A General Philosophy of Helping: Process Consultation

I WOULD LIKE to review some observations I have made over the last thirty years about the process of helping human systems. I say human systems rather than individuals or small groups because much of my work as a consultant has been with intergroup and organizational-level problems. Individuals are always centrally involved, but the definition of the client can get very complicated.

In fact, the systemic approach requires one to think simultaneously in terms of three clients: immediate or contact clients with whom one is interacting in the here and now; primary clients, who are the real targets of change, and who pay for the change efforts; and ultimate clients, who are the stakeholders that must be considered even though one might not ever interact with them directly.

I make this point at the outset because process consultation has been stereotyped as something one does primarily with small groups. My own experience is that one works on a daily basis with individuals, small groups, or large groups, but that one’s concerns are always systemic in the sense that one considers immediate interventions in terms of their consequences for other parts of the system. For example, one might choose not to help a manager to become more autocratic even if that was the manager’s wish, if such behavior would be dysfunctional for the department or harmful to subordinates.

I have three points that I wish to develop.

• Helping is a general human process that applies to parents, friends, teachers, and managers, not just to consultants or therapists whose central role is to help.
• Helpers make choices based on key assumptions that have to be examined continuously during the helping process. These choices are primarily on-line, real-time decisions about when to be in the role of expert, doctor, or process consultant. I will explore the contrast among these roles in some detail below.
• A central concern of consultants should be to improve the ability of clients themselves, especially managers, to become more helpful to superiors, subordinates, peers, customers, suppliers, and other stakeholders. In other words, the helping role is critical in all human affairs; more people should be taught to be effective helpers. It is an especially important role in hierarchical organizations and, therefore, needs to be taught especially to managers and leaders.

Historical Footnotes

I have come to these conclusions over a period of time and base them on a variety of experiences as a teacher, researcher, and consultant. My back-ground was in social and clinical psychology, and I had a chance to see these roles in operation while working from 1952 to 1956 in the Neuropsychiatry Division of the Walter Reed Army Institute of Research under David Rioch. I came to MIT in 1956 and was encouraged by Douglas McGregor to get involved with the activities of the National Training Laboratories for Group Development, where for fifteen years I served as a trainer in various kinds of human relations workshops, especially ones focused on managers.1

I learned a lot about how an effective group should operate, and I also learned that to make a group more effective I could not simply advise it. I had to find a way to make the data visible, so that the group could learn from its own experience. The essential skill of the trainer was to make observations in such a way that the group could learn from them, a model not unlike the psychoanalytic group model being proposed by Bion at the time and used by the A. K. Rice Institute in their workshops today.2

Given this background, I approached my first organizational consulting with models of effective interpersonal relations and group behavior in mind, and with an armamentarium of observational and intervention skills in my tool bag. I was fortunate in 1965 to have the opportunity to work with the top management team of a young high-tech company. My explicit mandate was “to help the group with communication and to make them more effective as an executive team.” I was to join the group at the weekly staff meeting, observe them at work, and intervene as appropriate.

There was more than enough to observe. The managers were very confrontational, constantly interrupted each other, often shouted at each other, revealed information of a negative sort about each other at the meetings, blamed each other, and in other ways behaved ineffectively. I shared my observations when I felt it appropriate and when I could get a word in edgewise, and I suggested that the group examine the consequences of their behavior. Their response was always one of interest. Members were grateful to have their behavior pointed out, and they expressed regret and some shame at what they themselves could easily see was “bad.” They complimented me on my perceptiveness, and then continued to do exactly what they had been doing. In other words, nothing changed.

At first I attributed my lack of influence to my lack of skill in making the consequences of the group’s behavior sufficiently visible. But as this scenario repeated itself over many months, and as my own frustration grew, I began to realize that I was making some inappropriate assumptions about helping. I realized that the helper has some real choices about how to help, and that I was making poor choices.

Specifically, I was assuming that I knew how the group should operate better than the group did itself. I was importing a model of effective group action from my training experience into a work setting. I was also imposing a set of humanistic values pertaining to how people should communicate, how they should not publicly embarrass each other, and how they should reach consensus on decisions.

In letting those assumptions guide me, I was missing a crucial point—the group had an agenda more important than all of the above considerations. That agenda was driving and stabilizing their group process. Specifically, the agenda was to resolve critical strategic issues around choices of technology and products in an industry where no one really knew what would and would not work, and where the academic tradition—that ideas had to be fought out in order to be tested and validated —prevailed. I was busy trying to civilize the group, while the group was searching for truth in a life and death struggle against its competitors. I was imposing my expertise about groups on a group trying to solve a problem far more important than how to be an effective group.

