Theory of Action

Chris Argyris and Donald Schon’s theory of action is a descriptive and normative framework that explains and prescribes behavior at the individual, group, and organizational levels. The intellectual roots of the theory of action are John Dewey’s theory of inquiry and Kurt Lewin’s formulations of action research. In particular, the theory of action aspires to the Lewinian ideal of contributing simultaneously to basic knowledge of human behavior and practical action in everyday life. In so doing, the theory of action integrates science and application to an extent that is unparalleled in the organizational behavior literature.

Espoused Theory versus Theory-In-Use

At its core, the theory of action maintains that, for virtually everyone, there is a discrepancy between what people say and believe is motivating their actions and what is actually motivating their actions. The former is termed espoused theory, and the latter is termed theory-in-use. In other words, there is a gap in awareness between the explanations people have for their own actions (espoused theory) and the cognitive structures that actually govern their actions (theory-in-use). This gap exists not only at the individual level, but at the group and organizational levels, as well (i.e., the cognitive structures that govern individual behavior give rise to interpersonal structures that regulate group behavior).

Theory-in-use must be inferred from people’s actual behavior and not from their descriptions of that behavior. Over decades of research with thousands of participants from a wide variety of cultures, Argyris and Schon have found overwhelming evidence of an implicit cognitive structure, or theory-in-use, that is common to most everyone; the authors refer to this theory-in-use as Model I. Model I is the result of socialization early in life. Specifically, from an early age, virtually all people in modern industrial societies are socialized to (a) individually define the task at hand and the purposes to be achieved, rather than work interdependently to develop mutual definitions of task and purpose; (b) maximize winning and minimize losing; (c) suppress negative feelings; and (d) be rational and minimize emotionality. These socialized tendencies are referred to as the governing variables of Model I.

Model I socialization carries a behavioral imperative in which the underlying strategy is unilateral control over others and the environment. Based on their extensive research, Argyris and Schon concluded that people vary greatly in the way they attempt to control others and the environment but that the attempt to do so is nearly invariant. Because this behavioral strategy does not produce valid feedback from others, it leads individuals to be defensive and closed. At the group and organizational levels, this strategy leads to defensive relationships that reduce the production of valid information and reduce free choice among organizational members. In general, the consequences of Model I behavior in organizations are poor decision making, low commitment, wasted resources, unproductive conflict, and limited learning/change on the part of organizational members.

Automatic Nature of Model I Actions

According to Argyris and Schon, most people are unaware of the fact that their theory-in-use conforms to Model I. This means that the Model I strategy of unilateral control tends to be highly automatic (in the sense that it operates outside of conscious awareness). In fact, not only are implementations of this strategy automatic, they are often very sophisticated. The difficulty is that most people have little awareness of how and when they implement this strategy. Consequently, people’s actions tend to remain consistent with the strategy of unilateral control—even when they say and believe otherwise. Especially when facing difficult human relations problems, people often unknowingly act in ways that are inconsistent with their words. That is, on the surface people may know—and espouse to others (i.e., espoused theory)—that unilateral control is a counterproductive strategy when attempting to resolve such problems, yet when they themselves are immersed in such a problem, they blindly implement this very strategy to some degree. Moreover, the higher the stakes, and especially in the midst of stress, threat, or embarrassment, the more strongly the strategy of unilateral control is activated and the more it interferes with the ability to work effectively with others.

According to Argyris and Schon, this automatic Model I programming is the primary source of the toughest and most persistent problems of organizational behavior (e.g., low morale, withdrawal from work, poor group decision making and problem solving, dysfunctional behavior in teams, employee-management strife, and ineffective leadership). Thus, to begin resolving these problems, organizational members must first become aware of their Model I programming and the ways in which it causes organizational problems. Then, after gaining this awareness, the existing program must be unlearned over time and replaced with a more useful and self-aware action model. Argyris and Schon advocate a replacement model they simply call Model II, and the primary focus of their work has been to disseminate this model while helping others learn how to assimilate and practice it.

Model II

The governing variables of Model II are to (a) maximize the use of valid information for solving problems, (b) maximize free and informed choice in solving problems, and (c) maximize internal commitment to problem solutions and the monitoring of solutions over time. In contrast to the Model I strategy of unilateral control, Model II requires mutual control if its principles are to be realized. Therefore, the action strategies of Model II involve creating shared purposes, expressing one’s own views openly while sharing the reasoning behind those views, inviting challenge from others while inquiring into one another’s views, designing ways to publicly test differences in views, and holding one another accountable. Not surprisingly, the organizational consequences of Model II are very different from those of Model I. These consequences are effective decision making, high commitment, faster adaptation to change, strong working relationships (characterized by high trust and openness), and mutual learning.

The transition from Model I to Model II requires what Argyris and Schon call double-loop, as opposed to single-loop, learning. Single-loop learning occurs when an individual learns new actions that are consistent with the core principles of his or her operative action model (e.g., Model I). Double-loop learning, by contrast, involves learning new core principles (e.g., Model II) and new actions that are consistent with those new principles. In numerous longitudinal studies, Argyris and Schon have found that the transition from Model I to Model II is generally neither fast nor easy, even for people who are highly committed to making the transition, because Model I actions tend to be highly automatic and deeply ingrained.

Practicing the Theory Of Action

The theory of action is unique because it is both a theory and a form of practice. As a form of practice, the theory of action has two key features: It is practiced both publicly and in real time. That is, groups of practitioners are brought together by a theory of action interventionist to inquire openly into their own work behavior and to identify whether Model I principles may be motivating their behavior and inhibiting organizational effectiveness. If Model I is found to be counterproductively operative, the interventionist then coaches the participants to behave consistently with Model II. With enough coaching and practice, the participants eventually learn to practice Model II on their own while becoming less and less dependent on the interventionist.

The theory of action approach stands in sharp contrast to the mainstream approach in organizational science. In the mainstream approach, organizational behavior is treated as an object of study separate from individuals’ immediate actions. The goal of the mainstream approach is to learn as much as possible about this object of study and create a bookshelf of knowledge from which practitioners can presumably draw for guidance. Argyris and Schon essentially argue that, although this mainstream approach has generated many noteworthy findings, the separateness of those findings inevitably limits their applicability. In their theory of action approach, the generation of bookshelf knowledge is a secondary goal. The primary goal is to generate firsthand, actionable knowledge for the practitioners involved—that is, knowledge of the automatic Model I actions being produced, how those actions lead to unintended and counterproductive outcomes, and alternative Model II actions that are more likely to resolve difficult organizational problems.


  1. Argyris, C. (1993). Knowledge for action: A guide to overcoming barriers to organizational change. San Francisco: Jossey-Bass.
  2. Argyris, C., & Schon, D. A. (1974). Theory in practice: Increasing professional effectiveness. San Francisco: Jossey-Bass.
  3. Argyris, C., & Schon, D. A. (1978). Organizational learning: A theory of action perspective. Reading, MA: Addison-Wesley.
  4. Argyris, C., & Schon, D. A. (1996). Organizational learning II: Theory, method, and practice. Reading, MA: Addison-Wesley.

See also: