidence begins to accumulate that people really haven't changed as much as was hoped. The all-school meeting fails. The school's tape recorders, which people used to be able to leave out, begin to disappear. The first interracial fight occurs. People begin to notice that, although whites and blacks are outwardly polite to each other, there is little communication and friendship. Cliques are mostly all-white or all-black. Severe interpersonal conflicts between strong-willed staff members surface, and their conflicts spill over into other issues debated in the school. Someone stuffs a roll of toilet paper into the toilet to make it overflow. A window is broken. Some kids consistently fail to follow through on their class commitments and other learning experiences; and since these kids have had a year to get themselves together, some people wonder whether the alternative school is doing any more for them than the old school. Community or staff meetings are held, and strongly felt resolutions are passed. But in practice both staff and students find it extremely difficult to confront individuals who don't abide by these resolutions, who persist in "doing their own thing."32

4. In the Gross et al. study, six of the ten teachers altered their behavior as time passed (four did not). All six reported changes of greater control of their classes and less open-ended activity. The others who did not change their behavior continued to conduct highly structured classes.33


The analysis, to date, suggests that an important cause of the ineffectiveness of alternative schools existed within the schools. The reader may ask, could not the causes for the demise of free schools exist outside the system, such as political, economic, and legal factors?

There are two answers inferable from the theoretical framework. The rhetoric of alternative schools is to create new learning environments where risky and threatening questions are asked of society. A world populated with Model I people will tend to see such innovations as hostile. Consequently, one would predict that economic, political, and legal pressures might indeed be used to inhibit the progress of these schools. We would not be surprised, therefore, to find alternative schools that have failed because of forces external to them. Our prediction, however, would be that such pressures were actually a response of the frightened establishment to the espoused theories of the creators and the participants. All the establishment would have had to do to bring the demise of these schools would be to let them alone; their creators and participants would tend to destroy them because, according to this analysis, they hold theories-in-use that will not permit the schools to succeed.

The alternative schools studied by the Center for New Schools, by Gross et al., and by Smith and Keith had adequate financial support; enthusiastic teachers, students, and administrators; and supportive parents and citizens. Yet these schools did not achieve their objectives. Indeed, as the research of these activities was ending, all three were beginning to "regress" to Model I learning environments.

The hard-hats who tend to see alternative schools as the creations of malcontents, people against capitalism, and people who basically cannot make it in the win-lose competitive world predict failure of the schools because such places produce ideas that are "woolly," or "pie-in-the-sky" dreams, or fantasies to "baby" children. Their prediction of failure is correct, but not their reasoning. Failure occurs because the creators of alternative schools behave in accordance with Model I. Rhetoric aside, the creators hold theories-in-use that are similar to those held by the hard-hats.

Interestingly, as the failure of alternative schools to achieve their objectives became increasingly evident, the basically authoritarian theories-in-use of their creators surfaced when they were forced to make explicit how they would govern such schools.

For example, Kozol condemns traditional schools for their involvement in extending the values of the established culture, thereby making it difficult for them to raise serious doubts about the indoctrinational and custodial functions of the public education apparatus.34 However, in designing his new schools, Kozol cautions against widespread participation by the students and teachers. He recommends that "a small, benevolent dictatorship is, to be quite blunt, a remarkably good and reasonable way to govern a small school."35 A small, but honest, power structure is recommended, but no model, or evidence that such a power structure is not corruptible, is given. Given his Model I design, the implication is that, if the leaders possess motives as clean and genuine as Kozol's, they are not corruptible. We hope that we have made the point that motivation is not the only important factor. Competence to design and behave in significantly new ways is required.

Why do alternative schools require benign dictatorships? Kozol suggests:

I have never yet seen a situation of this kind where the consequent disputation did not demean and undermine the character of those who were involved and when it did not also plant the seeds of future decimation.36

And later:

Also, some cases where a group packed the votes and went about the work of turning the school into an image of its own conceptions.37

But this is precisely what Kozol is doing. He is designing a school according to his own conception. True, there may be no attempt to pack the votes, but that is unnecessary because he has placed himself in control through the power structure. Model I is not only consonant with Kozol's description above, indeed it makes Kozol's findings understandable and predictable. People with Model I theories-in-use will tend to design Model I schools and Model I worlds.


