To download a pdf version of this article, click here.

Boundary Violations: Do Some Contended Standards of Care Fail to Encompass Commonplace Procedures of Humanistic, Behavioral and Eclectic Psychotherapies?

Psychotherapy, 1997, vol. 34, number 3, pp. 238-249

Martin H. Williams
Private Practice
Redwood Estates, California


Hugging, dining with, self-disclosing to, or making house calls to patients are among behaviors which have been termed "boundary violations" in psychotherapy. Although authors have asserted that boundary violations are both harmful and beneath the standard of care, some of the activities in question are consistent with the ethical practice of humanistic and behavioral psychotherapies, as well as with eclectic approaches deriving from those schools. Theoretical statements, survey research, and case examples are used to elucidate concerns about maintaining metaphorical boundaries in psychotherapy and to demonstrate that psychotherapy is diverse with respect to the behaviors at issue. It is concluded that even scrupulous humanistic, behavioral, and eclectic practitioners might appear to practice negligently by virtue of engaging in behaviors which some consider to be boundary violations and that innovative practice might be stifled by risk management concerns.

Copyright (1997) by Division of Psychotherapy of the American Psychological Association

Boundary Violations: Do Some Contended Standards of Care Fail to Encompass Commonplace Procedures of Humanistic, Behavioral and Eclectic Psychotherapies?

One matter that may be discussed at malpractice proceedings--before a civil court, licensing board or ethics committee--concerns the inappropriate crossing of boundaries by the psychotherapist. These “boundary violations” include, but are not limited to: hugging, dining with, self-disclosing personal information or feelings to, making house calls to, exchanging gifts with, engaging in non-sexual socializing with, or lending books to patients during treatment. The most egregious boundary violation is sexual intercourse during treatment--something that virtually all practitioners condemn. However, as discussed below, some contributors to the ethics literature assert that the occurrence of less severe boundary violations, like self-disclosure or gift-giving, lends validity to plaintiffs’ contentions that sexual activity--denied by the therapist--must have actually occurred, and some contend that a series of minor boundary violations shows a pattern of negligence and justifies licensing sanctions or financial settlements even in the absence of sexual activity.

The present paper is intended to show that two distinct and contradictory positions exist regarding boundaries in psychotherapy. On the one hand, authors argue that ethics concerns dictate a need for careful maintenance of boundaries as well as a need to sanction practitioners who violate. On the other hand, the traditions and practices of some forms of psychotherapy dictate that certain boundaries be routinely crossed. These two positions do not coexist well. For example, the possibility exists that a humanistic or behavioral psychotherapist, who practices in accord with published techniques and traditions deriving from those theoretical orientations, may appear to violate community standards should his or her work be adjudicated by a licensing board consultant or plaintiff’s expert who holds a conservative view of boundaries. To the extent that this occurs, behavioral, humanistic, or eclectic practitioners may be at risk for licensing sanctions or financial penalties should their practices be scrutinized by a sanctioning agency--regardless of the validity of the original allegations.

The discussion below is organized into three main parts. First, the viewpoint which holds that boundaries should be carefully maintained will be discussed. Second, humanistic and behavioral viewpoints on boundaries will be respectively presented. Next, surveys which depict therapists’ boundary related behaviors will be used to identify the extent to which some therapists cross boundaries. Finally, the issue of risk management--changing one’s practice to avoid the appearance of wrongdoing--will be discussed, and conclusions will be drawn .

The Logic of Boundary Concerns

Several authors have contributed arguments which emphasize the importance of maintaining boundaries in psychotherapy (Atkins and Stein, 1993; Bennett, Bricklin and VandeCreek, 1994; Brodsky, 1989; Brown, 1994; Borys, 1994; Epstein and Simon, 1990; Epstein, Simon, and Kay, 1992; Folman, 1991; Gabbard, 1994; Gechtman, 1989; Goisman and Gutheil, 1992; Gottleib, 1993, 1994; Gutheil, 1989, 1994; Gutheil and Gabbard, 1993; Johnston and Farber, 1996; Katherine, 1991; Notman and Nadelson, 1994; Pope, 1989,1994; Simon, 1991, 1992, 1994, 1995; Smith and Fitzpatrick, 1995; Sonne, 1989, 1994; Strasburger, Jorgenson, and Sutherland, 1992). The quotations which follow exemplify aspects of this viewpoint. Pope (1994), for example, states the following:

Establishing safe, reliable, and useful boundaries is one of the most fundamental responsibilities of the therapist. The boundaries must create a context in which therapist and patient can do the work of therapy. (p. 70) [emphasis added]

Katherine (1993) expresses a similar view:

A boundary violation is committed when someone knowingly or unknowingly crosses the emotional, physical, spiritual, or sexual limits of another. Whether a violation is intended or not, whether it is committed out of ignorance or malice, it is still a violation. It still harms (p. 86).

More subtle violations occur when the caregiver initiates interaction that is only appropriate among peers. Your doctor is not your peer. Your therapist is not your peer (p. 88).

Professional distance between therapist and client gives the client her greatest safety...

Social contact between therapist and client muddies the boundaries. I used to go to client weddings, but now I don’t... My presence in other contexts confuses the fact that I have a special, unique, protected role in her life with specific limitations (p. 91).

Johnston and Farber (1996) in summarizing Langs’ conservative psychoanalytic view of boundaries state:

Consequences of poor boundary management include the communication of the therapist’s intrapsychic conflicts to the patient, the contamination of the transference and consequent interpretations, the dissolution of the therapeutic “hold,” and the possibility of inappropriate gratification resulting from countertransference problems (p. 392).

