<#include virtual="/common.html">
Clifford IT Services OD Pages (continued)

Argyris, Schon, and Action Science

Chris Argyris

Sources of information on this page:


Action Science; Concepts, Methods, and Skills for Research and Intervention.

Chris Argyris, Robert Putnam, Diana McLain Smith.
Jossey-Bass Publishers, San Francisco, 1985.

Theory of Action (Chris Argyris and Donald Schön)

The theory of action approach was developed by Chris Argyris and Donald Schön. Its premise is that human beings design action to achieve results they intend.

Acting skillfully shows know-how or tacit knowledge that we can represent as a theory of action.

To learn, on this view, is to become able to produce the learning in action. This leads to distinguishing espoused theories, or theories of action that we believe we follow, from the theories-in-use that would be inferred from our actual behavior. We are aware of our espoused theories, but often unaware of our theories-in-use. The two are frequently inconsistent, notably in situations that trigger embarrassment or threat.

We seem to be socialized in a particular theory-in-use, called model I or the unilateral control model, that comes out in these situations. Model I leads to low trust, low commitment, and limited learning.

Argyris and Schön described an alternative theory-in-use, model II, for mutual learning on issues of fundamental importance. Developing competence in model II enables people to learn in the midst of difficult circumstances and to act as agents of organizational learning.

Argyris and Schön published Organizational Learning in 1978, more than a decade before the idea of learning organizations became popular.

Robert W. Putnam, Diana McLain Smith, and Philip McArthur of Action Design were doctoral students at Harvard and Teaching Fellows with Argyris. In 1985 Chris, Bob, and Diana co-authored Action Science.

Action science is a form of inquiry into how we design action and how we might create better organizations. It is concerned with practical knowledge for the conduct of human affairs. It proceeds by helping people reflect on and improve social practices that shape inquiry, choice, and action.

Argyris has focused on the reasoning processes individuals use to design and implement action, highlighting differences between productive and defensive reasoning. He has described how individuals create and maintain defensive routines in organizations and how to overcome those routines.

Donald Schön developed a theory of knowledge for action by reflecting on the performances of master practitioners. In the ordinary form of practical knowledge we do not think about what we are doing. But sometimes we do, especially when we are puzzled or surprised. Schön named this reflecting-in-action, and argued that it is central to our ability to act effectively in unique, ambiguous, or divergent situations. He also pointed out that our most difficult problems are characterized by clashing frames or definitions of the situation held by different parties. Addressing these problems requires a capability for frame reflection, for understanding and bridging different perspectives.

A short outline of Model I versus Model II:

Model I Theory-in-Use

Governing Variables:

Define goals and try to achieve them (unilaterally).

Maximize winning and minimize losing.

Minimize expressing or generating negative feelings.

Be rational and minimize emotionality.

Action Strategies:

Design, manage, and plan unilaterally.

Own and control the task.

Unilaterally protect self and others.

Evaluate others in ways that do not encourage testing the validity of the evaluation.




Interpersonal Manipulation

Self-serving action



Mis-use of power

Low freedom of choice

Low internal commitment

Low risk-taking


"Self-sealing, single-loop" learning


Little public testing of notions about why others behave as they do, what they need, etc.

Decreased effectiveness

Model II Theory-in-Use

Governing Variables:


Maximize valid information.

Have free and informed choice for all concerned.

Have high internal commitment to the choice and constant monitoring of its implementation.

Action Strategies:

Design situations where participants can originate actions and can experience

high personal causation and success.

Jointly control tasks.

Make protection of self and others a joint enterprise.

Craft positions or behaviors into action strategies that openly illustrate how the actors

reached their evaluations or attributions, and how they crafted them to encourage inquiry

and testing by others.


Minimally defensive interpersonal relations




High individuality

Open confrontation on difficult issues

High freedom of choice.

"Double-loop" learning (includes questioning of goals)

Processes can be disconfirmed

Public testing of theories and attributions.

Increased quality of life

Effective problem solving and decision making, especially for difficult issues

Increased long-run effectiveness.

Some good links:

The Action Science Network
Chris Argyris - Organisations@Onepine
Action research resource papers
Chris Argyris at infed.org
Rex C. Mitchell, Ph.D.

Back to OD Page