One might conjecture that as long as attention is directed toward the environment as the culprit, people do not have to explore the similarities between their theories-in-use and those of their ideological enemies. The analyses that condemn the larger society for the demise of alternative schools may therefore tend to distract people's attention from (a) their personal causality in designing and managing schools in ways that would bring about their demise, and (b) the possible new options that people could explore to re-educate themselves in the theories-in-use needed to make alternative schools successful. One example is the recent work of Graubard to which we have already alluded.38 Graubard presents several research studies that conclude that schools acculturate the child to become a passive, dependent person who competes with other children for scarce rewards, who may form cliques to support his or her competitiveness, and who focuses primarily on learning those intellectual and interpersonal skills that are salable in the capitalistic system. It is the political, social, and economics norms of the society that are the fundamental causes of these classroom behaviors.

Another example is the interesting work by the radical economists focusing upon education.39 When speaking as radical reformers, they view schools as the handmaiden of the free enterprise capitalistic system. Schools, they maintain, more than educate children in the three Rs. They prepare children to find their "niches" in the organizational hierarchy within the capitalistic system] For example, they prepare students to be appropriately dependent and docile toward authority, respectful of discipline, and disinclined toward making waves and upsetting applecarts.

Schools probably do all these things; indeed, it would be predictable from the model presented. Schools populated with Model I teachers and students will tend to teach Model I values and use Model I learning strategies. However, this prediction is not limited to the capitalistic system. It would be made about any school where the governing variables and action strategies approximated Model I. Thus one would expect to find this behavior not only within profit-oriented institutions, such as, factories and businesses, but also in schools, in scout troops, in church organizations, and in communes (of the kind that survive).40 Also one would find it in socialist states such as Russia and China as well as those like Finland and Sweden.41

The radical economists' view that it is the economic system that leads to Model I behavior, values, and institutions is, I believe, not complete or accurate. More important causes than the economic system teach the child to value winning, controlling, competing, suppressing emotions, and the behavioral skills congruent with these values. They are:

1. The organizational theories-in-use in the world. There is ample research now to show that the most varied economic systems known to man have the same organizational theories. The organizations, economic and noneconomic, in China, Cuba, Sweden, Russia, and the United States do not differ significantly in their genotypic properties, that is, in the thrust to specialize work, centralize power and control, and have information flow toward centers of power and control.42 Economists are finding out that the traditional pyramidal structure may not be economically efficient; indeed, the reason for its use may be to control people and economic goods.43

2. The nature of information processing of the human mind, no matter what its political persuasion. The mind, as an information processing mechanism, is apparently quite simple compared to the environment in which it exists. The mind deals with environmental complexity by creating concepts and hierarchies of concepts, and those hierarchies have properties that are similar to the hierarchies found in pyramidal organizations.44

The reader may argue that this is an alternative explanation of my position. If people have certain finite information processing capacities, then they may necessarily be limited in the innovations that are possible. I acknowledge this possibility but suggest that the research on information processing is based upon studying Model I behavior in Model I settings (real or contrived). Research is needed in other types of settings to help us understand the potential of human information processing activities. For example, dissonance theory can be shown to be derivable from a Model I world, but is not derivable from, let us call it, a Model II world. The latter is rarely studied because social scientists have conceived of their role as primarily describing reality, which, it is suggested, is a Model I universe.45

3. Given the first two factors, then people become programmed with the engineering-microeconomic (cost-benefit) principles that have dominated organizational design throughout the world, in the public and private sectors. For example, when asked what quality of human relationships people would design if they wanted to achieve any objectives important to them and maintain a high quality of life (with or without concern about financial costs), over 85 percent of our respondents (n = 400) have described human relationships that are so similar to those endemic to the pyramidal structure that they have been called "pyramidal values." They are (a) the most important human relationships are those that achieve the objective (almost no mention of relationships to enhance the quality of life), (b) human behavior is effective when it is rational and ineffective when it becomes emotional, and (c) the most effective way to manage human behavior is through unilateral direction, control, and reward and penalties as well as competitive win-lose dynamics. The respondents provided these types of answers whether they were designing relationships for business, government, churches, voluntary organizations, the arts, hospitals, trade unions, student organizations, or schools of many varieties.46