Of course, I could have pointed out to myself that I was responding to the very request the group had made. They had asked me to help them be more effective as a group. But I eventually realized that they themselves did not know exactly what they had in mind. They only knew that “some-thing” was wrong, and they were counting on me to help fix it. They sincerely tried to help me by paying attention to what I said, but they, like I, found that some other agenda was driving their behavior.

In retrospect, the essence of what I came to think of later as “process consultation” was derived from the insight that I could not be helpful until I gave up my own notion of what the group should be and began to pay attention to what the group was actually trying to do. I had to abandon theory about how individuals, groups, and organizations should function and learn to be a better observer of what was actually going on.

My learning process was aided by one fact: This executive group violated most of my preconceptions, yet was enormously successful in running their business. Their track record in decision making was extremely good, and the company was growing rapidly. Obviously, something was lacking in my own theories.

As I began to see what the group was trying to do, and as my curiosity overrode my need to be helpful, I actually became more helpful. Instead of focusing on the dysfunctional aspects of interrupting, I began to focus on the idea that was being cut off. Occasionally I restated the idea. As the group suffered information overload, I went to the flipchart and wrote down some of the ideas that might be getting lost. As I observed destructive conflict between two members, I asked them to elaborate on what they were saying instead of pointing out to them that they were in conflict. I began to intervene in the “real” process of the group, its task process, and allowed myself to get preoccupied with interpersonal issues only when there was time, a clear need to deal with them, and a readiness on the part of the group to do so. Needless to say, the feedback from the group was that “now I was really helping.”

I knew I was on to something, but the clearer articulation of what lay behind these insights was only later forced upon me by a colleague who perceived management consulting to be a waste of time. Why teach elementary psychology to a bunch of managers, or counsel them on their hang-ups, when one could be doing important research? This challenge angered me because I believed I was learning more from my training and consulting than I was in the research laboratory. But I also realized that what was going on with my clients was invisible to and misunderstood by my colleagues. This led me to write about the three models of consultation that I now understood more clearly and, more recently, to elaborate on the process consultation philosophy.3

Three Models of Helping

The essence of process consultation as a helping philosophy can best be articulated by contrasting it to two other helping models that seem to me substantively quite different. The helper has to make on-line choices about which model to use from moment to moment.

Model One: Providing Expert Information

There are times when it is most helpful to give information relevant to a client’s problem. The client wants to know how the workers in a given plant feel about something and asks the consultant to do a survey to find out. A subordinate asks the boss: “How do I deal with this problem employee in my group?” and the boss tells her how. A child asks a parent: “How do I do this math problem on my homework?” and the parent shows him how to do it.

This seems straightforward enough, but notice that the model makes several assumptions that often cannot be met. It assumes that the client knows what the problem is, that the client has communicated the real problem, that the helper has the needed information, and that the client has thought through the consequences of asking the question and receiving the answer.

It may be that doing the survey will raise expectations that the manager is not prepared to deal with. It may be that the subordinate or child is learning how to be dependent on the boss or parent at a time when it is more important that they learn to dig out the information themselves. It is also possible that the boss or parent is wrong and that the information will not be helpful.

Sometimes the helper—the consultant, the manager, or the parent—must think about these assumptions and assess the consequences of providing expert information. And sometimes, based on this assessment, the helper must choose not to operate in that model even if requested to do so. Yet to recognize that one may not be as expert as one assumed, or that the client may not really benefit from one’s knowledge, is extremely difficult.

Model Two: Playing Doctor

Clients often invite helpers to be, in effect, a doctor: to investigate, interview, psychologically assess, run tests, make a diagnosis, and suggest a cure. If consultants find that model one is an ego trip, think how we respond to model two. For example, the organizational client wants us to investigate what’s wrong in a department and suggest a cure; the subordinate goes to the boss with a broad request for diagnostic help in dealing with problem people; or the child comes to the parent with the lament that he can’t ever do the math and doesn’t know what’s wrong. The temptation to put on our stethoscopes and to launch into a diagnostic investigation is overwhelming. The popular notion of the consultant fits this model, and, similarly, some models of psychological management consulting start with individual assessments of the key actors as a basis for diagnosing the systemic problems.