Finally, Simon (1991) asserts the following regarding consequences of boundary violations:

The boundary violation precursors of therapist-patient sex can be as psychologically damaging as the actual sexual involvement itself. Unfortunately, professional ethics codes are usually silent concerning the specific boundary violations that often precede therapist sexual misconduct. (p.614) [emphasis added]

Most of the contributions listed above have included acknowledgment of the difficulty of establishing specific prohibitions on boundary crossing, considering that each case and each practice style is unique. As Smith and Fitzpatrick (1995) have written:

The effects of crossing commonly recognized boundaries range from significant therapeutic progress to serious, indelible harm. The issues are further complicated by the wide range of individual variation that exists in a field where what is normal practice for one clinician may be considered a boundary violation by another. (p. 505)

A similarly catholic approach appears to be espoused by Simon (1992) when he writes:

...considerable disagreement exists among psychiatrists concerning what constitutes treatment boundary violations. The therapy techniques of one therapist may be anathema to another therapist who considers such practices as clear boundary violations. (P. 269)

... boundary excursions inevitably occur in almost every therapy (p. 286).

Finally, Brown (1994) exemplifies how attempted solutions to the problem of therapeutic boundaries can inadvertently become codified as rules. She states:

What I have found, to my dismay, is that when I have shared strategies that evolved into solutions that work for me, carefully framing as my solutions and opinions rather than “the rule,” I find myself quoted two journals articles later as saying that “such and so behavior is not okay” (p. 30).

However, one can also find in the literature statements that would seem to imply that widespread agreement exists concerning a prohibition of certain boundary crossing behaviors, as well as the need to sanction them. For example, Strasburger et al (1992) write:

The slippery slope of boundary violations may be ventured upon first in the form of small, relatively inconsequential actions by the therapist such as scheduling a “favored” patient for the last appointment of the day, extending sessions with the patient beyond the scheduled time, having excessive telephone conversations with the patient, and becoming lax with fees. Violations can involve excessive self-disclosure by the therapist to the patient... Gifts may be exchanged. The therapist may begin to direct the patient’s work and personal life choices... Meetings may be arranged outside the office for lunch or dinner.

... Notice that in this scenario, the therapist has not touched the patient, nor has the therapist said or done anything that is overtly sexual. The treatment, however, has already become compromised, and the therapist may be found liable civilly. The therapist is also vulnerable to action by a licensing board, should the patient wish to make a complaint. (P. 547 [emphasis added])

In a similar vein, Simon (1995) has suggested that agreement exists among diverse schools of psychotherapy regarding boundaries. He states, “The boundary guidelines..., with appropriate clinical modifications, are a unifying element in the over 450 different forms of psychotherapy currently in existence” (p. 90). Simon provides the following list of “treatment boundary guidelines” which appears in several of his publications:

Maintain therapist neutrality. Foster psychological separateness of patient. Obtain informed consent for treatment and procedures. Interact verbally with clients. Ensure no previous, current, or future personal relationships with patients. Minimize physical contact. Preserve relative anonymity of the therapist. Establish a stable fee policy. Provide a consistent, private, and professional setting. Define length and time of sessions. (1994, p. 514)

Boundary violations versus transference abuse. Frequently mentioned in civil and licensing board actions to make reference to a standard of care, the term “boundary violation” has become prevalent during the past ten years. Although psychotherapeutic boundaries have been conceptually useful to some practitioners for many years (e.g., Langs, 1974; Stone, 1976), the concept of boundary violations has recently supplanted that of “transference abuse” (e.g., Gabbard & Pope, 1988; Pope & Bouhoutsos, 1986) in the malpractice arena as one which describes unethical, overly close relations between therapist and patient. In Pope and Bouhoutsos’(1986) seminal work on therapist-patient sex, as in Bates and Brodsky’s (1989) extensive case study, no mention is made in the index of the word “boundaries.” In contrast, Pope’s (1994) book on therapist-patient sex contains no fewer than 14 index citations under this term, perhaps giving some indication of the increasing popularity of this concept.

Transference abuse had been a problematic concept in malpractice litigation because of its overt theoretical linkage to psychoanalysis, causing the concept to be meaningless or offensive to numerous practitioners. As Gutheil (1989) has pointed out:

It seems that professionals who belong to a school of thought that rejects the idea of transference, behaviorists, or psychiatrists who provide only drug treatment, are being held to a standard of care they do not acknowledge. (p. 31)

The greater courtroom utility of boundary violations, in contrast to that of transference abuse, may rest on the universality of the former concept. It is free from the criticism that it derives only from psychoanalysis with relevance only to practitioners espousing allegiance to that school.

A therapeutic rationale for careful boundary maintenance. An argument relative to the import of maintaining boundaries in psychotherapy might be outlined as follows:

  1. Psychotherapy is a powerful tool that can evoke powerful emotions.
  2. Reported victims of therapist-patient sexual involvement often tend to be women and the perpetrators men.
  3. Psychotherapy patients tend to regress to infantile and vulnerable states.
  4. This regression places the patient in a situation which is reminiscent of a family, in which the child is vulnerable to the parent.
  5. Consequently, the male therapist, as metaphorical father, must insure that the vulnerability of the female patient, his metaphorical daughter, is not exploited. (This is equally true regardless of gender of patient or therapist, although most litigation involves this metaphorical “father-daughter” dyad (e.g., Pope, 1990a, 1990b).
  6. Steps must be taken to give the patient clear messages that any vulnerability will not be exploited.
  7. These steps include establishing boundaries between the therapist and patient, and between the professional and the social. These boundaries include clearly demarcating starting and stopping times of sessions, having no social contacts with patients, not touching patients, not disclosing personal information to the patient, always billing patients, and doing all related activities to continually remind the patient and make abundantly clear that this is a professional, rather than social or sexually intimate, relationship.
  8. Patients, because of their regressed and vulnerable state, may attempt to initiate behaviors that are more social or sexual than therapeutic. The therapist always bears the burden of preventing this from occurring. Because the patient’s judgment may be impaired, the therapist bears the exclusive responsibility for setting limits.
  9. Some patients, because of a prior history of childhood or adult sexual abuse may become confused regarding the difference between that which is therapeutic and that which is social or sexual. Consequently, even apparently harmless forays into social or personal activity, which are not part of treatment, must be avoided to avoid confusing the patient as to the nature of the relationship.
  10. In light of all of the above, even seemingly trivial “boundary violations,” such as giving a patient a ride should her car be broken, can indicate a disregard for the import of boundaries and can be a sign of negligence. Such negligence can approach metaphorical incest considering the father-daughter metaphor which is attributed to these therapeutic dyads.

From this point of view, one of the most significant curative aspects of psychotherapy may be that the patient undergoes his or her initial lifetime experience of an interpersonal relationship in which boundaries are adequately maintained. By virtue of this, the patient comes to learn that the therapist is a separate person, not dependent on the patient for love or any other emotional needs, and that it is possible for the patient to have other, mature, adult, independent and non-exploitative relationships. In addition, the patient learns that there can be healthy relationships such as psychotherapy in which the stated goals which gave rise to the relationship are in fact carried out without manipulation, hidden agendas or abuse. While this approach shares many elements with the ground rules of classical psychoanalysis (Greenson, 1967; Gutheil and Gabbard, 1993), it is also informed by modern concepts such as family dysfunction and the incest survivor movement (e.g., Russell, 1986), feminist therapy (e.g., Brown, 1994), and has become a cornerstone of the addiction-recovery movement (e.g., Katherine, 1993).

Although anecdotal reports provide support for arguments which hold that even non-sexual boundary violations lead to patient harm, e.g., Simon (1991), the reliable occurrence of such harm has not been established. The difficulties in systematically associating resultant harm with previous boundary violations of a sexual nature have been discussed elsewhere (Pope and Bouhoutsos, 1986; Williams, 1992; 1995).

Humanistic Psychotherapy: A Solution to a Different Problem

Humanistic psychotherapists, and those eclectic therapists influenced by humanistic concepts and methods, practice from viewpoints which lead them routinely and intentionally to engage in what some might argue are grossly negligent boundary violations. Humanistic psychotherapy includes, but is not limited to, approaches such as Gestalt (e.g., Fagen and Shepherd, 1970; Perls, 1969; Smith, 1976), encounter group (e.g., Korchin, 1976; Rogers, 1970), Transactional Analysis (e.g., Berne, 1961; Berne, 1964), Existential (e.g., Yalom, 1980) and Client-Centered psychotherapy (e.g., Rogers, 1951; Rogers, 1961). Bugental’s work (1986, 1987) exemplifies a more recent, eclectic approach to humanistic psychotherapy.

It is of great significance that the humanistic movement, as described by Korchin (1976) and others, has been devoted not to maintaining but to tearing down the boundaries between therapist and patient. Korchin describes the essence of Humanistic Psychology as follows:

Above all else, therapy involves an authentic encounter between two real individuals, free of sham and role-playing, rather than technical acts of an interpretive, advising, or conditioning sort. (1976, p. 352)

The following quotation from Bugental (1987) provides some insight into the humanistic ethos with respect to boundaries:

Long-term therapy of some depth inevitably involves times of warm communion and times of great stress--for both participants. Living through these together has a true bonding effect which is not always recognized by those who teach or practice more objective modes. Nevertheless, therapist and patient often have what can only be called a love relationship, which is by no means simply a product of transference and countertransference. Patient and therapist are two human beings, partners in a difficult, hazardous, and rewarding enterprise; it is unreal to expect otherwise. (P. 258).

Humanistic practitioners might argue that for the therapist to be self-disclosive makes the patient feel more equal to, rather than inferior to, the therapist. It allows the patient to see that all people have failures and other unresolved matters in their lives, and that there is no essential difference, in fact, between those people who are psychotherapists and those people who are patients. In discussing self-disclosure with patients, Bugental advocates, “First and foremost: strict honesty is required” (1987, p. 143.)

In the context of group psychotherapy, Vinogradov and Yalom (1990), offer the following recommendations:

Group psychotherapists may--just like other members in the group--openly share their thoughts and feelings in a judicious and responsible manner, respond to others authentically, and acknowledge or refute motives and feelings attributed to them. In other words, therapists, too, can reveal their feelings, the reasons for some of their behaviors, acknowledge the blind spots, and demonstrate respect for the feedback group members offer them. (p. 198)

In a similar vein, humanists might argue that many other activities which were excluded from classical psychoanalytic technique can take place without harming the therapeutic process, and, in fact, will help it. Some humanists, like Perls (e.g., 1969), largely defined their methods in contrast to what they considered to be stagnant psychoanalytic doctrine and unnecessary psychoanalytic prohibitions. Consequently, patients are urged to call humanistic therapists by their first names, therapists might socialize with patients--especially in the context of a personal growth center or retreat, gift-giving is not unheard of, sessions might take place outside of the office, and hugging, in sharp contrast to analytic reserve, becomes the order of the day. Although these activities are not specifically prescribed by any particular humanistic theory, they flourish in the open, innovative, and often spontaneous climate of humanistic psychotherapy. This excerpt from Jourard (1971) provides specific examples of the wide range of behaviors that might be part of a humanistic practice style:

In the context of dialogue I don’t hesitate to share any of my experience with existential binds roughly comparable to those in which the seeker finds himself (this is now called “modeling”); nor do I hesitate to disclose my experience of him, myself, and our relationship as it unfolds from moment to moment... I might give Freudian or other types of interpretations. I might teach him such Yoga know-how or tricks for expanding body-awareness as I have mastered or engage in arm wrestling or hold hands or hug him, if that is the response that emerges in the dialogue.

I do not hesitate to play a game of handball with a seeker or visit him in his home--if this unfolds in the dialogue (p. 159).

Humanistic approaches like Jourard’s were part of the training of many psychotherapists during the 1970’s. To what extent individuals trained in this approach continue to practice largely in this same manner is unknown, although it appears, at least to the present author, that many of today’s licensing board and malpractice defendants were trained during the 1960’s and 1970’s, when tearing down boundaries was encouraged.. Survey research by Pope, Tabachnick and Keith-Spiegel (1987) found that 14% of practitioners recently identified themselves as “humanistic,” and Borys and Pope (1989) found a relationship between theoretical orientation and boundary-related behaviors. Further, one finds no indication in the humanistically-oriented journals of a radical shift away from previous styles. In contrast, recent examples, like the following from the Journal of Humanistic Psychology, indicate that similar, boundary-blurring themes continue to prevail within the realm of humanistic psychotherapy:

At her moment of deep despair, this young woman found in her core the courage and strength to call the therapist to open herself to her as one human being to another. The therapist returned her embrace without words.

This moment of dialogue did not interfere in any way in the therapist’s observation and evaluation. On the contrary--it added most valuable data that a moment later the therapist could bring into her psychological-observational stance (Kron & Friedman, 1994, p. 71).

Recent humanistic writings have become less strident regarding the need to challenge therapeutic boundaries in contrast to humanistic publications of two or three decades ago. One wonders whether humanistic practice has become more conservative with respect to boundaries, or whether humanistic practitioners--fearing lawsuits or ethics enforcement--simply have become less forthcoming about their boundary-crossings. A relatively recent study by J. Simon (1990), although based on a small sample, indicates that the most self-disclosive therapists considered their mentors to be Albert Ellis, Carl Rogers, Fritz Perls and Werner Erhard. This supports the notion that the basic humanistic ethos regarding boundaries may still be intact.

Recently, some psychoanalytic psychotherapists have adopted a liberal position regarding the boundary of self-disclosure. Some analysts now seek a “shared experience” between analyst and analysand This requires authentic self-disclosure by the analyst, making aspects of this psychoanalytic approach indistinguishable from humanism, as illustrated by the following quotation from Fisher (1990):

Further, I contend that for a genuine encounter to occur between patient and therapist, and for authentic growth in intimacy to emerge (which is at the heart of the need for therapy to begin with) a truly shared experience must take place. Again, the belief herein suggested is that the encounter between patient and therapist (like that between parent and child) should take place between (psychological) equals: between the co-participants of dyadic psychotherapy. Lastly, that the sharing of experiencing, which leads to intimacy, is achieved through the process of (mutual) self-disclosure (p. 14).

One may surmise that the revised APA ethics code (1992), statutory changes, and changes in professional liability coverage (e.g., Pope, 1994) have had an impact on overt sexual boundary violations committed by humanistic therapists. Past examples of the humanistic approach to sexual boundaries, reckless at minimum and unethical by today’s standards, have been provided by Finney (1975), regarding client-centered therapy, and Shepard (1975), regarding Gestalt therapy.

Behavior Therapy: Another Rebellion Against the Culture of Psychoanalytic Technique

Behavior therapy is another approach to psychotherapy that arose, in part, as a rebellion against psychoanalysis (Korchin, 1976) Practitioners of behavior therapy, and eclectic therapists who have been influenced by this approach, might routinely commit what many today would call boundary violations. Many behavior therapists have never been part of the culture of psychoanalysis, and many have gone through their graduate school or residency training free from exposure to even the most basic ground rules of psychoanalysis, including those pertaining to boundaries. Indeed, should a behavior therapist-in-training encounter a list of psychoanalytic practice guidelines, the result might be little more than derisive laughter. Arnold Lazarus, the noted behavior therapist, has referred to psychoanalysis as a “speculative theory” and has called transference a “myth” (A.A. Lazarus, personal communication, September 27, 1990).

In practice, behavior therapists have cogent reasons to counter even the most basic of classical psychoanalytic dicta. For example, the analytic blank screen allows the patient to project fantasies as part of the transference. For this reason, classical psychoanalysts would not be self-disclosive; the less the patient actually knew about the analyst, the more the patient could project. Behavior therapists, in contrast, use the concept of modeling, which holds that the patient might learn from the therapist by copying behaviors or even attitudes and that this could be curative. To practice modeling, the behavior therapist might need to be self-disclosive. For example, a behavior therapist might tell the patient how that therapist overcame his or her own fear of flying (e.g., Ellis, 1977). In addition, while the psychoanalyst might prefer to leave the patient guessing about the degree and nature of his or her training--another source of material to analyze--the behavior therapist might carefully spell out the nature or his or her training and experience, to better be seen as a high status model, capable of influencing behavior. Indeed, a behavior therapist might, upon request, disclose his or her high success rate at treating similar cases to that for which the patient seeks help, while a classical psychoanalytically oriented practitioner would interpret such a request only in terms of the patient’s psychodynamics.

From the viewpoint of behavior therapy, there has been little concern about socializing with patients outside of therapy sessions. Some behavior therapists have argued that they merely apply a set of technical procedures and, consequently, require no special rules for the therapist-patient relationship beyond those rules that would apply to one’s relationship with one’s architect (e.g., Marquis, 1972). As Marquis states, “The resulting relationship is one in which I have felt quite comfortable having good friends as clients and good clients as friends” (1972, pp. 48-49).

Nothing in the theory of behavior therapy would or should preclude socializing with patients, taking meals with them, giving them gifts, or treating them at their homes, schools or offices. Hugging patients might increase the therapists’ potency as a reinforcer for the patient, and, thus, might be supported theoretically, and on the subject of touching patients, Marquis advises:

The depth of relaxation is tested carefully by observing the client visually and by manipulating the extremities...

Psychologists and social workers are often reluctant to touch a client, but even most clients with phobias about being touched accept the procedure easily (p. 55) [emphasis added].

A behavior therapist’s viewpoint on self-disclosure is provided in the recent handbook by Burns (1990), a strong proponent of the currently popular cognitive behavior therapy. Burns advises, “Let the patient know how you feel about what he or she is saying. This will make you appear more genuine and real” (p. 514). He provides the following case example to illustrate his use of self-disclosure:

I told Ronda that I felt inadequate. I said I felt as if every sentence that came out of my mouth was wooden and useless to her. I said that although I usually felt I had something to offer, it didn’t seem that way today. I told her I felt excluded and shut out, and that I felt angry with her. I said I wanted to give her something positive and I believed that the therapy could be successful, but I felt thwarted in my efforts (p. 521).

In a similar vein, Dryden (1990) argues that therapist self-disclosure is an integral part of Rational Emotive Therapy (RET). Dryden states;

Given that the effective RET therapist would accept herself for her errors and flaws and for past and present emotional disturbances, she will, as often as is therapeutically advisable, show her client how she upset herself about experiences similar to those with which her client is concerned and how she used RET to overcome such emotional disturbances (p. 66).

Lazarus (1994a, 1994b) recently took issue with those who would strictly enforce boundary maintenance. He provided numerous examples from his practice in which either treatment issues or common politeness dictated a need to dine with, self-disclose to, socialize with or befriend patients, and argued that such activities had either no ill effects on treatment or were helpful. Invited discussants reacted sharply, presenting Dr. Lazarus as naive and foolhardy and generally, lucky to have managed his boundary crossings without adverse consequences (e.g., Gabbard, 1994; Gutheil, 1994), or they suggested that while he might possess the skill to negotiate these boundary crossings successfully, others who would follow his example might not (Bennett, Bricklin & VandeCreek, 1994). None of the discussants seemed to accept his overriding point that “one of the worst professional or ethical violations is that of permitting current risk-management principles to take precedence over humane interventions” (1994, p. 260) [emphasis added].

Goisman and Gutheil (1992) provide a poignant example of what might occur when commonplace behavior therapy procedures are held up to psychoanalytic scrutiny regarding boundary maintenance:

We are aware of a case currently in litigation where a number of the charges against an experienced behavior therapist flowed from the testimony of a psychoanalytically trained expert witness, who faulted the behavior therapist for assigning homework tasks to patients, hiring present and former patients for jobs in psychoeducational programs and other benign interventions, and performing a sexological examination and sensate focus instructions in a case of sexual dysfunction. From a psychoanalytic viewpoint all of these would likely constitute boundary violations of a potentially harmful sort, but from a behavioral viewpoint this is not the case. The legal system in this lawsuit had some difficulty, as is commonly the case, in grasping the distinctions between therapies and the variations of boundary norms appropriate to each type of treatment. (p. 538)

As with humanistic approaches, some of the most direct challenges to therapeutic boundaries by behavior therapists were published some years ago, leaving the reader to wonder about the extent to which practitioners who were trained during a more liberal era may have moved towards more careful boundary maintenance in their contemporary practices.

Diversity of Practice

Research has demonstrated that practice is diverse with respect to boundaries. Table 1 shows the self-reported rates of some of the now controversial activities that Pope, Tabachnick and Keith-Spiegel (1987) found in the practices of a national sample of psychologists. Psychologists were asked to rate the frequency with which various boundary-blurring activities occurred in their practices, e.g., lending money to patients, socializing with them, asking favors of them, and touching them in various ways. As Table 1 indicates, many of the controversial boundary related activities were surprisingly prevalent in the practices of psychologists. An interesting aspect of this sample is that humanistic approaches only comprised 14% of the psychologists surveyed, behavioral only 2.6%, while practitioners identifying themselves as psychodynamic comprised 32.9%. Considering the theoretical positions taken by the various schools with respect to these boundary-blurring activities, one would suspect that homogeneous samples of either humanistic or behavioral practitioners might yield even higher reported rates. As it is, these authors found that the use of self-disclosure, at least on rare occasions, is an “almost universal” aspect of psychotherapy (p. 998) [emphasis added]. This is significant in view of Simon’s (1994) injunction, noted above, that one must “preserve relative anonymity of the therapist,” and it becomes especially significant considering the assertion by Simon (1991) and others that the degree of self-disclosure in psychotherapy is a better predictor of sexual involvement with patients than is the degree of non-sexual touching.

A similar study by Borys and Pope (1989) also supported the notion that practice is diverse with respect to boundaries. In this study, an interdisciplinary national sample of psychiatrists, psychologists, and social workers was surveyed. Table 2 shows the rates of occurrence of boundary blurring behaviors in this sample. As was the case with the Pope et al study (1987), these rates were much higher than one would expect were Simon’s guidelines, listed above, to be widely observed. Both tables show that gift giving, dining together, non-sexual touching, asking favors and self disclosure are not uncommon in psychotherapeutic practice. Borys and Pope’s findings also provided support for the present contention that maintenance of boundaries is related to therapeutic orientation, with a statistically significant relationship being found between “frequency of reported social involvements with clients...and theoretical orientation” (p. 288). Statistically significant differences in behavior were also found when psychodynamic practitioners were contrasted with humanistic ones.

A recent study by Johnston and Farber (1996) surveyed psychologists’ attitudes and behaviors regarding “everyday” boundaries in psychotherapy, such as starting and ending sessions on time, requiring timely payment of fees, and demanding adherence to specified appointment times. Although this set of studied boundaries were subtle, and less likely to constitute behaviors that were of questionable ethics, therapists’ behavior was once again diverse. Findings indicated that:

“patients make relatively few demands and psychotherapists accommodate them most of the time. This finding stands in opposition to the generally accepted image of the psychotherapist standing firm in the face of persistent attempts by the patient to challenge existing boundaries and suggests a spirit of cooperation and good faith under emphasized in theoretical writings (p. 397).

The particular set of boundaries investigated by Johnston and Farber did not show a relationship between boundary maintenance and any therapist variables, such as theoretical orientation. However, it should be noted that “humanistic” was not used as one of the choices for assigning theoretical orientation.

A generation gap. The possibility exists that attitudes are changing, and that psychotherapists who have been trained within the past few years may tend to more closely adhere to the boundary guidelines promulgated by Simon (1994) and others. Indeed, the survey by Borys and Pope (1989) found a significant relationship between therapists’ years of experience and their ratings concerning the ethics of maintaining dual professional roles with patients, with less experienced therapists more likely to consider dual roles unethical. Perhaps professionals’ attitudes regarding the ethics of certain boundary-related behaviors have changed dramatically over the past two decades. One consequence of this would be a schism between older and younger practitioners, with the older practitioners appearing to the younger ones to be less ethical. This possible interaction between experience and ethics is a matter that requires additional research. One implication of such a “generation gap” would be that a younger expert witness, in adjudicating behaviors that are not specifically enumerated by an ethics code, might honestly testify that an elder practitioner seems to have practiced negligently, when, in fact, the elder practitioner may have practiced well within the standard of care of the cohort with which he or she trained.

