jap a

Andrea Celenza / Glen O. Gabbard 51/2

ANALYSTS WHO COMMIT

SEXUAL BOUNDARY
VIOLATIONS: A LOST CAUSE?

The causes of sexual misconduct by analysts are discussed, as is the

viability of rehabilitation for different types of transgressors. Common
misunderstandings about the transgressor (such as the assumption of
psychopathy and the likelihood of multiple offenses) are countered
with a summary of data derived from the evaluation and /or treatment
of over two hundred cases, most of them one-time transgressors. The
typical characteristics of the analyst or therapist who engages in sexual
misconduct are presented and discussed as qualities that are to
some extent present in analysts generally. The temptation to deny
this universal vulnerability is viewed as effectively replicating the kind of
vertical splitting or compartmentalization that makes one vulnerable
to sexual misconduct in the first place.

S exual misconduct is not a problem we can observe from a com-

fortable distance. The analyst who becomes sexually involved
with a patient may be a former teacher, supervisor, colleague, or friend.
Moreover, the persistence of this problem, even among the most highly
educated, respected, and trusted of us, requires a microscopic examina-
tion that may be painful, humbling, and at times threatening. Perhaps
this explains why, until this past decade, the problem has eluded our
attention despite having been identif ied since the dawn of psycho-
analysis.
Any question regarding rehabilitation, whether or not one considers
it viable, presupposes both a general and a specif ic (case-based) under-
standing of the origin of sexual misconduct. We will discuss here both

Andrea Celenza, Faculty, Boston Psychoanalytic Society and Institute; Faculty

and Supervising Analyst, Massachusetts Institute for Psychoanalysis; Assistant Clinical
Professor, Harvard Medical School, The Cambridge Hospital. Glen O. Gabbard, Chair
of Psychiatry, Brown Foundation; Professor of Psychiatry, Baylor College of Medi-
cine; Training and Supervising Analyst, Houston-Galveston Psychoanalytic Institute.
Submitted for publication April 22, 2002.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

the causes of the problem and the prospects for rehabilitation, since
both of us have had substantial experience in assessing and develop-
ing rehabilitation plans for this population. We also address certain
misguided assumptions that are often made about sexual boundary
violations by analysts and therapists. We are summarizing our com-
bined experience evaluating and/or treating over two hundred cases of
physicians, mental health professionals (including analysts), and clergy
who have committed sexual boundary violations. It is our impression
that a carefully monitored rehabilitation plan can be successful if certain
factors are present at the time of the initial evaluation. However, we
recognize that such efforts are futile where the potential for rehabili-
tation does not exist.
Sexual misconduct is commonly misunderstood. This misunder-
standing is particularly apt to surface when a shocking or confusing
case comes to light. The misunderstanding often involves three
assumptions: (1) therapists and analysts who have engaged in sexual
misconduct are probably psychopathic; (2) they have likely exploited
more than one patient; and (3) they are not amenable to rehabilitation.
618
While these three assumptions are indeed valid for a specif ic sub-
group that accounts for a substantial number of victims, this prof ile
does not f it the majority of transgressors.
In our combined experience, most cases do not f it the psycho-
pathic profile, either at the descriptive level or on the basis of deeper,
psychoanalytic understanding. Additionally, a review of the literature
on this topic reveals that all of those who systematically evaluate
signif icant numbers of transgressors come to the same conclusion.
Schoener et al. (1989) have found, as we have, that a myriad of com-
plex factors go into sexual misconduct and that psychopathic predators
constitute a minority of transgressors. One of us (AC) has treated,
supervised, evaluated, and/or consulted with 48 therapists or analysts
who have engaged in sexual misconduct. Of these, only a small sub-
set were multiple transgressors (n=12; 25%), while the remaining
36 (75%) were one-time transgressors who had been involved in an
ostensibly romantic long-term relationship. While we are not sug-
gesting that these relationships are based on mutual or healthy modes
of relating (despite frequent reports by the patient-victim of having
consented to the relationship at the time), we are asserting that most
therapists or analysts are not predatory. The other coauthor (GG) has
evaluated, treated, or consulted with over 150 therapists, analysts, and

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

pastoral counselors who had engaged in sexual relations with a

patient. Fewer than 25 percent were psychopathic or severely narcis-
sistic predators. Moreover, not all of those involved with more than
one patient could be characterized as psychopathic.
The objection can be raised that those sent for evaluation, treat-
ment, or supervision are a skewed sample. Many psychopathic predators
do not admit to the misconduct (even when multiple complaints are
lodged) and refuse evaluation and/or treatment. Our data are consistent,
however, with the f indings of Schoener et al., whose experience over
the last twenty years (with mental health professionals of all disci-
plines) often involved clinicians against whom one or more complaints
were lodged and for whom rehabilitation was not viable.1 They report
that more than half of their cases involve single-victim transgressors
who present with genuine remorse and are recommended for rehabili-
tation (Schoener, personal communication, August 18, 1991). Gartrell
et al. (1987) found in their prevalence study that single-victim trans-
gressors accounted for 66 percent of the transgressors they identified.
These f indings have been reported in the literature over the last
619
twenty years in a wide variety of journals, including psychoanalytic
ones (Gabbard 1995; Celenza 1998; Gabbard and Peltz 2001). It is our
impression, however, that much of the psychoanalytic community
ignores such data. Misunderstanding appears to be especially wide-
spread when institutes or overseeing professional organizations are
required to adjudicate a prominent member of their group. It appears
to us that when members of an institute are confronted with a shameful
and shocking case, previously grasped knowledge about the nature of
sexual misconduct seems to vanish and the stereotype of the psycho-
pathic predator becomes the only explanation available.
This reaction invites the question of why these data have not been
integrated into our general understanding in an enduring way. We
believe that some of the problem lies in the dramatic presentation of the
psychopathic predator. He (such transgressors are almost always male)
captures our attention by committing perverse, egregious acts and
by seducing a large number of victims. Such cases are also highly publi-
cized in the media, by virtue of the lawsuits that invariably follow.
1
An often cited prevalence study (Holroyd and Brodsky 1977) has unfortunately
contributed to confusion regarding the prevalence of one-time versus multiple
transgressors by failing to distinguish between multiple contacts with the same
patient versus multiple offenses with different patients. That study also made use of
the term recidivism in an idiosyncratic way to denote multiple offenses.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