Given our training as outside “experts,” this all sounds eminently logical and appropriate, but what assumptions does it imply? First, the doctor model assumes that the client has correctly identified the sick area. Second, it assumes that the “patient” will reveal the information necessary for a good diagnosis. (I have often found that my data was invalid because a grateful client exaggerated the problems, or a resentful client denied them completely.) Third, a correlated assumption that applies especially to consulting clinical psychologists is that they have the expertise necessary to arrive at a correct diagnosis. Fourth, this model assumes that the client will accept the diagnosis arrived at. Fifth, it assumes that the client will accept the prescription and do what the “doctor” recommends. And finally, it assumes that the client will be able to remain healthy after the doctor leaves.

We are all frequently frustrated in our helping efforts by clients who do not accept our expertise, who file our reports instead of acting on them, who misunderstand and subvert our recommendations, or who revert to their disease the minute we leave. As an aside, I might comment that this last condition may not bother us as consultants inasmuch as it keeps us employed, but in fact if the patient becomes too dependent, we are no longer consulting or helping. We become de facto managers wearing consultant hats.

The model often goes awry because one or more of these assumptions cannot be met. It is a perfectly good model when it applies, but only then. Incidentally, physicians are themselves questioning this model as they observe their own patients resisting diagnosis or prescriptions and as those prescriptions themselves become more complex.

Model Three: Process Consultation

That brings us to the philosophy I have labeled process consultation and to a set of assumptions that seem to better fit human systems with which we typically deal.

First, I assume that clients, whether managers, subordinates, children, or friends, often seek help when they do not know exactly what their problems are. They know something is wrong but the help they really need is in figuring out exactly what that is. Once that question has been answered, they can often figure out their own solution.

Second, I assume that most clients do not know what kinds of help are available and what kinds of help are relevant.

Third, I assume that many problems in human systems are such that clients would benefit from participation in the diagnosing process—particularly since they are so often part of the problem and need to be led to this insight.

Fourth, I assume that only clients know what form of remedial intervention will really work, because only they know what will fit their personalities and their group or organizational culture.4

Fifth, I assume that clients have “constructive intent” and will benefit from the process of learning how to solve problems, so that future problems can be dealt with more effectively. The implication is that, if the client’s goals are not acceptable to the consultant helper, he or she should not enter into a helping relationship in the first place.

What these assumptions mean is that helpers must suspend most of their own biases initially; they must develop a mutual inquiry process that not only creates a shared sense of responsibility for figuring out what is wrong and how to fix it, but also enables helpers to pass on some of their own diagnostic and intervention skills. Helpers must help their clients to learn how to learn.

Another way of making this point is to note that, in the expert or doctor model, the consultant allows and even encourages the client to pass his or her presumed problem on to the consultant. Once the helper has accepted the problem and the responsibility to do something, the client can relax and wait for answers or recommendations. The client is then in an ideal position to distance him or herself from whatever the consultant proposes.

If we allow clients to distance themselves, we have already lost the war, because it is, after all, the client’s problem we are dealing with. In process consultation, it is essential to create a situation in which clients continue to own their own problems; the consultant becomes a partner or a helper in diagnosing and dealing with those problems. But they will never be the consultant’s problems, and we should not allow clients to feel that we can take their problems off their shoulders onto our own.

As the relationship between the consultant and the organization evolves, the identity of the client is gradually broadened; the consultant may work with individuals, groups, and organizational units at different times. But the basic assumptions about how to work with these various client systems remain the same.

The Three Models in Practice

Having articulated the three models of helping and some of the assumptions that underlie them, let me be practical and discuss how they really work in my own experience. I am continually moving from one model to another as the situation dictates: my greatest problem is to know when to operate from which model. Two guidelines reflect the process consultation philosophy.

Always Start in the Process Consultation Mode.

When a client, manager, friend, child, subordinate, or boss comes to you with a request for help or a question that invites you to give advice, assume that you may not know exactly what you are being asked for and should, therefore, adopt a spirit of inquiry.

I assume that whatever I say will be an intervention; the process of helping, then, begins with my first response to the inquiry. I therefore need a category of interventions, which I label “exploratory,” whose strategic goal is simultaneously to provide help; provide some diagnostic insight; and ensure that the client will continue to own the problem and begin to feel that we are a team working on it together.

The initial interventions are the most important ones: they communicate my strategic intent and create the right kind of psychological contract between me and the client. The client is typically overtrained to expect me to take an expert or doctor role, especially if I am being paid for the help, so I have to simultaneously be helpful and also correct the stereotype. I may later realize that I should be the expert or the doctor, but I have no way of knowing that until I have thoroughly explored the situation.

