Barry Grant

Chicago, Illinois


Nondirectiveness is a focal point in the debate about the nature of person-centered and client-centered therapy. In my view, the debate is essentially about the morally best way of doing therapy. Conceptions of nondirectiveness differ primarily in whether they emphasize pragmatic concerns for promoting growth and "meeting needs" or respect for persons. Two conceptions of nondirectiveness — instrumental and principled — are described and compared. Principled nondirectiveness is elaborated, and a justification for it is sketched.

Barry Grant is a part-time psychotherapist and college instructor. His first psychotherapy class was in client-centered therapy, which at the time seemed to him to be on to something important, but simplistic and unexciting. Later, upon reflection, and the (nondirective) influence of Barbara Brodley and Marjorie Witty, he came to see some of the depth of its simplicity.

The defining characteristics of person-centered therapy and client-centered therapy have been debated in Renaissance, the Person-Centered Review, and at ADPCA meetings for at least the past three years (e.g., Bozarth & Brodley, 1986; Cain, 1986; Rogers, 1987). One of the focal points of the debate, perhaps the focal point, is nondirectiveness: what nondirectiveness is, and the extent to which consistent adherence to it is a feature of person-centered therapy and client-centered therapy. Conceptions of nondirectiveness seem to differ primarily in the extent to which they emphasize either pragmatic concerns for promoting growth and "meeting needs," or respect for persons. What I will call "instrumental nondirectiveness" and "principled nondirectiveness" embody the different emphases. Instrumental nondirectiveness is seen as essentially a means of facilitating growth; principled nondirectiveness is essentially an expression of respect. In this paper, I describe and compare the two notions, and elaborate on and sketch a justification for principled nondirectiveness.1

In the instrumental version of nondirectiveness, which I take to be exemplified in a recent essay by Cain (1989), the purpose of the therapist’s actions is to bring about growth or empower clients. If being nondirective facilitates growth for a client in a particular instance, then it is valuable; if it does not, then the therapist decides whether continuing to be nondirective or adopting a different approach would be more effective. Respect for the client is not absent, but is allied to or tempered with a pragmatic concern with promoting growth. In the principled version, the therapist’s actions are derived from the fundamental idea of respect for persons (Brodley, 1986; Grant, 1986). The therapist does not attempt or intend to make anything happen — growth, insight, self-acceptance — in the client, but rather provides the therapeutic conditions in the belief that they are expressions of respect and with the hope that the client will make use of them. Following Brodley’s (1988) distinction between the person-centered therapies and client-centered therapy, principled nondirectiveness is essential to client-centered therapy, while instrumental nondirectiveness is part of the other person-centered therapies.


PERSON-CENTERED REVIEW, Volume 5 Number 1, February 1990, pp. 77-88 © Sage Publications

The two conceptions of nondirectiveness are part of different moral visions. To describe them in this way may seem odd and out of place, but I believe it precisely describes the most important difference between the two views. The debate about the nature of person-centered therapy, and about nondirectiveness, is essentially an argument about right and wrong in psychotherapy. Before discussing nondirectiveness, I want to give this claim some substance.



One way of looking at the question of the nature or definition of person-centered therapy is to see it as asking what is the correct interpretation of Rogers’s theory of psychotherapy. Rogers originated person-centered therapy. When we ask what it is, we ask about what he thought. This is important both to historians of psychology and to those of us who call ourselves person-centered or client-centered. We have chosen these labels because we believe our views on therapy are identical, or significantly similar, to Rogers’s views.

Any body of complex thought that has developed over a period of time will be open to more than one compelling reading. There is no one correct answer to the question of what Rogers’s therapy is, any more than there is one correct understanding of Freud’s, Jung’s, or any other person’s therapy. Debate about the correct interpretation of Rogers’s work can continue for as long as his readers can come up with new and persuasive ways of making sense of his writings. The debate will end when we all agree or lose interest. Neither is likely to happen soon.