Perhaps of greater signif icance is the sense we often share that

these transgressors are different from you and me, at least on the sur-
face. Because their overt grandiosity and psychopathy inspire a defensive
disowning, an “us versus them” attitude, their behavior is with relative
ease dismissed as a threat to our professional identity. These boundary
violations are intentional and premeditated, and appear to cause no
pangs of conscience in the transgressor. It is diff icult for us to conceive
of engaging in such blatant self-deception. Our resorting then to the
psychopathic prof ile to explain all cases of sexual misconduct may be
seen to involve at least three problematic tendencies. The f irst is the
temptation to master our confusion by simplifying a complex problem.
Second, the maneuver serves our need to have idealized f igures in our
midst by dismissing the analyst in question as not only fallible but
fatally f lawed (with the implication that our idealized role models
remain pure and untainted by comparison). Finally, and most problem-
atic, the psychopathic prof ile denies our own fear of vulnerability by
sharply demarcating the type of character who engages in transgressive
behavior. We are thus reassured that this cannot happen to us.
620
In contrast, the prof ile of the one-time transgressor is, at least on
the surface, too much like you and me. While most of us might still
have diff iculty imagining ourselves engaging in frank sexual behavior
with a patient, it is much easier to resonate with the particulars of this
type of case; we too have found ourselves in a transference-counter-
transference dilemma, and have entertained the fantasy of extreme mea-
sures that might solve an acute crisis. There are typical precipitating
factors in this subgroup, most of which we can understand. We can all
recall times when we crossed minor boundaries in a similar circum-
stance with a similar patient. Examples of such crossings come easily
to mind but are usually shared only in private, informal contexts with
colleagues and friends.
Another reason this prof ile is not securely f ixed in our minds is that
the analysts and therapists in this group can be cooperative and gen-
uinely remorseful about their transgression. In addition, the analyst or
therapist himself often has trouble advocating on his own behalf and
defending himself effectively.2 The adjudication of these cases therefore
is often “silent,” absent media attention or secondary complications.
2
Masculine pronouns are used because most of the transgressors referred to us are
male. However, we will briefly address particular characteristics of women analysts
who have engaged in sexual misconduct.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

Even with those who do not initially present with remorse and are
uncooperative with the assessment process (at least in the beginning),
we have found that much of what determines the decision to rule out
rehabilitation is related to the particular time in which the assessment
or adjudication occurs and the way in which the transgressor presents
himself. Those who remain attached to the patient, who still view the
relationship as based on a special type of love, may feel an untoward
(and, from an outsider’s perspective, incomprehensible) reaction of
betrayal and victimization by the patient-complainant. This reaction is
particularly common when the transgressor had terminated the thera-
peutic relationship and so, in his mind, had adequately dealt with the
boundary transgression and power imbalance. The transgressor may
also feel attacked or threatened by the ethics committee or overseeing
professional organization and may continue to rationalize or defend his
behavior in response to the perceived attack. He may temporarily
minimize the damage to the patient-victim, thereby presenting himself
in a particularly unsympathetic way.
It is understandable for anyone being sued to experience anxiety
621
over the prospect of an investigation. This reaction may heighten
defensiveness and cause attempts at self-defense to include angry
counterattacks. Charges of ethical violations, founded in truth or not,
engender extreme anxiety, defensiveness, and even rage, as one’s
livelihood and professional reputation are threatened. Under such
stress, people often behave in extreme ways that may not be typical
of them and that present them in an unfavorable light. While people
have characteristic ways of reacting to stress, we have found that
these reactions taken alone are incomplete and misleading indicators
of the professional’s competence and underlying character structure.
Extreme reactions—bona f ide posttraumatic stress reactions occa-
sionally accompanied by suicidal ideation—are not at all unusual, but
are usually short-lived if the assessment process can be experienced
as an attempt to understand the complexity of the process (while
still holding the transgressor fully responsible). Some uncooperative
transgressors may eventually become cooperative and amenable to
rehabilitation, but the personal acceptance and ownership of what they
have done may take some time. Still, it remains true that rehabili-
tation is not an option until this acceptance occurs. For some trans-
gressors, the acceptance of full responsibility for the misconduct
never comes to pass.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

