(1995). Int. J. Psychoanal.

, (76):475-485

Countertransference: The Emerging Common Ground

Glen Gabbard 

In the last decade or so, the understanding of countertransference has become

an emerging area of common ground among psychoanalysts of diverse
theoretical perspectives. This convergence can be traced to the development of
two key concepts—projective identification and countertransference
enactment. Projective identification has evolved from a patient's intrapsychic
fantasy in Klein's original work to an interpersonal interaction between patient
and analyst. The notion of countertransference enactment has been widely
used to capture clinical situations in which a countertransference reaction in
the analyst corresponds to the patient's attempt to actualise a transference
fantasy. These ideas, in conjunction with the contributions of social
constructivists and relational theorists, as well as Sandler's conceptualisation of
role-responsiveness, have led to an understanding of countertransference as a
‘joint creation’ by analyst and patient. The relative contributions of analyst and
patient vary somewhat according to the theoretical perspective espoused by
the analyst. This common ground is best regarded as comprising a gradient or
continuum in which more weight is given to the analyst's contribution on one
end of the continuum and more emphasis to the patient's contribution on the
other. While countertransference enactments are widely regarded as inevitable,
their role in creating intrapsychic change is more controversial.

In his search for common ground, Wallerstein (1990) noted that there are fewer
differences in technique than in theory among the diverse psychoanalytic
traditions. He proposed that systematic attention to the core psychoanalytic
phenomena of transference and resistance shows that they are probably more
similar than different in the clinical setting, regardless of one's theoretical
perspective. In this communication I am proposing that recent psychoanalytic
writings suggest another area of emerging common ground—namely, the
understanding of countertransference. Abend (1989) recently acknowledged
that the notion that an analyst's countertransference can be a crucial source of
information about the patient has now become widely accepted. This has
occurred in parallel with a gradual recognition that contemporary Kleinians and

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
classical ego psychologists have a good deal in common (Richards & Richards,
in press).

This rapprochement between groups of analysts on opposite sides of the

Atlantic has been paralleled by another common theme in contemporary
psychoanalytic writing. Hoffman has noted how many recent contributions
have reported some form of deviation from a traditional or more accepted way
of working, what he has termed a feeling of ‘throwing away the book’ (1994, p.
188). He related this to the spontaneous introduction of the analyst's
subjectivity into the process. From a broader perspective, there is a growing
recognition in all quarters that the analyst is ‘sucked in’ to the patient's world
through an ongoing series of enactments that dislodge the analyst from the
traditional position of the objective blank screen. By examining the interface of
two related concepts, projective identification and countertransference
enactment, we can track the evolution of this new common ground in

(MS. received January 1995)

Copyright © Institute of Psycho-Analysis, London, 1995

475

contemporary psychoanalysis, a region that occupies a portion of the landscape

that we regard as the analyst's psychic reality.

The concept of projective identification

The link between the notion of countertransference and the concept of

projective identification has evolved considerably over the last 50 years.
Although Melanie Klein is correctly credited with coining the term, her writings
on the subject are relatively sparse and marked with ambiguity. The term first
appears in her classic 1946 paper, ‘Notes on some schizoid mechanisms’, and
was actually a relatively peripheral aspect of that communication. Projective
identification was considered as one of several defence mechanisms connected
with the paranoid-schizoid position. In this context the term is used to describe
how the infant expels ‘bad parts’ of the ego into the mother in an effort to
possess and control the mother.

British Kleinian analysts have understood this usage to reflect Klein's

conceptualisation of projective identification as an intrapsychic fantasy (Segal,

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
1964; Spillius, 1992). Some contemporary American contributors to the
literature on projective identification (Ogden, 1979, 1982, 1994; Scharff, 1992)
have called attention to a footnote in Klein's paper in which she stressed that
she prefers to conceptualise the projected contents as going into rather than on
to the mother. This attempt at clarification by Klein may be viewed as possibly
signalling an interpersonal dimension to the process. This perspective is further
bolstered by the usage of projective identification in her subsequent 1955
paper, ‘On identification’. Ogden (1994) pointed out that in the novella, If I Were
You, by Julian Green, which is the centrepiece of Klein's paper, there is a clear
implication that the target of the projection is transformed by the process.