So I say things like, “go on,” or “tell me a bit more,” or “can you describe the situation?” or “what do you have in mind?” before I leap in with answers and advice. The key is to be genuinely curious and to communicate that I feel no obligation to take the problem onto my own shoulders. But I do want the client to feel that I am being helpful.

I also want to communicate my genuine ignorance of the deeper psychological and cultural issues that may lie behind what I am told. Even with my diagnostic tools and experience, I know very little about what is going on in any new situation. This ignorance is one of my most important assets, because it permits me to ask all sorts of dumb questions that might offend if I really understood my contact client’s situation better.

As the conversation develops, my focus gradually shifts to what I call “diagnostic interventions,” but I am still operating in the process consultation mode because these interventions invite diagnostic thinking from the client, not from me. Examples might be, “why do you feel this is an issue or a problem?” or “why do you think this is happening?” or “why did you come to me with this question?” Where the focus of exploratory questions is on the “what,” the focus of diagnostic questions is on the “why.”

These questions may, of course, be perceived by the client as stalling. If I irritate the client, I am not being helpful. The choice of intervention, there-fore, has to be guided all along by the strategic intent to be helpful; that may require moving more rapidly to what I call “action alternative interventions.” Here the focus shifts from why something might be happening to what the client has done or is intending to do about it. I might ask, “what have you tried to do?” or “what do you plan to do?” or “what alternatives have you considered?”

The characteristic common to these categories of intervention is that they all keep clients actively solving their own problems without having to deal with advice or new information that comes from the helper. The helper is steering the process, but not adding new content.

If new content is clearly called for—if the client signals that he or she really wants new information or ideas or advice—the helper can, of course, provide them. But we are then dealing with what I call “confrontive interventions”; these force the client to think about new facts, ideas, or alternatives that might not have been considered before.

If I want to be confrontive and yet be consistent with the assumptions of process consultation, these interventions must be couched in a way that does not make me into an expert or a doctor, yet that gets across my hypotheses about what may be going on. The easiest method is to provide the new information or ideas in the form of alternatives, hypotheses, or possibilities. “Have you considered the following items of information?” “Have you thought about options A or B?” “Maybe you are having one of the following feelings—you are anxious or maybe angry?”

If I state the alternatives and state them in question form, the client is forced to stay in the active problem-solving mode. As the helper, I am maintaining the realistic posture that I do not really know what is going on, but have begun to consider some hypothetical alternatives.

The difference between offering alternatives in question form and giving advice may seem stylistically trivial, but it is philosophically crucial. Clients operating from their stereotype of the consultant as doctor sometimes try to make consultants feel like a coward or a cop-out if they don’t offer a single recommendation. Therefore, consultants must be able to argue for their style on the logical grounds that they cannot possibly get inside the client’s system and culture to a sufficient degree to recommend a single course of action.

I have been surprised by how rarely one needs to be confrontive, and by how much can be accomplished early in the relationship with an inquiry, diagnostic, and action alternative mode. Let us now turn to how things evolve, which brings us to the second guideline.

Do Not Withhold Your Expertise if the Client Really Needs It.

Just as it is not helpful to leap in with premature advice, it is also not helpful to withhold advice if the helper realizes that the client is about to make an error. If the initial inquiry process reveals enough to enable the helper to be an expert or a doctor, and if this seems necessary and appropriate, then of course the helper should shift to either of those roles.

For example, in the high-tech company meetings previously referred to, once I realized what the group was trying to do, I found myself in an expert role with respect to two crucial issues. First, I was more expert at listening than other group members and so was able to restate or write down what members were saying, thus making this information available to the whole group. Second, as the group began to redesign their meeting format, I realized I was more expert at meeting design; therefore, I was able to make recommendations on how meetings should be run.

The key to moving into and out of these roles appropriately is to know enough about what is happening and to know what one’s own areas of expertise really are. When I have not spent enough time in the process consultation mode to figure out what the problem really is, or when I develop the illusion that I know what the client should do, I get into trouble. My recent inquiries into organizational culture have consistently shown how idiosyncratic organizations really are and how difficult it is to prescribe what managers should do.

Individual Assessment and Employee Surveys

To further illustrate the contrast between the consultation models, I would like to comment on two kinds of interventions typical of psychologically oriented management or organizational consultants: individual diagnostic profiles based on testing, interviewing, or assessment centers; and opinion or morale surveys of employees.