We believe that the role of the patient-victim, especially in the

case of the one-time (i.e., single-victim) transgressor, warrants efforts
both to understand the misconduct as a transference-countertrans-
ference enactment and to revise (as necessary) the conventional wisdom
regarding sexual misconduct. Note that our perspective places no
responsibility for the misconduct on the victim’s shoulders. It does,
however, attempt to understand the obstacles to integrating the com-
plexity of the problem within our professional psyches and in the
general public as well (see Gutheil and Gabbard 1992). To put matters
simply, the profile of the psychopathic predator coincides with sex-role
stereotypes, avoids a politically incorrect “blaming the victim,” and
perhaps, more subtly, appeases the enraged and potentially litigious
patient-victim in the case at hand. However, this stereotyped scenario
obfuscates our efforts to understand fully the role of both parties in co-
constructing the transference-countertransference enactment.
Finally, the problem of the one-time transgressor involves subtle
and covert defensive processes, along with denied or disavowed
impulses. The dynamics are not easily explained, especially in cases
622
of respected and trusted leaders of psychoanalytic organizations.
Many one-time transgressors lack the charm and charisma of the preda-
tor. Transference-countertransference scenarios in these cases are
neither immediately accessible nor easily formulated; nor are they easily
held in mind, especially when we consider that the analyst has been
previously analyzed, may be a training analyst, or has even held posi-
tions on ethics committees. The defensive splitting or compartmental-
ization that must be posited here is of a type we usually attribute to
more primitive personality organizations.

COMMON CHARACTERISTICS
OF SINGLE-VICTIM CASES

The following is a presentation of the characteristics typical of the

analyst or therapist who engages in sexual misconduct with only one
patient. In order to ref lect the complexity and diversity of such cases,
we f ind it helpful to conceptualize them as occupying a spectrum, with
lovesickness at one end, masochistic surrender at the other, and many
positions between these extremes showing elements of both (Gabbard
and Lester 1995). We also present common precipitating scenarios
that often provide the context within which the misconduct occurs.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

Some of these aspects are situational, either in the life of the analyst
or therapist or in the therapeutic context. Others are deep-seated
personality characteristics or defensive styles that appear to represent
long-standing vulnerabilities in the character of the transgressor. Still
other elements are best described as intersubjective, as arising from
within the codetermined transference-countertransference enactment
of the treatment dyad.
The psychopathology of these analysts is variable. Most have
narcissistic and masochistic issues, but these may not be apparent.
Depression or despair may be overt or just beneath the surface. Omnipo-
tence and grandiosity may also be at work. However, we are reluctant
to pathologize this spectrum of analysts. Who among us can claim to be
entirely free of similar struggles? Indeed, one could argue that a healthy
dose of grandiosity is necessary to treat our most disturbed patients.
From the point of view of the analyst or therapist, the sexual
involvement with the patient represents a one-time occurrence. By this
we mean that though the relationship may continue over time (from
weeks to years), there is only one patient with whom he has engaged
623
in sexual misconduct. Boundary crossings with other patients may be
reported, but these are not generally beyond the type or level that are
committed at one time or another by all of us. It is common for a trans-
gressor’s other analysands to report acceptable analytic work being
done concurrently with his ongoing sexual relationship with that one
patient. In other words, the misconduct, though an extreme ethical
violation, can occur in an otherwise ethically sound and competent
practitioner.
For the lovesick group, the sexual relationship is experienced by
both participants, at least for a time, as a true love affair. They usually
share a fantasy that each is rescuing the other and that they are soul-
mates who understand each other’s needs like no one else. Rationaliza-
tions are employed, especially by the analyst or therapist, to support the
therapeutic nature of the relationship (Celenza 1995). Role reversals
occur in which the analyst or therapist discloses personal problems to
the patient, often justifying this by citing examples of famous analyts
(e.g., Ferenczi, Winnicott, or Jung) who had similarly transgressed.
Finally, the analyst or therapist is typically in the midst of a life crisis
such as divorce, a lifeless marriage, illness or death in the family, the
suicide of a patient, malpractice litigation, or bankruptcy (Gabbard
and Lester 1995).

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

Intrapsychic factors vary, but common themes have been observed

(Twemlow and Gabbard 1989; Gabbard and Lester 1995; Celenza and
Hilsenroth 1997; Celenza 1998). A prominent intrapsychic factor in
many cases is unconscious guilt in a male analyst or therapist, usually
coinciding with a childhood history of having felt responsible for his
mother’s unhappiness. Needs for mirroring, aff irmation, and recog-
nition may have been neglected early in life. The analyst or therapist
thus presents as a narcissistically vulnerable person who looks pri-
marily or even exclusively to professional activities and relationships
for sustenance and aff irmation of self-worth.
Many clinicians on this spectrum display a compensatory grandi-
ose defensive structure that functions in a covert, disavowed manner
(such as identif ications with a martyr-like “Jesus” imago or heroic
and idealized self-representation). The therapist may enact this struc-
ture in a self-depriving and self-destructive rescue fantasy. The sexual
misconduct represents one such enactment and often stems from con-
f licts around omnipotence, grandiosity, or neediness established as
long-standing characterological features. Superego lacunae may co-
624
incide with this state of mind (a structural mechanism organized by a
vertical split with separate id, ego, and superego features) whereby
certain behaviors that are usually prohibited are permitted in the con-
text of the misconduct. This contributes to a “slippery” type of morality
different from that of the psychopathic character, whose immorality is
based on a more stable and ego-syntonic absence of conscience.
A common theme in this group is a great intolerance of aggres-
sion, an intolerance stemming from childhood family situations in
which conf lict and aggression were not acknowledged, tolerated, or
expressed. This diff iculty in tolerating anger and destructiveness may
occur within the context of a highly moralistic (or even frankly reli-
gious) family with a high level of constrained or unacknowledged
conf lict. While many pastoral counselors describe this background, we
have found that similar childhood histories are common in analysts
and therapists as well.
On an interpersonal level, it is important to recognize that for this
type of analyst or therapist the treatment situation is experienced as
a replication of the childhood trauma (in both preoedipal and oedipal
conf igurations) in the sense that it is at once overstimulating, depriv-
ing, and forbidden. Often the analyst experiences the treatment of the
patient with whom he eventually transgresses as especially challenging