Avoiding the Appearance of Wrongdoing

An additional, very significant contribution to conceptualizations of boundaries has been proposed by Gutheil and Gabbard (1993). These authors reviewed a wide range of “boundary crossings,” which may or may not constitute boundary violations, and they argued that even certain boundary crossings which are justified and consistent with good care should be eschewed on the basis of their possible adverse appearance in court. This process of avoiding behaviors, not because they are wrong, but only because they appear wrong, is known as “risk management.” Gutheil and Gabbard state that plaintiff’s attorneys will make these crossings appear to the jury to be incontrovertible evidence of wrongdoing. Thus, regarding the question of scheduling patients late in the day, they state:

In the fog of uncertainty surrounding sexual misconduct (usually a conflict of credibilities without a witness), this factor has gleamed with so illusory a brightness that some attorneys seem to presume that because the patient had the last appointment of the day, sexual misconduct occurred! (p. 191)

On the subject of self-disclosure of matters which include information concerning the therapist’s family, they advise that such behavior:

may be used by the legal system to advance or support a claim of sexual misconduct. The reasoning is that the patient knows so much about the therapist’s personal life that they must have been intimate.(p. 194)

Finally, in discussing a case in which a psychiatrist, who was ultimately exonerated, was found guilty of sexual misconduct, they state the following:

recent court decisions suggest a trend toward findings of liability for boundary violations even in the absence of sexual contact. On this basis, the risk-management value of avoiding even the appearance of boundary violations should be self-evident. (p. 189)

Gutheil and Gabbard may be correct that avoiding behaviors in psychotherapy which, although harmless and appropriate, might appear negligent is a good risk management strategy. However, one may question, in terms of both principle and practicality, whether practitioners ought to change their deeply held beliefs and established methods to avoid “looking bad.” As a practical concern modifying one’s methods in the interest of risk management becomes a more difficult decision to the extent that one’s practice relies on boundary-crossing techniques, as it becomes easier to extent that one’s practice does not. Indeed, some of the distinguishing characteristics of one’s practice might be its boundary-crossing procedures. Prospective clients seeking treatment, or colleagues making referrals, might seek out a given practitioner only because that practitioner truly is a dyed-in-the wool behaviorist or humanist, known to use methods specific to those approaches. Why, for example, would a patient seek treatment at a “humanistic growth center” if the psychotherapy offered were largely indistinguishable from that found in other offices? Clearly, there would be professional or business risks for some practitioners were they to adopt the risk management strategies proposed by Gutheil and Gabbard.

As a final point, it should be noted that Gutheil and Gabbard (1993) have also advocated restraint in court regarding accusations of boundary violation, urging that attention be paid to the context in which the boundary crossing occurred prior to drawing conclusions regarding harmfulness or standard of care. However, as a practical matter, whether a boundary crossing is perceived as a boundary violation is very much in the eye of the beholder. My own consulting experience indicates that plaintiff and defense experts are capable of arriving at opposite, but equally firm, conclusions when confronted with the same clinical data. I firmly believe that this is not the result of the use of “hired guns” as expert witnesses who will testify on one side or the other for a fee. Instead, these experts appear to be sincere individuals who, like the authors quoted above, simply hold a range of divergent views of what constitutes appropriate psychotherapy.

The plaintiff’s purpose in establishing that non-sexual boundary violations occurred. In some legal cases the occurrence of sexual negligence is not contested, yet the occurrence of other boundary crossings is nevertheless disputed in court. The question arises as to why a plaintiff would bother to establish that self-disclosure, for example, had taken place in the context of psychotherapy that was known to have been grossly negligent because it had included sexual intercourse. The answer concerns the nature of malpractice insurance: Most, if not all, of today’s malpractice insurance policies exclude coverage for damages resulting from therapist-patient sexual involvement. Thus, a finding for the plaintiff in a civil case that awards damages due to sexual transgressions would not result in payment by the insurance carrier. Many defendants lack the resources to pay such awards which can exceed a million dollars. To prevent such fruitless courtroom victories, plaintiffs often file complaints intended to establish, consistent with Simon’s (1991) writings cited earlier, that the non-sexual boundary violations which occurred were harmful in and of themselves, constituted negligent psychotherapy in and of themselves, and would have, with a reasonable degree of certainty, resulted in equivalent damages to those observed even had sexual contact not taken place. This strategy places the insurance carrier in a position such that the damage award will be paid, even when sex was involved, because the sexual involvement had been shown not to be the only, or even chief, source of harm to the patient. In fact, one strategy would have the plaintiff intentionally omitting instances of documented sexual contact as part of the complaint, thereby enhancing the argument for insurance coverage as negligent psychotherapy.


With the above in mind, it may be argued that strict boundary maintenance, although supported by some authors, should not constitute a minimum standard of care. While there is near unanimity among practitioners that therapist-patient sexual involvement is unethical during treatment (Pope et al, 1987), the other boundary related behaviors which have been discussed are neither specifically prohibited by ethical guidelines, nor are they universally recognized as deviations from the standard of care.

The complex endeavor of psychotherapy, which includes dozens of theoretical perspectives, several different licensed professions with disparate training requirements, and innumerable continually evolving techniques, cannot facilely be delimited by an overly simple and restrictive set of rules (e.g., those proposed by Simon, 1994), deviation from which would seem to place one beneath the standard of care. This is no better exemplified than by the text of the Distinguished Professional Contribution Award presented by the American Psychological Association to Kenneth Pope in 1994 (“Awards,” 1995). Although Pope’s substantial contributions to the literature have emphasized the importance of maintaining clear boundaries in psychotherapy (e.g., 1989, 1994), his award biography describes several instances, in his treatment of a female patient, of what Gutheil and Gabbard might term boundary crossings (although not boundary violations). The exemplary treatment carried out by Pope had included daily meetings without fee and his arranging for a personal friend of his to loan the patient money and to provide her with an airline ticket and a place to stay. In the context of the particular case, these boundary excursions appeared to be both humane and sensible. However, some practitioners might, in the interest of risk management, avoid making similar modifications. Clearly, any efforts that would discourage the sort of flexibility of practice exemplified by Pope’s case serve no end but to increase the likelihood that undue damage awards will be paid, that dedicated clinicians will be improperly sanctioned, and that the practice of psychotherapy will stagnate as practitioners become more concerned with risk-management than with innovation.