Neither of these interventions is an appropriate way to identify system problems initially because they both cast the consultant in the expert or doctor role and stimulate client dependency that will later undermine joint problem solving. They assume that the consultant, by virtue of special secret knowledge, must shoulder responsibility for diagnosing the situation and making prescriptive recommendations.

Does this mean we should never use individual assessments or surveys? Not at all. It means we should use these techniques only when the primary client has decided with the consultant that such an intervention would be helpful and when the client accepts the responsibility for the consequences of the intervention. A good test is whether the client is willing to explain to others what the intervention will be and why it is being used. This usually means that a good deal of diagnostic problem solving has gone on before a decision is made to use such a major intervention.

The three consulting models imply different ways of handling the intervention itself. In the expert or doctor model, the consultant uses proven tools that the client is not professionally trained to administer or interpret. The consultant has to interpret the results and make recommendations. The client is dependent on assessments of individuals based on validated tests and interviews, or, in the case of surveys, is given tables with statistical interpretations of their meaning and implications.

In the process consultation model, if individual assessment seems relevant, the primary client must help specify what areas need to be assessed. To avoid using invalid tests, the consultant would probably also move toward an assessment center concept. The main role of the consultant would be to teach members of the client system how to develop an effective individual assessment process.

If the client wanted outside professional assessment, the process consultant would probably recommend an appropriate resource and help the client to develop an internal process for feeding back and using the assessment information. The emphasis would be on providing such information only to the individual being assessed, and giving that individual the right to decide whether to pass the information upward in the organization.

In employee surveys, the contrast between the models is equally sharp. In the expert or doctor model, the consultant uses a proven, reliable, and valid questionnaire, tells the client how to administer it for maximum participation, collects the data, analyzes it, and then communicates the data to the top of the organization with advice and training about feeding it back to other levels. This is usually a “cascading down” process; each level is given its own data and told to work on it before it goes down to the next level. The consultant typically trains supervisors to handle the feedback sessions in ways that elicit appropriate responses from the employees.

In the process consultation model, one would proceed quite differently. Top management decides jointly with the consultant that a survey could identify problems in such a way that the organizational level “owning” those problems could get to work on them. The goal is not to gather data but to solve problems.

The survey questions are based on individual and group interviews with diagonal slices of the organization; all employees are consulted on what should go into the questionnaire. The intent is to get the whole organization to think diagnostically from the outset. Such involvement typically results in a higher response rate and a feeling of data ownership.

Once the data is gathered, it is aggregated by group from the bottom up. Each lower-level group is given its own results (without the presence of the supervisor) in order to do two things: correct the data or enhance it; and sort the results into those problems that the group can address and those that need to be passed upward for higher-level attention. All this happens before anyone higher up has seen aggregated data.

This bottom-up process is dramatic in that it clearly establishes in the employees’ minds their ownership of the data and of some of the problems. The group meeting itself is a clear signal that management expects employees to diagnose and fix their own problems. They cannot assume that, having told management about the problems, they are off the hook—in other words, they cannot become dependent.

As this process works its way up the organization, problems get identified, sorted out, and worked on by those who have the appropriate resources and responsibility. There may, in fact, never be a summary aggregate report. Top management may never see any statistics on different departments; instead, they will see a highly motivated organization working out solutions.

One may well ask why top management would pay for a survey if they never saw the results. Paradoxically, once top managers become convinced that the bottom-up method starts to solve problems, they realize that they’d rather get solutions than long lists of problems. The process consultant must spell the issue out for company presidents: would they rather see fancy tables and statistics that leave them having to motivate the “problem departments,” or, on the other hand, would they rather initiate a process that identifies problems in such a way that they get worked on immediately? Most managers I have worked with prefer the second alternative once they understand its potential.

Conclusions

Periodically we all find ourselves in the role of a helper. If we are to play that role effectively, we must be conscious of the choices we make about being a process consultant, an information expert, or a doctor. Each of these major models rests on assumptions that have to be examined; with human systems, the assumptions that underlie the process consultation model are most likely to be the correct ones.

In almost all helping situations, the initial interventions must be guided by an inquiry mode that establishes an appropriate helping relationship; I believe that the process consultation model is the most appropriate way to do that. Finally, as we glean some insight into what is going on, we must shift into and out of the expert and doctor roles according to the client’s needs and a realistic assessment of our own expertise.

I sincerely believe that helping relationships are a basic category of all human relationships. We must not only be better at managing such relationships when we are in the formal role of helper and consultant; we must also teach effective helping to parents, managers, and all others who are involved with people.