There are no data, to my knowledge, that indicate that the capitalistic economic-political or socialistic system (to give two examples) create these "human programs." These human programs are taught to people because industrial cultures are controlled by concepts of organizational efficiency whose meanings are tangentially affected by notions of the distribution of income and the ownership of resources. To repeat, the pyramidal values are found where the issues of ownership of income and allocation of economic resources are resolved in opposite ways.47

Another reason why blaming society may be a less fruitful strategy than focusing on the behavior of human beings is that, in the long run, it may inhibit change. For example, Katz recently concluded that social revolution should start outside the schools because schools can do no more than mirror the society in which they are imbedded.48 If our analysis is correct, Model I individuals will tend to create Model I institutions of all kinds. Thus all institutions will mirror the society in which they are imbedded.

The prediction that follows from our perspective is that beginning such innovations in other societal organizations would not eliminate the problem because those participants would also be programmed with Model I theories-in-use. Moreover, if we had available new and workable societal designs of organizations that facilitated and encouraged human growth, such designs would meet with considerable difficulty because the participants would have trouble in behaving according to the new behavioral specifications. If these new designs are to succeed, individuals would have to be unfrozen from their Model I programs.

Another unfortunate consequence of not examining the theories-in-use of those proposing and carrying out educational innovations is that certain questionable tactics, parochial vision, and rigidity about research techniques are not adequately explored. For example, if the educational professionals who resist change are momentarily ignored (and admittedly that may rule out quite a few), the remaining may be characterized as follows. There are those who are strong on rhetoric about change in schools but are weak on implementation. The rhetoric may be angry or romantic, or as in the cases already described, romantic about children and angry about society. There is a curious phenomenon that may be characteristic of many of these individuals. On the one hand, they may have an idea, such as alternative schools, and they will unhesitantly try it put on children with a minimum of prethought, pre-evaluation of possible relevant literature, pilot programs, or on-going evaluative research designed to make a difference in the actual operation of these schools. On the other hand, most of these same people shout angrily against pharmaceutical manufacturers who market a product with inadequate research. In defending themselves, these people protest that they cannot conduct laboratory studies, that they must try out their ideas directly on children. Why is this necessarily the case? Why cannot these people experiment with themselves for a year or two? If this analysis is correct, they would have developed adequate data about their inability to create the learning environments that they valued, or at least they would have discovered some of the difficulties that were reported in the studies cited in this paper.

There is a second group that tends to react against the lack of rigorous research and theory by striving to surpass the rigor of the experimentalists. The difficulty with this group has been that they have tended to ignore the nature of the world implicit in and created with the use of the technology of rigorous research. It appears that rigorous research technology is basically a Model I technique.49 Thus the rigorous people not only lose vigor and relevance,50 but actually may create knowledge on the basis of a Model I technology, and therefore, may be limited in producing a Model I vision of the world. A famous example of this outcome may be Skinner.51

Next is the position, illustrated by Gross et al., that educational innovations may lack clear, comprehensive maps of the objectives of the innovations, as well as administrative clarity about the processes by which these innovations are to be implemented. There is validity to the view of Gross et al. that it helps to achieve a goal if one can define the goal precisely. But for two reasons this view is difficult to apply to innovations striving toward a genuinely new world view and options. First, if one could spell out an innovation ahead of time, the question arises as to how much of an innovation it would be. Second, if one can spell it out, then the tendency will be to teach it to others (a strategy that Gross et al. explicitly recommend). Such an action will tend to create Model I conditions and will be at variance with the intended innovations of the new schools.