On the other hand, Spillius (1992) did not interpret this usage to imply a change
in the external object as an integral part of projective identification. To the
extent that the analyst was influenced by the patient's behaviour, Klein
understood it to reflect countertransference in the narrow Freudian sense,
implying that the analyst needed further analysis. Spillius (1992) has argued
that Klein was not enthusiastic about the broadening of the concept to include
the analyst's emotional response to the patient's provocative behaviour, as
Heimann (1950) suggested. She was concerned that such a connotation might
allow analysts to blame their patients for their own countertransference
difficulties.

The ambiguity in the usage of projective identification by the Kleinians is also

reflected in Segal's (1964) somewhat contradictory definition of the term.
Although she generally regarded it as an intrapsychic fantasy, she also had
acknowledged that the person targeted by the projection may identify with that
which has been projected through the process of introjective identification
(Scharff, 1992).

The interpersonal dimension of projective identification was made explicit in

the 1950s by a small group of British analysts, including Bion (1955), Rosenfeld
(1952), and Money-Kyrle (1956). Bion (1957, 1958, 1959, 1962a, b, 1970), in
particular, led the way in redefining projective identification in a manner that is
now common in contemporary psychoanalytic parlance.

Bion linked projective identification with his container-contained model. In

other words, the infant projectively disavows affects and internal states that are
intolerable and thus facilitates their containment by the mother. These feelings
are ‘detoxified’ and metabolised by the mother and reinternalised by the infant,
who is able to experience them more fully through means of identification with
the mother. Bion was explicit in stressing that an interpersonal interaction
occurs above and beyond the projector's unconscious fantasy. He stressed that

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
in the analytic situation, the analyst actually feels coerced by the patient into
playing a role in the patient's fantasy.

Elaborating on Bion's construct, Ogden (1979, 1982) has described three

aspects of projective identification: (1) an aspect of the self is projectively
disavowed by unconsciously placing it in someone else; (2) the projector exerts
interpersonal pressure that coerces the other person to experience or
unconsciously identify with that which has been projected; and (3) the recipient
of the projection (in the

476

analytic situation) processes and contains the projected contents, leading to

reintrojection by the patient in modified form. Ogden (1992) stressed, however,
that these aspects should not be construed in a linear sequence of steps, but
rather should be conceptualised as creating a dialectic in which the patient and
analyst enter into a relationship in which they are simultaneously separate but
also ‘at one’ with each other. A unique subjectivity is created through the
dialectic of interpenetration of subjectivities.

Modern Kleinian analysts have widely accepted that the analyst's

countertransference may reflect the patient's attempt to evoke feelings in the
analyst that the patient cannot tolerate. Such feelings can therefore be
regarded as an important communication. Joseph (1989) observed that patients
often attempt to ‘nudge’ the analyst to act in a manner that corresponds to
what the patient has projected. She suggested that analysts must allow
themselves to respond to such pressures in an attenuated way so that they
become consciously aware of the projected contents and can bring it to bear
constructively through interpretation. Spillius (1992) has noted that, in practice,
the analyst is always influenced to some degree by what the patient is
projecting, a view shared by Ogden (1982). She pointed out, though, that
virtually everyone agrees with Klein's caveat that patients should not be
‘blamed’ for all the feelings experienced by the analyst. In this regard she
emphasised that analysts may well confuse their own feelings with those of the
patient and that ongoing psychological work by the analyst is necessary to
differentiate feelings that originate in the patient from those that originate in
the analyst.

In the light of the broad agreement that projective identification relies on

interpersonal pressure or ‘nudging’, rather than mystical exchange of psychic
content, there is a growing consensus that the process requires a ‘hook’ in the
recipient of the projection to make it stick (Gabbard, 1994a, b, c). In other
words, the pre-existing nature of intrapsychic defences and conflicts, as well as

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
self-object-affect constellations in the internal world of the recipient, will
determine whether or not the projection is a good fit with the recipient. Even
when the countertransference response is experienced by analysts as an alien
force sweeping over them, what is actually happening is that a repressed self-
or object-representation has been activated by the interpersonal pressure of
the patient. Hence, the analyst's usual sense of a familiar, continuous self has
been disrupted by the emergence of these repressed aspects of the self.
Symington (1990) has described this process as one in which the patient ‘bullies’
the analyst into thinking the patient's thoughts rather than the analyst's own
thoughts.