Atkins, E.L. & Stein, R. (1993). When the boundary is crossed: a protocol for attorneys and mental health professionals. American Journal of Forensic Psychology, 11, 3-21.

Awards for Distinguished Professional Contributions. (1995). American Psychologist, 50, 236-247.

Bates, C.M. & Brodsky, A.M. (1989). Sex in the therapy hour: a case of professional incest. New York: Guilford.

Bennett, B.E., Bricklin, P.M. & VandeCreek, L. (1994). Response to Lazarus’s “how certain boundaries and ethics diminish therapeutic effectiveness. Ethics and Behavior, 4, 263-266.

Berne, E. (1961). Transactional analysis in psychotherapy. New York: Grove Press.

Berne, E. (1964). Games people play. New York: Grove Press.

Borys, D.S. & Pope, K.S. (1989) Dual relationships between therapist and client: a national study of psychologists, psychiatrists, and social workers. Professional Psychology: Research and Practice, 20, 283-293

Brodsky, A.M. (1989). Sex between patient and therapist: psychology’s data and response. In G. Gabbard (Ed.) Sexual exploitation in professional relationships. Washington, D.C.: American Psychiatric Press, pp. 15-26.

Brown, L.S. (1994). Boundaries in feminist therapy: a conceptual formulation. Women and Therapy, 15, 29-38.

Brown, L.S. & Walker, L. Feminist therapist perspectives, In G. Stricker & M. Fisher (Eds.), Self-disclosure in the therapeutic relationship. New York: Plenum, pp. 135-154.

Bugental, J.F. (1986). Existential-humanistic psychotherapy. In I.L. Kutash & A. Wolf (Eds.), Psychotherapist’s casebook, San Francisco: Jossey-Bass, pp. 222-236.

Burns, D.D. (1990). The feeling good handbook. New York: Plume.

Bugental, J.F. (1987). The art of the psychotherapist. New York: Norton.

Dryden, W. (1990). Self-disclosure in rational emotive therapy. In G. Stricker & M. Fisher (Eds.), Self-disclosure in the therapeutic relationship. New York: Plenum, pp. 61-74

Ellis, Albert (1977). How to master your fear of flying. New York: Institute for Rational-Emotive Therapy.

Epstein, R.S. & Simon, R.I. (1990). The exploitation index: an early warning indicator of boundary violations in psychotherapy. Bulletin of the Menninger Clinic, 54, 450-465.

Epstein, R.S., Simon, R.I. & Kay, G.G. (1992). Assessing boundary violations in psychotherapy: survey results with the exploitation index. Bulletin of the Menninger Clinic, 56, 150-166.

Fagen, J. & Shepherd, I.L. (1970). Gestalt therapy now. New York: Harper and Row.

Fenichel, O. (1945). The psychoanalytic theory of neurosis. New York: Norton.

Finney, J. (1975). Therapist and patient after hours. American Journal of Psychotherapy, 29, 593-602.

Fisher, M. (1990). The shared experience and self-disclosure. In G. Stricker & M. Fisher (Eds.), Self-disclosure in the therapeutic relationship. New York: Plenum, pp. 3-15.

Folman, R.Z. (1991). Therapist-patient sex: attraction and boundary problems. Psychotherapy, 28, 168-173.

Gabbard, G (1994). Teetering on the precipice: a commentary on Lazarus’s “how certain boundaries and ethics diminish therapeutic effectiveness. Ethics and Behavior, 4, 283-286.

Gabbard, G. & Pope, K. (1988). Sexual intimacies after termination: clinical, ethical, and legal aspects. Independent Practitioner, 8, 21-26.

Gechtman, L. (1989) Sexual contact between social workers and their clients. In G. Gabbard (Ed.) Sexual exploitation in professional relationships. Washington, D.C.: American Psychiatric Press, pp. 27-38.

Goisman, R.M. & Gutheil, T.G. (1992). Risk management in the practice of behavior therapy: boundaries and behavior. American Journal of Psychotherapy, 46, 533-543.

Gottleib, M.C. (1993). Avoiding exploitative dual relationships: a decision-making model. Psychotherapy, 30, 41-48.

Gottleib, M.C. (1994). Ethical decision making, boundaries, and treatment effectiveness: a reprise. Ethics and Behavior, 4, 287-293

Greenson, R. (1967). The technique and practice of psychoanalysis. New York: International Universities Press.

Gutheil, T.G. (1989).Patient-therapist sexual relations. The California Therapist, November/December, pp. 29-39.

Gutheil, T.G. (1994). Discussion of Lazarus’s: how certain boundaries and ethics diminish therapeutic effectiveness. Ethics and Behavior, 4, 295-298.

Gutheil, T.G. & Gabbard, G.O. (1993). The concept of boundaries in clinical practice: theoretical and risk-management dimensions. American Journal of Psychiatry, 1993, 200(2), 188-196.

Johnston, S.H. & Farber, B.A. The maintenance of boundaries in psychotherapeutic practice. Psychotherapy, 33, 391-402.

Jourard, S.M. (1971). The transparent self. New York: D. Van Nostrand.

Katherine, A. (1993). Boundaries: where you end and I begin. New York: Fireside/Parkside.

Korchin, S. (1976). Modern clinical psychology. New York: Basic Books.

Kron, T. & Friedman, M. (1994). Problems of confirmation in psychotherapy. Journal of Humanistic Psychology, 34, 66-83.

Langs, R.J. (1976). The therapeutic relationship and deviations in technique. In R.J. Langs (Ed.) International Journal of Psychoanalytic Psychotherapist: Vol. 4, New York: Jason Aronson, pp. 106-141.

Lazarus, A.A. (1994a). How certain boundaries and ethics diminish therapeutic effectiveness. Ethics and Behavior, 4, 255-261.

Lazarus, A.A. (1994b). The illusion of the therapist’s power and the patient’s fragility: my rejoinder. Ethics and Behavior, 4, 299-306.

