Chris Argyris was born in Newark, New Jersey on 16 July 1923, the son of Stephan and Sophia Argyris. He grew up in Greece and New Jersey. According to Lundberg (1998), Argyris had in some respects a difficult childhood, particularly at school where he was a member of a minority group and had – initially – a limited command of English. This experience ‘instilled in him two enduring characteristics: a propensity to examine himself carefully to discover his deficiencies, and a desire to work hard to change himself’ (Lundberg 1998: 19). He served as an officer in the US Army Corps of Signals during the Second World War, and afterwards went to Clark University, from which he graduated with an AB in 1947. He completed his MA at Kansas University in 1949, and his Ph.D. at the School of Industrial and Labor Relations at Cornell University in 1951. He married Renee Brocoum in 1950; they have two children.
On completing his Ph.D., Argyris joined the faculty of Yale University. He served as director of research in labour from 1951 to 1954, associate professor from 1954 to 1959, and professor of business administration from 1960 to 1965; he was Beach Professor of Academic Services from 1965 to 1971. He then moved to Harvard University, where he has been James Bryant Conant Professor of Education and Organizational Behavior since 1971. He has held a number of other posts, including membership of the board of directors of National Training Laboratories and special consultant on human relations to the secretary of health, education and welfare, and has been associated with institutions as varied as the Ford Foundation, the National Institute of Mental Health, and the Air Force Personnel and Training Center.
The capacity for self-analysis which Argyris developed at an early age was soon extended to the study of others, both as individuals and groups. In his early work (Argyris 1957, 1960, 1962), he noticed what he termed a basic incongruence between the needs of individuals and the demands of organizations. Organizations, as they are usually structured, are hierarchical and control-centred. Communication is largely top-down. Managers, in guiding the efforts of those below them, impose strict limits on those efforts and on those people. Individuals, on the other hand, are independent, active, self-aware entities. In fact, as people grow more mature and wiser, these attributes tend to be enhanced. Experienced and knowledgeable people – who arguably make the best employees – are those most likely to be independent in thought and to find the internal climate of organizations restrictive. In other words, it is those employees with the greatest potential who are being most heavily restricted by the organization and its structure.
The result, in terms of organization culture, tends to be disbelief, distrust and inhibition. The trust and loyalty which the organization should be encouraging are not there. Instead, in frustration, people may even take negative steps towards the organization, reducing their work output, ‘gold-bricking’ or even leaving the organization altogether; high staff turnover can be read as one sign of an organization in which individual needs are being suppressed by organizational constraints.
One of the consequences of this incongruence can be the erection of barriers to organizational change. Argyris (1970, 1971, 1980, 1982) describes how what he calls ‘defensive routines’ inhibit change, retarding its progress or even blocking it altogether. These defensive routines are actions taken by employees or groups of employees for the specific purpose of warding off changes which they perceive to be dangerous or threatening. While some of these routines are obvious, others are much less so. The most problematic routines to deal with, says Argyris, are those undertaken for what are perceived as positive reasons. Defensive routines are not necessarily undertaken for selfish reasons: people may use them to ‘support’ the organization against changes which they believe would damage it, or to ‘protect’ other employees whom they believe to be vulnerable to unwanted change, or because they wish to be ‘realistic’ about the prospects for change. Many of these routines become so embedded in the organization that even the change strategies designed to overcome them begin to take on a defensive nature themselves.
How, then, to overcome these barriers and manage change? In the next phase of his work, Argyris turned to both his own background in psychology and to colleagues in sociology, most notably Donald SCHÖN, with whom Argyris had an extremely fruitful collaboration in the 1970s. Together, they developed the theory of ‘action science’. Argyris and Schön decided to switch their attention from observed behaviour to the actual processes of reasoning, to get at the causes and sources of behaviour, specifically the knowledge and routines employed when planning and undertaking actions. The term ‘action science’ was developed as an alternative to the concept of ‘normal science’ as elaborated by T.S. KUHN. For Argyris, the static studies employed by conventional rigorous research are divorced from reality; science needs to be part of action, and vice versa. As he says: ‘Action is how we give meaning to life… Actionable knowledge is not only relevant to the world of practice, it is the knowledge that people use to create that world’ (Argyris 1993: 1).
At the core of action science is learning and the circulation of knowledge. It is the circulation and use of knowledge, says Argyris, that leads to effective change; therefore the emphasis should be on the knowledge, not the change processes. He distinguishes between single-loop learning, in which feedback is used to alter actions, and double-loop learning, in which feedback is used to question the underlying assumptions on which action is based. Truly effective change, he says, must be based on double-loop learning. Here the problem of defensive routines becomes important once more: many people within the organization, through motives as various as pride in the status quo and fear of uncertainty, may resist double-loop learning as it threatens their own previously held beliefs and convictions. Time and effort must be spent breaking down these routines before double-loop learning can be put into place.
Action science is not a panacea, and the pitfalls for the researcher and change agent are many. But the concept has two important advantages: first, the research aspect of action science generates considerable quantities of new knowledge about the organization; second, the action this research calls for ensures a wide circulation of this knowledge. The breaking down of defensive routines calls for patience and persistence, but the new knowledge generated can itself be a powerful resource for doing this. Above all, perhaps, action science provides a clarity which other forms of research do not; as Lundberg comments, ‘perhaps the most distinctive feature of action science is that is has the potential to uncover its own contradictions and alter its own learning processes accordingly’ (Lundberg 1998: 22). This dynamism and flexibility means action science can be configured to a variety of organizational needs.
In the 1970s and again in more recent work (for example, Argyris 2000), Argyris has also paid attention to the role of the change agent. He is particularly critical of the forms of research which change agents use to gather information. He notes that research into the functioning of organizations tends to have some of the faults found in those organizations themselves. That is, it tends to be top-down and directive, with the researchers controlling the research programme and defining the tasks that subjects would undertake. This tends to distort the results. Argyris discovered that the results of studies of organizational behaviour seldom matched actual behaviour; there was a further incongruence, between what people actually did and what they said they did. This problem further fuels his belief in the need for action science, which, by shifting away from behaviour to knowledge and beliefs, has a capacity to get closer to the heart of the problem. The importance of a neutral, value-free change agent cannot be overstressed, but even these agents – such as consultants – can unwittingly reinforce or even create defensive routines. Ultimately, it is not so much the character of the agent that matters as the quality of the change created.
Argyris’s work has not been universally accepted. His writing can be dense and impenetrable to the layman, and he has been accused of focusing too much on conceptual thinking and not enough on the practicalities of actually implementing action science. However, his conceptualizations of learning flows, knowledge in action and defensive routines are becoming increasingly relevant in modern knowledge-based businesses as we learn more about how knowledge functions and is controlled and distributed within organizations.