Kurt Lewin's Change Theory
7. Personal and Relational Refreezing
The main point about refreezing is that new behavior must be to some degree congruent with the rest of the behavior and personality of the learner or it will simply set off new rounds of disconfirmation that often lead to unlearning the very thing one has learned. The classic case is the supervisory program that teaches individual supervisors how to empower employees and then sends them back into an organization where the culture supports only autocratic supervisory behavior. Or, in Lewin's classic studies, the attempt to change eating habits by using an educational program that teaches housewives how to use meats such as liver and kidneys and then sends them back into a community in which the norms are that only poor folks who can't afford good meat would use such poor meat.
The implication for change programs are clear. For personal refreezing to occur, it is best to avoid identification and encourage scanning so that the learner will pick solutions that fit him or her. For relational refreezing to occur, it is best to train the entire group that holds the norms that support the old behavior. It is only when housewives groups met and were encouraged to reveal their implicit norms that change was possible by changing the norms themselves, i.e. introducing collectively a new set of standards for judging what was"ok" meat.
In summary, what I have tried to show above is that Lewin's basic model of change leads to a whole range of insights and new concepts that enrich change theory and make change dynamics more understandable and manageable. It is a model upon which I have been able to build further because its fundamental concepts were anchored in empirical reality. Intellectual knowledge of the change process is not the same as the know-how or skills that are learned in actually producing change. In the next section I examine the implication of Lewin's thinking for the practice of change management.
II. "You Cannot Understand a System Until You Try to Change It:" Process Consultation and Clinical Research
The change and consulting literature is filled with the notion that one first diagnoses a system and then intervenes to change it. I learned early in my own consulting career that this basic model perpetuates a fundamental error in thinking, an error that Lewin learned to avoid in his own change projects and that led him to the seminal concept of "action research." The conceptual error is to separate the notion of diagnosis from the notion of intervention. That distinction comes to us from scientific endeavors where a greater separation exists between the researcher and the researched, particularly from medicine where the physical processes are assumed to be somewhat independent of the psychological processes (an assumption that is not even holding up in many parts of medicine).
The classical model is that the doctor makes an examination, runs certain tests, decides what is wrong, and writes a prescription which includes recommendations for therapy or, if necessary, for other interventions such as surgery. The consulting industry has perpetuated this model by proposing as a major part of most projects a diagnostic phase in which large numbers of interviews, questionnaires, and observations are made the basis of a set of recommendations given to the client. Consultants differ on whether they feel they should also be accountable for the implementation of the recommendations, but they tend to agree that there is a discrete billable period in any project that is basically considered necessary--namely a diagnosis of the problem--and that the consultant's basic job is done with a set of recommendations "for future intervention." If interviews or surveys are done, the attempt is made to be as scientifically objective as possible in gathering the data and to interfere minimally during this phase with the operation of the organization. What is wrong with this picture?
If Lewin was correct that one cannot understand an organization without trying to change it, how is it possible to make an adequate diagnosis without intervening? So either consultants using the classical model are getting an incorrect picture of the organization, or they are intervening but are denying it by labeling it "just diagnosis." Isn't a better initial model of work with organizations something like the stress test that the cardiologist performs by putting the heart under pressure to see how it will perform, even knowing that there are some risks and that some people have been hurt during the test itself? This risk forces the diagnostician to think about the nature of the "diagnostic intervention" and to apply clinical criteria for what is safe, rather than purely scientific criteria of what would seemingly give the most definitive answer.
It is my contention that Lewin was correct and that we must all approach our consulting work from a clinical perspective that starts with the assumption that everything we do with a client system is an intervention, and that, unless we intervene, we will not learn what some of the essential dynamics of the system really are. If we start from that assumption, we need to develop criteria that balance the amount of information gained from an intervention with the amount of risk to the client from making that intervention. In other words, if the consultant is going to interview all the members of top management, he or she must ask whether the amount of information gained will be worth the risk of perturbing the system by interviewing everybody, and, if the answer is "yes," must make a further determination of what is to be learned from the reactions of the management to being interviewed. That is, the interview process itself will change the system and the nature of that change will provide some of the most important data about how the system works, i.e. will respondents be paranoid and mistrusting, open and helpful, supportive of each other or hostile in their comments about each other, cooperative or aloof, and so on. The best information about the dynamics of the organization will be how the organization deals with the consultant, because his or her very presence is de facto an intervention.
In actual practice what most change agents have learned from their own experience is that "diagnostic" activities such as observations, interviews, and questionnaires are already powerful interventions and that the process of learning about a system and changing that system are, in fact, one and the same. This insight has many ramifications, particularly for the ethics of research and consulting. Too many researchers and consultants assume that they can "objectively" gather data and arrive at a diagnosis without having already changed the system. In fact, the very method of gathering data influences the system and, therefore, must be considered carefully. For example, asking someone in a questionnaire how they feel about their boss gets the respondent thinking about an issue that he or she might not have focused on previously and it might get them talking to others about the question in a way that would create a common attitude that was not there before.
The concept of process consultation as a mode of inquiry grew out of my insight that to be helpful one had to learn enough about the system to understand where it needed help and that this required a period of very low key inquiry oriented diagnostic interventions designed to have a minimal impact on the processes being inquired about (Schein, 1969,1987,1988). Process consultation as a philosophy acknowledges that the consultant is not an expert on anything but how to be helpful, and starts with total ignorance of what is actually going on in the client system. One of the skills, then, of process consulting is to "access one's ignorance," to let go of the expert or doctor role, and get attuned to the client system as much as possible. Only when one has genuinely understood the problem and what kind of help is needed, can one even begin to think about recommendations and prescriptions, and even then it is likely that they will not fit the client system's culture and will, therefore, not be refrozen even if initially adopted. Instead, a better model of help is to start out with the intention of creating in insider/outsider team that is responsible for diagnostic interventions and all subsequent interventions. When the consultant and the client have joint ownership of the change process, both the validity of the diagnostic interventions and the subsequent change interventions will be greatly enhanced.
The flow of a change or managed learning process then is one of continuous diagnosis as one is continuously intervening. The consultant must become highly attuned to his or her own insights into what is going on and his or her own impact on the client system. Stage models which emphasize up front contracting do not deal adequately with the reality that the psychological contract is a constantly evolving one and that the degree to which it needs to be formalized depends very much on the culture of the organization.
In summary, Lewin's concept of action research is absolutely fundamental to any model of working with human systems, and such action research must be viewed from a clinical perspective as a set of interventions that must be guided primarily by their presumed impact on the client system. The immediate implication of this is that in training consultants and change agents one should put much more emphasis on the clinical criteria of how different interventions will affect client systems than on the canons of how to gather scientifically valid information. Graduate students should be sent into field internships as participant observers and helpers before they are taught all the canons of how to gather and analyze data. Both are necessary, but the order of priority is backward in most training programs.
What can be done to enhance an understanding of these models and to begin to build the necessary skills to implement them? We turn next to an experimental course that attempts to teach "the management of planned change."