Marquis, J.M. (1972). An expedient model for behavior therapy. In A.A. Lazarus (Ed.), Clinical behavior therapy, pp. 41-72.

Notman, M.T. & Nadelson, C.C. (1994). Psychotherapy with patients who have had sexual relations with a previous therapist. Journal of Psychotherapy Practice and Research, 3, 185-193.

Perls, F.S. (1969). Gestalt therapy verbatim. Lafayette, Calif.: Real People Press.

Pope, K.S. & Bouhoutsos, J.C. (1986) Sexual intimacy between therapists and patients. New York: Praeger.

Pope, K.S. (1989). Therapist-patient sex syndrome: a guide for attorneys and subsequent therapists to assessing damage. In G. Gabbard (Ed.) Sexual exploitation in professional relationships. Washington, D.C.: American Psychiatric Press, pp. 39-56.

Pope, K.S. (1990a). Therapist-patient sex as sex abuse: six scientific, professional, and practical dilemmas in addressing victimization and rehabilitation. Professional Psychology: Research and Practice, 21, 227-239.

Pope, K.S. (1990b). Therapist-patient sexual involvement: a review of the research. Clinical Psychology Review, 10, 477-490.

Pope, K.S. (1994). Sexual involvement with therapists. Washington, D.C.: American Psychological Association.

Pope, K.S., Tabachnick, B.G. & Keith-Spiegel, P. (1987). Ethics of practice: the beliefs and behaviors of psychologists as therapists. American Psychologist, 42, 993-1006.

Rogers, C.R. (1951). Client-centered therapy. Boston: Houghton-Mifflin.

Rogers, C.R. (1961). On becoming a person. Boston: Houghton-Mifflin, .

Rogers, C.R. (1970). Carl Rogers on encounter groups. New York: Harper and Row.

Russell, D.E. (1986). The secret trauma: incest in the lives of girls and women. New York: Basic Books.

Shepard, M. (1975). Fritz. New York: Bantam.

Simon, J.C. (1990). Criteria for therapist self-disclosure, In G. Stricker & M. Fisher (Eds.), Self-disclosure in the therapeutic relationship, New York: Plenum, pp. 207-225.

Simon, R.I. (1991). Psychological injury caused by boundary violation precursors to therapist-patient sex. Psychiatric Annals, 21, 614-619.

Simon, R.I. (1992). Treatment boundary violations: clinical, ethical, and legal considerations. Bulletin of the American Academy of Psychiatry and Law, 20, 269-287.

Simon, R.I. (1995). The natural history of therapist sexual misconduct: identification and prevention. Psychiatric Annals, 25, 90-94.

Simon, R.I.(1994). Transference in therapist-patient sex: the illusion of patient improvement an consent, part 1. Psychiatric Annals, 24, 509-515.

Smith, D. & Fitzpatrick, M. (1995). Patient-therapist boundary issues: an integrative review of theory and research. Professional Psychology: Research and Practice, 26, 499-506.

Smith, E.W. (1976). The growing edge of gestalt therapy. New York: Brunner-Mazel.

Sonne, J.L. (1989). An example of group therapy for victims of therapist-client. In G. Gabbard (Ed.) Sexual exploitation in professional relationships. Washington, D.C.: American Psychiatric Press, pp. 101-114.

Sonne, J.L. (1994). Multiple relationships: does the new ethics code answer the right questions? Professional Psychology: Research and Practice, 25, 336-343.

Stone, M.H. (1976). Boundary violations between therapist and patient. Psychiatric Annals, 6, 670-677.

Strasburger, L.H., Jorgenson, L. & Sutherland, P. (1992). The prevention of psychotherapy sexual misconduct: avoiding the slippery slope. American Journal of Psychotherapy, 46, 544-555.

Stricker, G. (1990). Self-disclosure in psychotherapy, In G. Stricker & M. Fisher (Eds.), Self-disclosure in the therapeutic relationship, New York: Plenum, pp. 277-290.

Vinogradov, S. & Yalom, I. (1990). Self-disclosure in group psychotherapy, In G. Stricker & M. Fisher (Eds.), Self-disclosure in the therapeutic relationship, New York: Plenum, pp. 191-204.

Williams, M.H. (1992). Exploitation and inference: mapping the damage from therapist-patient sexual involvement. American Psychologist, 47 (3), 412-421.

Williams, M.H. (1995). How useful are clinical reports concerning the consequences of therapist-patient sexual involvement? American Journal of Psychotherapy, 49 (2), 237-243.

Wolpe, J. & Lazarus, A.A. (1966). Behavior therapy techniques. New York: Pergamon.

Wolpe, J. (1958). Psychotherapy by reciprocal inhibition. Stanford, Calif.: Stanford University Press.

Yalom, I.D. (1980). Existential psychotherapy. New York: Basic Books.

Table 1

Self Reported Rates of Boundary-Related Behaviors as reported by Pope, Tabachnick, and Keith-Spiegel (1987, p. 995-997).

Description Reported Rate of Occurrence
Telling a client you are angry at him or her 42.5% responded from “sometimes” to “very often.”
Using self-disclosure as a therapy technique 69.4% responded from “sometimes” to “very often.”
Having a client address you by your first name 96.2% responded from “rarely” to “very often.”
Hugging a client 41.7% responded from “sometimes” to “very often.”
Kissing a client 23.5% responded “rarely”
Accepting a client’s gift worth at least $50 19.1% responded “rarely”
Accepting a gift worth less than $5 from a client 58.1% responded from “sometimes” to “very often.”
Asking favors (e.g., a ride home) from clients 35.7% responded “rarely”
Lending money to a client 23.9% responded “rarely”
Inviting clients to a party or social event 15.4% responded “rarely” or “sometimes”

Table 2

Self Reported Rates of Boundary-Related Behaviors as reported by Borys and Pope (1989, p. 288).

Description Reported Rate of Occurrence
Disclosed details of current personal stresses to a client 38.9% did this with a few or more clients.
Went out to eat with a client after a session 11.6% did this with a few or more clients
Accepted a client’s invitation to a special occasion 35.1% did this with a few or more clients

To download a pdf version of this article, click here.