The danger of debating the definition of person-centered therapy or client-centered therapy solely in terms of what Rogers wrote is that it can distract us from explicitly addressing a more important question. Certainly none of the disputants are just arguing about how to read Rogers. We are exercised about "misreadings" of Rogers because we have judgments about the practices they imply. We think it is wrong or right, better or worse, to help people to focus, to give advice, to be nondirective, and so on. We don’t just think Rogers did or did not advocate these things. The debate is not essentially (or should not be) about the correct understanding of Rogers. It is a way of addressing a basic question every form of psychotherapy must answer: How ought one to practice therapy?

The debate about what person-centered therapy or client-centered therapy is or is not is better understood as a dispute about what is the morally best way of doing therapy. Taking a position on the matter begins with Rogers’s work, but it should not end there. It begins there because the participants in the debate have been heavily influenced by Rogers. We take our vocabulary and guiding ideas from him. Because we are different and he isn’t unitary, we don’t take the same things in the same way, and we argue about what he said. But, defining terms, offering interpretations, and declaring ourselves are not adequate justifications for our practices. We must also give moral reasons for our views. We must be able to argue why it is good to engage in the practices we advocate. We should be able to do this because psychotherapy is a moral enterprise. It is an activity which can and does affect the well-being of others, and which is based on ideas about right and wrong in human relationships (see, of many authors who make this point, Grant, 1985; London, 1986; Szasz, 1978; Watson, 1958). Thus, principled and instrumental nondirectiveness are not only competing interpretations of Rogers, they are competing ideas about how one ought to conduct oneself as a therapist, what is most important in therapy, what it serves, what its goal is.


In instrumental nondirectiveness, the therapist’s primary commitment is to the client’s growth. Cain (1989) argued that strict adherence to nondirectiveness does clients a disservice when it does not serve the client’s growth. He sees nondirectiveness as a way of bringing about certain effects in clients; it is valuable insofar as it can do this, and an obstacle insofar as it cannot. He reads Rogers as advocating nondirectiveness because it promotes freedom and growth: "Rogers’s intent.., was to free the client to grow in his or her own way" (p. 124). The "paradox" of nondirectiveness in the title of his paper refers to the therapist, whose purpose in being nondirective is to empower, free, and bring about growth in clients, and who may, at times, actually disempower clients, and hamper and frustrate their growth. Cain (1989) wrote:

To the degree to which person-centered practitioners feel compelled to be nondirective, the more likely they are to impose limits on their clients because they constrain their own behavior... . In order for our clients to feel genuine freedom it seems necessary that we too must feel free to interact in a variety of ways that may or may not include nondirectiveness. (pp. 129-130)

The major point of his essay is that this is especially true for clients whose "learning style" is not compatible with nondirectiveness. He argued, in effect, that it is wrong to be nondirective when being so is counterproductive. Cain did not explicitly describe what actions are consistent with his conception of nondirectiveness, but he seems to have a common view of it which holds that it is usually expressed through empathic responses. Indications from the client for a different response are likely to be met with empathic responses, by simply listening or by encouraging the client to experiment or figure things out alone.

In this view, efficacy is the most important criterion that nondirectiveness or any attitude or technique has to meet. "Does it facilitate growth?" is the first question a therapist of this ilk asks. Growth is necessarily defined to some extent independently of the client’s frame of reference. The therapist must have an idea of what constitutes growth in order to decide when to be nondirective.



"Does it respect the client?" is the first question a therapist who has a principled conception of nondirectiveness asks about an intervention. For this sort of therapist, the question about efficacy is absent. Because therapists claim to offer service, they may ask clients if therapy is helpful. But the client-centered therapist’s rationale for being nondirective is not that nondirectiveness works. Being nondirective in a principled manner is not a way of making something happen, not a way of causing growth or freedom or empowerment or self-acceptance. Client-centered therapists are not nondirective because they believe that by declining to exercise power they are more likely to "free" clients than if they did something else. They are nondirective because they respect their clients’ voices. Principled nondirectiveness is an attitude that provides a "space" for growth, not one that intends to cause it. Nondirectiveness, like love, is not acted upon for what it achieves, but for what it honors. Excepting instances in which other moral considerations prevail, it is, within this framework, always right to be nondirective and always wrong not to be.