The challenge for the designers may be not so much to be perfectly clear ahead of time about their innovation, but about the learning processes necessary so that faculty and students will learn from their experimentation. A concept of on-line, real time learning is needed, by which teachers and students can collect and process information about their experiences and share them in such a way that corrective actions can be suggested within real time constraints and knowledge can become additive to the participants and eventually to others interested in the experiment. Such a concept requires a different set of human relations and organizational structural arrangement.52

On the other hand, Gross et al. are correct when they state that much effort is probably needed in training teachers and students if organizational innovations such as alternative schools are to be effective. To date, I know of no published material that suggests that such learning can be acquired in less than six months' full-time effort or two years' part-time effort with periodic help (one or two days a week) from expert consultants.53

Finally, the analysis also sheds light on the recent attacks on teachers for their middle-class values. Such criticism is harmful for many reasons, not least because it inhibits educational innovation. For example, Henry suggests that teachers are biased against children who are dirty, noisy, unruly, and confronting, and they react by becoming authoritarian and prejudiced.54 He explains this behavior by saying that teachers are primarily middle-class and value cleanliness, quiet, discipline, courtesy, etc.

The assertion that teachers value these qualities is probably true. But to explain this by asserting that teachers hold middle-class values is, at best, to take a variable that is distant in the causal chain. An alternative explanation, which follows from our framework, is that teachers, like most others in our society (black, white; lower-, middle-, and upper-class), are finite information-processing systems. School classes can easily become an information overload for teachers. After experiencing the exhausting consequences of overload, teachers may try to place constraints on the amount of information that they are bombarded with and are expected to process. They may focus, understandably, on such factors as discipline, neatness, quietness, etc., because these are valid ways to reduce the overload and, therefore, to make their lives more livable. To translate this into our framework, teachers programmed with Model I, when confronted with overload, can either withdraw (which would increase the overload) or become Model I2.

The difficulties with middle-class as an explanatory concept are that (a) the concept misses the point that the lower-, middle-, and upper-class are finite human beings, (b) the lower-, middle-, and upper-class are programmed with Model I behavior, and (c) the concept provides the teachers with little insight into what to do to change their behavior. Even if their behavior was largely related to middle-class values, then what is the strategy for change? Make them upper-class? Lower-class? The concept may be useful to anthropologists and sociologists who have built theories with it. The concept is not very helpful in bringing about effective change.


Individuals who are interested in creating innovations with objectives such as those described at the outset should first become aware of their espoused theories and their theories-in-use. They should also become aware of inconsistencies within their theories and incongruities between their espoused theories and their theories-in-use.

Next, there is a need for them to learn and to behave according to a theory-in-use that would serve to facilitate the human growth and learning espoused by alternative school theory. Donald Schon and I have attempted to develop such a theory. We call it Model II.55 The theory has increased effectiveness as its objective; it encourages learning that questions the status quo (double-loop learning), public testing of hypotheses (instead of private testing), and people learning through action.

We have also begun to experiment with experiential learning environments that help people move from Model I theories-in-use toward Model II. Although it is too early to be certain about the nature of these learning environments, we have attempted to explicate a model of the transition process, as well as to provide empirical illustrations from our experimentation.

As people are able to behave according to Model II, which means they would have developed a new set of values and skills, they can use these values and competences to begin to design the new learning systems. Effective interpersonal skills, however, are not enough. New types of organizational structures and task requirements will have to be evolved. These structures and other systemic properties will have to be designed in accordance with the criteria provided by Model II. Little is known about how such structures can be designed because few experiments have been attempted. Much research is needed in this area.

However, the important first step is for individuals to learn to behave according to Model II. With these skills and competences, it should be possible for the participants to work effectively toward building new schools that achieve the intended purposes and that monitor themselves effectively. In other words, given the internalization of Model II by the participants, one has the genes for designing and redesigning until an effective learning system is created.