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

and frustrating, such that the patient takes on the persona of a “bad”
maternal object.
The transgression with the patient can be understood as an attempt
to manage and dominate (through sexualization and triumph) the
patient’s aggression, as well as the analyst’s own. On a conscious or
unconscious level, the transgression can function as a masochistic sur-
render to the patient’s hostility, experienced by the analyst or therapist
as intolerable and unbearable because of the way it replicates (for him)
the despair experienced within the mother-child dyad. The sexual mis-
conduct, then, establishes a sadomasochistic relationship that includes
a projection of the analyst’s aggression into the patient, rendering the
analyst masochistically tied to the patient’s cruelty and control. This
description of masochistic surrender is not a sanitized way of explain-
ing or excusing the transgressor’s or therapist’s behavior. It is an attempt
to formulate the defensive transformations of his aggression, which
have resulted in a pathological compromise formation, perhaps of
the kind to which we may all be vulnerable.
By examining common transference-countertransference dynam-
625
ics, we are in no way attributing any ethical responsibility to the patient
for their actualization. It is our experience, however, that many of the
victims in these cases are actively suicidal at the time of the mis-
conduct, lending an urgency to the treatment. Thus, sexual misconduct
can often be understood as an egregious mismanagement of suicidality
(Eyman and Gabbard 1991; Gabbard and Lester 1995).
Many patients who are victims of sexual misconduct have histories
of severe childhood trauma. For some of these patients, the enactment
with the analyst is experienced as a loving and protective act, a sacri-
f icing of herself to meet the analyst’s needs. Just as the patient may
have rescued an abusive parent in childhood, so she may seek to rescue
the analyst from his depression and despair (Gabbard and Lester 1995;
Apfel and Simon 1985). We have found that such a rescue fantasy in the
patient often defends against awareness of her own competitiveness,
envy, and destructiveness, all of which, of course, remain unexamined
in the context of the misconduct.
Many patient-victims also fall into the diagnostic spectrum of
Cluster B personality disorders or dissociative disorders. As we all
know, treatments of such patients are among the most challenging for
any analyst or therapist, and phases of suicidality are often part of the
process. The prospect of the patient’s suicide may arouse narcissistic

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

mortif ication and separation anxiety in the analyst, who feels he must
do anything in his power to avoid losing the patient. The patient’s deep
sense of abandonment and intense ambivalence about closeness may
also stir up countertransference hate in the analyst or therapist (see
Maltsberger and Buie 1974).
Most of us abhor being cast in the role of the bad object and deny
the hatred engendered in response to the patient’s transference hate. In
such a situation, we may choose to provide a defensive form of love
that directs all of the aggression in the dyad to an outside party, usu-
ally the malevolent father or mother. This “disidentif ication with the
aggressor” (Gabbard 1997) maintains our professional identity as a help-
ful and benevolent professional rather than a loathsome practitioner
who is hostile toward or overwhelmed by a dif f icult patient.
Overreliance on reaction formations of this nature is a problematic
defensive style that we regard as a potential precursor of therapist-
patient sexual misconduct (Celenza 1995, 1998).
The threat of suicide, of course, is the ultimate injury to the thera-
pist’s sense of professional mastery and self-worth. Patient and analyst
626
may collude in the magical hope that the therapist has a capacity to
omnipotently heal the patient’s lifelong sense of badness and self-
loathing. Meanwhile, the patient may secretly sense the therapist’s
defensive posture and recognize that enormous feelings of hatred lie
just beneath the surface. The therapist’s despair and rage at the patient’s
tendency to thwart all therapeutic efforts may lead to an escalation of
boundary transgressions under the impression that only extraordinary
measures can reach the patient. Sexualization may transform such
despair into something exciting and positive (Coen 1992; Gabbard 1996).
Ultimately, the childhood trauma is repeated through sexual boundary
violations under the guise of “loving” the patient back to health.
In addition, the analyst or therapist may be prone to feeling in-
tensely degraded, as indicated by the warding off of his own hostility
and its projection into the patient. The analyst or therapist may engage
in a hypomanically charged fantasy of being special to the patient, a
fantasy that is fueled in part by the analyst’s or therapist’s reaction-
formed hatred of the patient. Often in the course of a rehabilitative
treatment process the analyst or therapist comes to recognize his esca-
lating sexualization as an attempt to avoid the countertransference
hate generated by the patient’s undermining efforts, depressive despair,
and frank suicidal threats (Celenza 1991, 1998).

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

ASSESSING THE POTENTIAL

FOR REHABILITATION

Any discussion of rehabilitation potential must begin with the acknowl-

edgment that many analysts and therapists who have committed sexual
misconduct are simply not amenable to rehabilitation. They may have
serious character pathology, complete lack of remorse, and no moti-
vation to examine their role in what has occurred. Others may deny the
charges. Assessment under such circumstances is futile since the trans-
gressor refuses to acknowledge a problem for which a rehabilitation
plan is a solution (Schoener et al. 1989; Gabbard and Lester 1995).
However, many analysts and therapists are amenable to rehabilitation,
and the presumption that all such professionals are beyond redemption
is unwarranted.
Independent assessments have become standard practice in the area
of sexual misconduct cases (Schoener et al. 1989; Schoener 1995;
Gabbard and Lester 1995). Because of the intense reactions of the
psychoanalytic community to a transgressing analyst, the capacity for a
627
local colleague to make such an assessment with any degree of objec-
tivity is seriously in question; for that reason, independent comprehen-
sive evaluations are usually provided by experienced colleagues in
other locations. These evaluations are sometimes requested by a licens-
ing board, sometimes by an advocacy organization (such as a physi-
cians’ health organization associated with a state medical society), and
at other times by the ethics committee of a psychoanalytic society or
institute. Transgressors themselves occasionally request such an evalu-
ation when they realize they have made an egregious error in judgment
and want to ensure that boundary violations do not occur in the future.
Analysts who perform these assessments should have demonstrable
expertise in the area of boundary violations, because evaluating the
potential for rehabilitation is complicated and diff icult.
When an analyst is referred for evaluation, all parties must under-
stand that rehabilitation is clearly distinguished from any disciplinary
measures or sanctions from an ethics committee, licensing board, or
professional organization. In other words, the possibility of rehabili-
tation is not to be regarded as a way to evade punishment or substitute
for it. The analyst performing the evaluation should have no personal
or professional ties to the transgressor and should have no stake in the
internal concerns of the referring organization or institute. The analyst

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

who is being evaluated must sign a release of information at the begin-

ning of the assessment so that the evaluator can communicate the
f indings of the evaluation to the referring body. When transgressing
analysts themselves request the evaluation, of course, no report is made
to anyone except the analyst requesting the consultation.
Before an evaluation begins, most evaluators will insist on having
collateral information about the boundary violation. The boundary-
violating analyst may have a perspective widely divergent from that of
the victim, and both perspectives are essential to reach conclusions
in these Rashomon-like situations. In addition, investigative reports by
licensing boards, written statements by family members of the patient,
and perspectives supplied by the transgressing analyst’s colleagues are
all valuable. Some assessors will insist also on evaluating the spouse
of the transgressing analyst, as marital diff iculties are frequently part
of the overall context. Collateral interviews may be conducted in per-
son or by telephone, but letters or written reports are discouraged so
that subtle implications and nuances in tone may be followed up.
At some centers the evaluation is done by a team that includes
628
an analyst, a psychologist responsible for testing, and various con-
sultants in specif ic areas of expertise, such as substance abuse, affec-
tive disorder, or complicated marital dynamics. Some evaluators work
independently and collect all the information on their own. Most eval-
uations take several hours over two or three days, although some
require more extended periods. In the course of the evaluation, a detailed
account of the misconduct is constructed from the transgressing ana-
lyst’s point of view, with an emphasis on the analyst’s current under-
standing of what happened and why. Discussion of the analyst’s
understanding of why it is unethical to engage in sex with a patient is
an essential aspect of the evaluation, though it frequently happens that
the transgressor recites a sound but intellectualized rendition of the
ethical code. In addition, the analyst’s personal life circumstances,
recent stressors, and diff iculties in intimate relationships with spouses
or partners are essential information.
Despite disparities among various accounts, evaluators must fre-
quently remind both the transgressing analyst and the referring body
that they are not a court of law and cannot determine the actual facts of
what happened. Thus, the purpose of the evaluation is not to “f ind the
facts” but to assess psychopathology in the analyst and the psycho-
dynamics that are relevant to the misconduct.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

Many analysts under evaluation feel they have been traumatized by

the process of adjudication in their local setting and may plead their
case to the evaluator that they are more “sinned against than sinning.”
It is important for evaluators to empathize with what the analyst has
been through and to make it clear that they are genuinely interested
in the transgressing analyst’s point of view. Analysts who conduct the
independent assessment must often struggle with countertransference
wishes to rescue the transgressing analyst; contempt toward the analyst;
anger at the mismanagement of the misconduct by the local institute or
society; collusion with the transgressing analyst against the patient,
who may be seen as “seductive”; and disbelief when the details are
recounted (Gabbard and Lester 1995). In such cases, the evaluator may
wish to seek consultation with an experienced colleague who is knowl-
edgeable about boundary violations.
The assessment of rehabilitation potential hinges more on the atti-
tude of the transgressor than on objective facts provided to the assessor.
An essential characteristic that argues for a potential for rehabilitation
is the presence of genuine remorse. Does the transgressor take full
629
responsibility for the misconduct and demonstrate that he or she
profoundly regrets what happened? Can he or she empathize with the
damage inf licted on the patient and on the profession? Analysts who
are remorseful show the capacity to evaluate their behavior and often
recoil in horror at the rationalizations they employed, the compartmen-
talization of their behavior that they used, and their contextually shift-
ing values. While some analysts may not have reached this point by the
time of the evaluation, they are often progressing in that direction, and
this may become apparent to the evaluator.
Remorse must be rigorously differentiated from narcissistic mor-
tif ication. The latter refers only to regret for the ways in which the
misconduct has damaged one’s sense of self-regard, self-worth, and
reputation. Narcissistic mortification is present to some degree in most
transgressing analysts, but an absence of appreciation for the damage
brought to the patient and to the profession is a disturbing sign.
Similarly, the assessment of whether remorse is genuine or feigned is
crucial. Often the way the transgressor speaks about his responsibility
may be more useful than the actual narrative of the misconduct.
Evaluators look for an attitude of curiosity about his behavior, a sense
of shock at the way his own values have been betrayed, and a desire
to explain the behavior without disclaiming responsibility.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

Other signs of rehabilitative potential may be found in the way the

transgressor behaved during the enactment itself. Evidence of restraint
such as terminating the therapeutic relationship, stopping the acting
out at some point, or referring oneself for evaluation or treatment may
indicate such potential.
Finally, the assessment process entails some efforts to enlist the
transgressing analyst in a collaborative exploration that has a thera-
peutic component. Trial interpretations of the misconduct may be of-
fered to discern whether the transgressing analyst is ready and willing to
ref lect on meanings relevant to understanding what happened. In other
words, a process is begun in which the analyst who is undergoing evalu-
ation is pointed in the direction of intensive therapy or further analysis.
As a general principle, transgressing analysts who are on the con-
tinuum from lovesickness to masochistic surrender are much more likely
to be amenable to rehabilitation than those who are predators.
Nevertheless, the categories sometimes overlap, and even those who
may be passionately in love with the patient may have signif icant
superego pathology that renders them unsafe to practice. Hence the fact
630
that only one victim was involved does not automatically imply that the
analyst is amenable to rehabilitation. Moreover, lovesick analysts who
insist that there is nothing fundamentally unethical in their misconduct
and that they are simply involved in “true love” lack the ref lective
capacity to undergo a rehabilitation process. At a later point the situa-
tion may change.
Our major focus has been on analysts who are men, largely because
they constitute the vast majority of the boundary violators referred
to us. Female analysts and therapists, though, are responsible for about
15 to 20 percent of the cases we see. Some of the dynamics specif ic
to this gender have been elaborated elsewhere (Gabbard and Lester
1995), but much of what we have described about the continuum
from lovesickness to masochistic surrender applies equally to female
clinicians. They are often intensely involved in a rescue ef fort with
a particularly diff icult patient, and their overidentif ication with the
patient’s suffering leads them to heroic and misguided treatment efforts.
In our experience, the patient with whom the female therapist becomes
involved is as likely to be female as male. Most of the female boundary
violators we have seen are involved with only one patient. Because
predatory psychopathy is rare in women, the potential for rehabilitation
among female transgressors is particularly promising.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

THE REHABILITATION PROCESS

Does rehabilitation work? We have seen numerous examples in our

experience that convince us that in many cases it can be effective.
However, we accept that no data will ever satisfy the skeptic, because
follow-up information is limited to independent and measurable events
such as the subsequent f iling of complaints or the revocation of one’s
professional license. An analyst’s self-report of no subsequent trans-
gressions and meticulous attention to the maintenance of professional
boundaries is often entirely convincing to the rehabilitation team, but
would not satisfy a skeptical scientist because of the possibility that
the transgressor is concealing subsequent violations. Self-report, in
any case, cannot be regarded as the sole outcome measure in a well-
designed empirical study. Still, it is our experience that rehabilitation
processes can result in an expansion of the transgressor’s insight, affect
tolerance, and self-control such that he can be regarded as a trusted
and valued colleague and practitioner. Further, due to the intensive self-
scrutiny involved in the rehabilitation process, some former trans- 631
gressors may end up with a keener appreciation of the subtleties of
boundary management than has the average practitioner who has never
transgressed.
Rehabilitation plans are individually tailored to the specif ic situa-
tion of the analyst being evaluated. Typical components of such a plan
include the following (Gabbard and Lester 1995).
Assignment of a rehabilitation coordinator. A rehabilitation coordi-
nator is assigned to monitor the overall plan and make any necessary
reports to licensing bodies, professional health organizations, or other
monitoring bodies. This position is usually held by an analytic colleague
knowledgeable about boundary violations and agreeable to interviewing
the transgressing analyst at regular intervals. The position must not be
filled by a friend of the analyst undergoing rehabilitation. The coordinator
may receive reports from supervisors. Assigning the reporting function
to the coordinator allows for (1) a synthesis of potentially disparate
points of view (in the case of multiple supervisors as well as mentors,
educators, or other persons involved in the rehabilitation) and (2) in
the case of the treating analyst, a more narrow breach of confidentiality
as compared to reporting to the overseeing professional agency directly.
Individual psychotherapy or psychoanalysis. The transgressing
analyst must return to treatment to explore the factors that led to the

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

boundary violation and to discuss any diff icult countertransference

situations that arise as he or she returns to practice. The reports of
the treating analyst go to the coordinator and are generally limited to
one-sentence statements of whether the patient is attending the treat-
ment or whether a premature termination has occurred. Some licens-
ing boards try to encroach on this conf identiality, but the treatment
is irreparably compromised if the analyst undergoing rehabilita-
tion knows that frank disclosures of erotic countertransference
will be reported to a monitoring agency. In the ideal rehabilitation
plan, the treating analyst only reports attendance to the coordinator.
An important component of this treatment usually is the thorough
exploration of the transgressor’s negative transference to the treating
analyst, especially as it involves aggression, competitiveness, and strug-
gles over power and authority. To aid in the identif ication of problem-
atic areas, the treating analyst or therapist is encouraged to request
and read a copy of the evaluation before beginning the treatment.
However, there may be an explicit understanding that the report, to
guard against intellectualizing the treatment process, will not be shared
632
with the transgressor.
Supervision. In most rehabilitation plans, one or more supervisors
may be assigned to the analyst. The analyst being rehabilitated should
not be allowed to choose these supervisors unaided, lest personal friends
or sympathetic colleagues be chosen. The rehabilitation coordinator or
the oversight body usually selects two or three supervisors with appro-
priate expertise among whom the transgressor may choose.
Each supervisor should be thoroughly familiar with the indepen-
dent assessment and with the circumstances of the transgression. Super-
visors must focus on blind spots in their supervisee and should examine
boundary issues as they arise. An understanding of the misconduct and
the management of boundaries in general must be the principal goals
of the supervision. Usually, special attention is paid to enhancing the
transgressor’s appreciation of power imbalances and ability to identify,
tolerate, and manage countertransference hate and anger.
Practice limitations. Depending on the evaluation, some analysts
may be allowed to work only with certain subgroups of patients. In
extreme cases, for example, a male analyst may be limited to treating
male patients. More commonly, an analyst’s practice might be restricted
in such a way that patients with severe childhood trauma and severe
personality disorders are referred elsewhere. Some analysts thought

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

to be at high risk in an independent practice situation may be limited

to work in an institutional or group practice setting.
Education. Many analysts have very little education about profes-
sional boundary issues. Tutorials or seminars in ethics and boundaries
may be useful as part of the overall strategy. However, it has been our
experience that this aspect of the rehabilitation plan addresses only
conscious intellectual factors that are easily denied.
Mediation. In some cases an evaluator may decide that a mediation
process is indicated as part of the rehabilitation plan. An experienced
analyst knowledgeable about boundary violations meets with the patient
and the transgressing analyst for several sessions to provide an opportu-
nity for the patient to recount how he or she experienced the betrayal
by the analyst. Such feedback to the transgressing analyst may help him
or her empathize with the impact that the transgression has had on the
victim. Mediation also provides an opportunity for the transgressing analyst
to offer an apology to the patient. Finally, mediation often facilitates
the reimbursement of the patient for fees spent on a process that was
seriously misguided and harmful. Such restitution may obviate the need
633
for further litigation. However, the process of mediation should not
be initiated as a substitute for the formal f iling of a complaint by the
patient. Mediation is much more effective after the patient’s concerns
have been heard by an ethics committee or similar body.
The duration of a typical rehabilitation plan ranges from three
to six years (Gabbard and Lester 1995). Before the transgressing
analyst can return to unsupervised practice, the analyst must undergo
a careful reassessment. Sometimes the independent evaluator who
made the original assessment is enlisted to reevaluate the analyst. In
other situations, an independent assessment from a fresh perspective
may be sought. The rehabilitation coordinator’s regular reports to the
ethics committee or oversight body may help determine at which point
return to unsupervised practice is appropriate. Even in cases where
rehabilitation has been successful, analysts who have undergone the
program are advised to continue using supervisors or consultants as
long as they practice.

CAN THIS HAPPEN TO YOU OR ME?

One of the most common responses to news of serious boundary viola-

tions by a colleague is “I can’t imagine how that could ever happen

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

to me.” Yet we are all more similar than different. The characteristics
of the transgressor as outlined above do not have hard-and-fast bound-
aries between them. They often shade into one another. Elements of
each can be found in the others, and we can f ind aspects of each in
most of us. This raises again the persistent question, How different are
these transgressors from you and me?
A useful way to address this question is to note the multilayered
way in which the problem of sexual misconduct currently exists in our
own minds. A common reaction on f irst hearing of misconduct is to
disbelieve the allegations or to see the analyst as victim of a destruc-
tive patient. An extreme reaction that may follow is to demand the
transgressor’s immediate expulsion from the analytic community. This
reaction usually arises from an inability to imagine ever trusting the
transgressor again, an inability or refusal to identify with his or her
circumstances and state of mind, or extreme anger at the damage
done to the reputation of the analytic community.
Once the initial shock has worn of f, however, questions linger
in our minds as we struggle to come to terms with the problem. In
634
response to the precipitating factors and in an attempt to put ourselves
in the analyst’s place, we f ind ourselves saying, “I can resonate with
this dilemma.” Then a countervailing thought emerges: “But I wouldn’t
do that.” Another reaction follows: “I can imagine a fantasy of surren-
dering to . . .” or “With a patient like this, I have crossed . . . .” And
then a counterresponse: “But I can’t imagine feeling that way or doing
that. What is it about him that’s different from me? He does have
those inhibitions or that specif ic vulnerability.” And f inally: “Is that
different from me? Maybe yes, maybe no.”
Such obsessional ruminations are common in all of us. They are
part of an introspective process in which we should all continuously
engage as we monitor the impact our work is having on us. We believe
that the “us versus them” mentality in effect replicates the very com-
partmentalization or vertical splitting that makes one vulnerable to
committing boundary violations. It disavows vulnerability and, perhaps
grandiosely, denies fallibility as well.
Another way to address the problem turns the whole question of
sexual misconduct on its head. Rather than ask why sexual miscon-
duct occurs, we might usefully consider why in most cases it does
not. Love of one sort or another regularly finds its way into the intensely
intimate experience of the psychoanalytic process. Why, then, aren’t

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 ANALYSTS WHO COMMIT SEXUAL BOUNDARY VIOLATIONS

boundary violations more common? What are the safeguards most of

us employ to prevent our crossing boundaries all the time?
The answer is not a simple one. A fundamental belief in the value
of analytic treatment is essential. An altruistic determination to put the
patient’s needs ahead of our own is basic to the analytic process.
However, we know that excessive self-deprivation over long periods
can be a precursor of sexual misconduct. Regular use of consultation
from trusted colleagues early in a process, before transgressions occur,
may help prevent disaster.
We end by addressing the question in our title: Are analysts who
have engaged in sex with their patients a lost cause? We think the
question must be answered on a case-by-case basis. We are convinced,
however, that many transgressors, perhaps even most, are amenable to
rehabilitation efforts and that it is misguided to dismiss every analyst
who has made a serious error in judgment. If these analysts are always
a lost cause, then so are we all. The fundamental issue is this: Do we
believe people can change? If we do not, we are in the wrong business.
635
REFERENCES

APFEL, R.J., & SIMON, B. (1985). Patient-therapist sexual contact: I.

Psychodynamic perspectives on the causes and results. Psychotherapy
and Psychosomatics 43:57–62.
CELENZA, A. (1991). The misuse of countertransference love in cases of
sexual intimacies between therapists and patients. Psychoanalytic
Psychology 8:501–509.
——— (1995). Love and hate in the countertransference. Psychotherapy:
Theory, Research, and Practice 32:301–307.
——— (1998). Precursors to sexual misconduct: Preliminary f indings.
Psychoanalytic Psychology 15:378–395.
——— & HILSENROTH, M. (1997). A Rorschach investigation of sexualized
dual relationships. Bulletin of the Menninger Clinic 61:1–20.
COEN, S. (1992). The Misuse of Persons: Analyzing Pathological Dependency.
Hillsdale, NJ: Analytic Press.
EYMAN, J.R., & GABBARD, G.O. (1991). Will therapist-patient sex prevent
suicide? Psychiatric Annals 21:669–674.
GABBARD, G.O. (1995). The early history of boundary violations. Journal of
the American Psychoanalytic Association 43:1115–1136.
——— (1996). Love and Hate in the Analytic Relationship. Northvale, NJ:
Aronson.

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015

 A n d r e a C e l e n z a / G l e n O. G a b b a r d

——— (1997). Challenges in the analysis of adults who were sexually

abused as children. Canadian Journal of Psychoanalysis 5:1–25.
——— & L ESTER , E. (1995). Boundaries and Boundary Violations in
Psychoanalysis. New York: Basic Books.
——— & PELTZ, M. (2001). Speaking the unspeakable: Institutional reac-
tions to boundary violations by training analysts. Journal of the Ameri-
can Psychoanalytic Association 49:659–673.
GARTRELL, N., HERMAN, J., OLARTE, S., FELDSTEIN, M., & LOCALIO, R. (1987).
Reporting practices of psychiatrists who knew of sexual misconduct by
colleagues. American Journal of Orthopsychiatry 57:287–295.
GUTHEIL, T.G., & GABBARD, G.O. (1992). Obstacles to the dynamic under-
standing of therapist-patient sexual relations. American Journal of
Psychotherapy 56:515–525.
HOLROYD, J., & BRODSKY, A. (1977). Psychologists’ attitudes and practices
regarding erotic and nonerotic physical contact with patients. American
Psychologist 32:843–849.
MALTSBERGER, J.T., & BUIE, D.H. (1974). Countertransference hate in the treat-
ment of suicidal patients. Archives of General Psychiatry 30:625–633.
636 SCHOENER, G.R. (1995). Assessment of professionals who have engaged in
boundary violations. Psychiatric Annals 25:95–99.
——— MILGROM, J.H., GONSIOREK, J.C., LUEPKER, E.T., & CONROE, R.M. (1989).
Psychotherapists’ Sexual Involvement with Clients: Intervention and
Prevention. Minneapolis: Walk-In Counseling Center.
T WEMLOW, S.W., & G ABBARD , G.O. (1989). The lovesick therapist. In
Sexual Exploitation in Professional Relationships, ed. G.O. Gabbard.
Washington, DC: American Psychiatric Press, pp. 71–87.

Andrea Celenza
32 Ingleside Road
Lexington, MA 02420
E-mail: [email protected]

Glen O. Gabbard
Department of Psychiatry
Baylor College of Medicine
One Baylor Plaza
Houston, TX 77030
Fax: 713–798–3138
E-mail: [email protected]

Downloaded from apa.sagepub.com at Freie Universitaet Berlin on May 12, 2015