Principled nondirectiveness is an attitude, not a set of behaviors. Having the attitude does not mean having a stock or a "hands off" approach to relating to clients, although the most common expression of it is empathic understanding responses (Brodley, 1988). Living the attitude means being open and responsive to clients’ requests and indications for other types of response. So, for example, a tennis instructor who holds to principled nondirectiveness would not respond to a request for instruction by encouraging the student to experiment, as Cain’s, or a directive instructor would, unless he or she thought that this was the best way to develop a stroke. If the instructor thought that there were better ways, those would be offered instead. A request for help would get the instructor’s best answer. If the instructor did not know, or was not sure, that would be indicated.

Principled nondirectiveness is an expression of an absence of the intention to make anything in particular happen, and of an openness to following the client’s direction.

How can the principled nondirective attitude be justified? I think there are a couple of reasons for it which I will sketch briefly. One derives from the "Attitudes and Orientation of the Counselor" chapter of Client-Centered Therapy (Rogers, 1951). There, Rogers wrote about the dignity and worth of the individual, the individual’s capacity for self-direction and right to self-direction, all values of liberalism. Client-centered therapy can be understood as liberal values enacted in a therapeutic setting (see Van Belle, 1980). "The liberal idea of freedom . . . claims for man, by reason of his humanity, the right, within limits . . ., to order his life as seems good to him" (Plemenatz, 1973). Client-centered therapy fully respects clients’ right to determine their path in life. It makes no assumptions about what people need or how they should be free. It respects clients as authors of their own lives and provides them with a space to rewrite their story, if they want to. Ibsen describes liberty as giving each individual the right to liberate himself, each according to his personal needs. Client-centered therapy offers this liberty. It does it in such a way that its ends are consistent with its means (Tomlinson & Whitney, 1970). This is very important. The liberation that can come from client-centered therapy is accomplished by respecting clients as autonomous beings, not by making them autonomous beings.

Principled nondirectiveness can also be seen as an expression of a nearly religious attitude toward the world. The attitude is religious in that it has to do with what Edwards (1982) calls "a fundamental and pervasive stance to all that is" (p. 236). Edwards very beautifully expresses an attitude that can serve as a basis for nondirectiveness in his description of an aspect of Wittgenstein's thought:

Wittgenstein’s [notion of sound human understanding] is the expression of a religious commitment; it is the expression that is, of a fundamental and pervasive stance to all that is, a stance which treats the world as a miracle, as an object of love, not of will. The sound human understanding is the mark of such love, for it is a feature of love that it never literalizes any perception; love is always ready to go deeper, to see through whatever has already been seen. From the perspective of loving attention, no story is ever over; no depths are ever fully plumbed. The world and its beings arc a miracle, never to be comprehended, with depths never to be exhausted. Thus the sound human understanding is essentially a religious response .... It is a response that makes sheer acknowledgement, not control, central. (p. 236)

This attitude is one of being humbled before the mystery of others and wishing only to acknowledge and respect them. Client-centered therapy is a way of making this acknowledgment — an almost aesthetic appreciation for the uniqueness and otherness of the client.

The paradox of principled nondirectiveness is that a therapist who wholeheartedly lives the attitude may at times appear to be extremely directive. A client may request direction, advice, interpretations, or instructions, and the therapist may offer these. The decision to do so does not depend on a determination of the client’s needs, best interests, diagnosis, or learning style. It rests, rather, on whether the therapist wants to honor the request, judges himself or herself competent to honor it, and believes it moral to do so. The therapist has no stake in having things come out one way rather than another.

Answering questions is not only not inconsistent with nondirectiveness, it may at times be required by it. The leading edge of growth is the client’s voice — the client’s requests, stated intentions, wishes, self-descriptions, and desires. Growth is what the client says and does. It is not a hypothetical process happening somewhere "inside" the client; nor can it be defined from outside of the client’s frame of reference. Answering questions meets clients at the edge of their growth — at the actions they initiate. Giving a personal response to a question is an encounter of one sovereign being with another. It is not a distortion or subversion of a process; it is a response to the only thing a person can know about the particular form of growth of a particular client2.

Principled nondirective therapists take their lead from the client and do not adapt to any determination of the client’s "learning style." They do not try to figure out clients’ learning styles any more than they try to figure out their diagnoses or anything else about them — unless they are asked and have an opinion. Principled nondirectiveness is incompatible with any programmatic use of the hundreds of notions such as "learning style," "diagnosis," "level of experiencing," which a therapist can use to conceptualize the client and link to an understanding of the nature of growth.

Client-centered therapists are, to use Cain’s word, "constrained" by the attitude of principled nondirectiveness. The freedom he advocates for the therapist as a means of facilitating freedom in clients is not consistent with principled nondirectiveness. Client-centered therapists do not intend to free or constrain their clients. They do limit themselves and the variety of ways in which they interact with clients. But this does not, as Cain argues, impose freedom or anything else on clients. To say so is to misuse language. It is not an imposition to not wish to direct or influence someone anymore than it is an imposition to speak English rather than Spanish, to love rather than hate.

The claim that client-centered therapists limit or constrain themselves needs to be clarified. Beginning and developing therapists should constrain themselves. It is wrong to act on every impulse or idea that occurs while doing therapy. Client-centered therapy is a discipline. Those who are gifted or who have done it for a long time may come to do it "naturally" and spontaneously. As one develops as a therapist, there is less need for self-restraint because there is less inclination to exercise power and control and more regard and love for others. The desire to change other people, to make things happen, dissipates.

Nondirectiveness, and the attitudes that express it — unconditional positive regard, empathy, and congruence — are moral virtues. They are praiseworthy characteristics of persons. Being able to act spontaneously and idiosyncratically on them does not happen by instructing oneself to loosen up and be free. This emerges out of discipline in the attitudes — e.g., by working on seeing in loving and accepting ways, by detecting and getting rid of the desire to exercise power, or by accepting oneself. No religious or moral tradition that I am aware of trusts that people left to their own devices will acquire the virtues valued by the tradition. They all recognize that character development takes work.

Because there is no telling exactly how a client-centered therapist will develop the attitudes, or what "new and more subtle ways of implementing [the] client-centered hypothesis" (Rogers, 1951p. 25a person will devise, empathic responses and personal responses to direct and indirect questions are not the only forms of response consistent with the nondirective attitude. Offering unsolicited opinions, suggestions, and the like can also be consistent with having the attitude. Nondirective client-centered therapy is a way of being, and not a method, because it allows the therapist to make novel, personal, unplanned responses. There is a place in client-centered therapy for

the therapist . . . in a spontaneous and non-systematic way [to] offer reactions, suggestions, ask questions, try to help the client experience feelings, share aspects of his or her own life, etc., while maintaining a basic and continuing respect for the client as the architect of the process (Raskin, 1988).

Cain sanctions these sorts of responses if they are well-intentioned attempts to empower the client, offered as options, and based on an understanding of the client’s needs and preferences. This rationale implies that the therapist is sizing up the client and judging what might be of most help. They are consistent with principled nondirectiveness if they are just things the therapist feels like saying. They are not justified by what they intend to accomplish, but by being personal expressions of the attitudes. These spontaneous and nonsystematic actions must be understood as coming from someone in whom the attitudes are deeply ingrained. Buddhists say that whatever an enlightened person does is faultless. Maybe so, but those of us who are less than perfect should usually stay to the well-worn path of common virtue. I am often suspicious of these actions in myself when I do them, and in others when I hear about them. I doubt the purity of the motivation. I doubt that they are gifts of love and regard, given with no desire to influence.

Perhaps those who are gifted with the attitudes or who have acquired them through discipline agree with the following lines from Rogers and Segel’s (1955) film and from Weil (1951), and have no desire to do anything but understand. According to Rogers:

The most that can be given to another person, not the least, but the most that can be given to another person, is to be willing to go with them in their own separate feelings as a separate person. (Rogers & Segel, 1955 [Film])

In a similar vein, Weil writes:

Those who are unhappy have no need for anything in this world but people capable of giving them their attention.. . . The love of our neighbor in its fullness simply means being able to say to him, "What are you going through?" (pp. 114-115) 


I have discussed two conceptions of nondirectiveness. Instrumental nondirectiveness is expressed by empathic responses and by encouraging clients to find solutions to problems within themselves. It is an often helpful and sometimes unhelpful way of promoting growth. On the principled conception, nondirectiveness is an ideally unwavering expression of respect that is manifested in many and unpredictable ways, but usually in the form of empathic responses. Many of the same behaviors are consistent with both versions of nondirectiveness, but they are done for different reasons, in different spirits.

Different conceptions of nondirectiveness, and of person-centered therapy and client-centered therapy, are not just private personal ideas about how to do therapy that are justified because Rogers supports them or because they are an expression of one’s process of growth. They are ideas about how one should conduct oneself as a therapist, and so they require moral justification. I have sketched two justifications for nondirectiveness in client-centered therapy. One appeals to a principle of respect for persons, the other to an attitude of "sheer acknowledgement" of others. The adequacy, implications, and appropriateness of these notions can, of course, be argued about. Indeed, we may disagree about what constitutes adequate justification. These are the sorts of discussions that come up when the moral dimension of psychotherapy is seen as central.


I. This essay is a revision of a paper presented at the Third Annual Meeting of the Association for the Development of the Person-Centered Approach in Atlanta, May 1989.

2. Rogers often defers answering questions. Sometimes, apparently, he does this because he believes the client ought to figure things out without his ideas and gives indirect instruction to do this. This is not principled nondirectiveness. Other times, apparently, he doesn’t answer questions because he doesn’t have an answer and doesn’t believe he could have one. This is consistent with principled nondirectiveness. Sometimes he does answer questions (Brodley, 1989).


Bozarth, J., & Brodley, B. (1986). Client-centered psychotherapy: A statement. Person-Centered Review, 1(3), 262-271.

Brodley, B. (1986). Client-centered therapy: What is it? What is it not? Unpublished manuscript.

Brodley, B. (1988). Untitled article. Renaissance, 5(3-4), 1-2.

Brodley, B. (1989). Cart Rogers’s responses to questions in client-centered interviews. Unpublished manuscript.

Cain, D. (1986). What does it mean to be "person-centered"? Person-Centered Review, 1(3), 251-256.

Cain, D. (1989). The paradox of nondirectiveness in the person-centered approach. Person-Centered Review, 4(2), 123-131.

Edwards, 1. (1982). Ethics without philosophy: Wittgenstein and the moral life. Tampa, FL: University of South Florida Press.

Grant, B. (1985). The moral nature of psychotherapy. Counseling and Values, 29(2). 141-150.

Grant, B. (1986). Etica y psicoterapia: Prolegomeno a una terapia centrada en el cliente sin una teoria centrada en el cliente [Ethics and psychotherapy: A prolegomenon to client-centered therapy without client-centered theory]. Revista de Psiquiatria y Psicologia Humanista, 17(4), 82-86.

London. P. (l986). The modes and morals of psychotherapy. New York: Hemisphere.

Plemenatz, J. (1 973). Liberalism, in P. Weiner (Ed.), Dictionary of the history of ideas (Vol. III, pp. 36-6 1). New York: Scribner.

Raskin, N. (1988). Untitled article. Renaissance. 5(3-4), 2.

Rogers, C. (195 1). Client-centered therapy. New York: Houghton Mifflin.

Rogers, C., & Segel, R. (Producers). (1955) Psychotherapy in process: The case of Mrs. Mun [Film]. Transcribed in J. Bozarth (1984), Beyond reflection: Emergent modes of empathy. in R. Levant & J. Shlien (Eds.), Client-centered therapy and the person-centered approach: New directions in theory, research, and practice (p. 62). New York: Praeger.

Rogers, C. (1987). Client-centered? Person-centered? Person-Centered Review, 2(1), 11-14.

Szasz, T. (1978). The myth of psychotherapy. Garden City, NY: Anchor Press.

Tomlinson, T., & Whitney, R. (1970). Values and strategies in client-centered therapy: A means to an end. In Hart & Tomlinson (Eds.), New directions in client-centered therapy.Boston:

Houghton Mifflin.

Van Belle, H. (1980). Basic intent and therapeutic approach of Carl R. Rogers. Toronto: Wedge.

Watson, G. (1958). Moral issues in psychotherapy. American Psychologist, 13(9), 574-576.

Weil, S. (1951). Waiting for God. New York: Harper & Row.