1 Allen Graubard, "The Free School Movement," Harvard Educational Review, Vol. 42, No. 3, August 1972, p. 352.

2 Chris Argyris. Integrating the Individual and the Organization. New York: John Wiley, 1964; Chris Argyris, "The Incompleteness of Social Psychological Theory," American Psychologist, Vol. 24, No. 10, October 1969, pp. 898-908; and Chris Argyris, "On Beyond Freedom and Dignity," Harvard Educational Review, Vol. 41, November 1971, pp. 550-567.

3 Center for New Schools, "Strengthening Alternative High Schools," Harvard Educational Review, Vol. 42, No. 3, August 1972, pp. 313-350; Neal Gross, Joseph Giacquinta, and Marilyn Bernstein. Implementing Organizational Innovations. New York: Basic Books, 1971; and Louis M. Smith and Pat M. Keith. Anatomy of an Educational Innovation. New York: John Wiley, 1971.

4 Chris Argyris. Organizations of the Future. Beverly Hills, Calif.: Sage Publications, 1973.

5 Argyris, "The Incompleteness of Social Psychological Theory," op. cit,; Chris Argyris. Intervention Theory and Method. Reading, Mass.: Addison-Wesley, 1970.

6 Chris Argyris and Don Schon. Action, Theory, Practice, in press 1974.

7 Professional fields, such as, education, law, divinity, and business, are composed of clusters of subjects. Each subject has an espoused theory and a theory-in-use. The former may be called technical espoused theories, while the latter, technical theories-in-use. There exist, in the different professions, varying degrees of incongruity between the technical espoused theory and the theory-in-use. In the field of education, the incongruity appears to be high. Argyris and Schon, op. cit.

8 Herbert A. Simon. The Sciences of the Artificial. Cambridge, Mass.: M.I.T. Press, 1969.

9 Graubard, "The Free School Movement," op. cit.; Gross et al., op. cit.; and Allen Graubard. Free the Children. New York: Pantheon Books, 1972.

10 Graubard, Free the Children, op. cit.; Jonathan Kozol. Free Schools. New York: Bantam Books, 1972; and Graubard, The Free School Movement," op. cit.

11 Center for New Schools, op. cit., p. 336.

12 Gross et al., op. cit., p. 11.

13 Center for New Schools, op. cit., pp. 334-335.

14 Philadelphia's Parkway Program: An Evaluation. Organization for Social and Technical Innovation, Newton, Massachusetts, April 1972.

15 Smith and Keith, op. cit., pp. 245 and 250.

16 Chris Argyris. Organization and Innovation. Homewood, 111,: R.D. Irwin & Co., 1965; Chris Argyris, "Some Unintended Consequences of Rigorous Research," Psychological Bulletin, Vol. 70, 1968, pp. 185-197; Argyris, Intervention Theory and Method, op. cit.; and Argyris, "On Beyond Freedom and Dignity," op. cit.

17 Argyris, "The Incompleteness of Social Psychological Theory," op. cit.

18 Chris Argyris. Interpersonal Competence and Organizational Effectiveness. Homewood, Ill.: R. D. Irwin & Co., 1962; Argyris, Organization and Innovation, op. cit.; Argyris, Intervention Theory and Method, op. cit.; and Argyris, "On Beyond Freedom and Dignity," op. cit.

19 Smith and Keith, op. cit., p, 157.

20 Smith and Keith, op. cit., p, 157.

21 Center for New Schools, op. cit., p. 318.

22 Ibid., p. 319.

23 Ibid.

24 Ibid., pp. 318-19.

25 Ibid., p. 320.

26 Ibid., p. 324.

27 Ibid.

28 Philadelphia's Parkway Program: An Evaluation, op. cit.

29 Center for New Schools, op. cit., pp. 321-322.

30 Ibid.,p. 326.

31 Smith and Keith, op. cit., p. 157; Gross et al., op. cit.

32 Center for New Schools, op. cit., pp. 336-337.

33 Gross et al., op. cit., p. 160.

Cite This Article as: Teachers College Record Volume 75 Number 4, 1974, p. 429-452
https://www.tcrecord.org ID Number: 1454, Date Accessed: 1/16/2022 5:57:34 PM

Purchase Reprint Rights for this article or review
Member Center
In Print
This Month's Issue