Racker (1968) divided the analyst's reactions into concordant and

complementary countertransferences. The former involves an empathic link
between the analyst and patient; in other words, the analyst identifies with a
self-representation within the patient. Complementary countertransferences
involve the analyst's identification with a projectively disavowed internal object-
representation of the patient, which Racker regarded as an instance in which
the analyst's own conflicts were activated by the patient's projections. Grinberg
(1979) made a further distinction in the analyst's response by using the term
projective counteridentification. He argued that Racker's complementary
countertransference reaction is always a function of the patient's projection
corresponding with certain aspects of the analyst's own unconscious conflicts.
On the other hand, in projective counteridentification he proposed that the
analyst introjects an affective state associated with the patient's object-
representation that comes almost entirely from the patient.

Grinberg's (1979) view would be considered extreme by many contemporary

thinkers. Although many would agree that an introjective identification process
takes place in the analyst, if a ‘good fit’ is not present, feelings and internal
representations projected by the patient may be shaken off as alien by the
analyst (Gabbard, 1994c; Scharff, 1992). Hence, the analyst's valency (Bion,
1959) to respond to a certain projection must be taken into account. Ogden
would appear to agree with this notion:

It is my experience that projective identification is a universal

feature of the externalization of an internal object relationship, i.e. of
transference. What is variable

477

is the degree to which the external object is enlisted as a participant

in the externalization of the internal object relationship (1983, p.236).

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
Some critics (Kernberg, 1987; Porder, 1987; Sandler, 1987) of the broadened
conceptualisation of projective identification have felt that the original Kleinian
notion has been extended too far and distorted in the process. Kernberg (1987)
argued for a narrower definition that includes projecting intolerable aspects of
intrapsychic experience on to the analyst, maintaining empathy with the
projected contents, attempting to control the analyst in the service of defensive
efforts, and unconsciously inducing feelings in the object that correspond to
what has been projected in the here-and-now interaction with the analyst. He
felt that extending it to include the analyst's intrapsychic elaboration of the
projected contents and the return of what has been projected in the form of an
interpretation is unwarranted.

Writing from an ego-psychological perspective, Porder (1987) believed that

projective identification can be alternatively conceptualised as an example of
identification with the aggressor. In his model, the patient unconsciously turns
passive into active by casting the analyst in the role of the bad child, while the
patient takes on the role of a demanding, critical, masochistic, or sadistic
parent. Porder has suggested that rather than something being projected into
the analyst, an affect is induced in the analyst because of the patient's ‘acting in’.

Sandler (1987, 1993) cautioned that it is extremely risky to assume a one-to-one

correspondence between what goes on in the analyst and what is in the
patient's mind. He regarded projective identification as a defensive process
involving two steps: first, there is an intrapsychic projection of a split-off and
unwanted aspect of a self-representation into an object-representation, and
second, the object-representation (revised in fantasy to include the unwanted
aspect of the self) is externalised via an actualisation process in which the
analyst is pushed (through unconscious verbal and nonverbal manoeuvres) to
play a particular role vis-à-vis the patient (Sandler, 1987).

Sandler's (1976) original concept of ‘role responsiveness’ is closely related to the

contemporary view of projective identification. Consider Sandler's observation
in his classic paper:

Very often the irrational response of the analyst, which his

professional conscience leads him to see entirely as a blind spot of his
own, may sometimes be usefully regarded as a compromise-formation
between his own tendencies and his reflexive acceptance of the role
which the patient is forcing on him (1976, p. 46).

Sandler conceptualised the patient as unconsciously actualising in the

transference an internalised object relationship, in which the analyst is playing

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
a role derived from the patient's intrapsychic world. Spillius (1992) has
characterised Sandler's concept of actualisation as a colloquial term for the
same process described by Joseph (1989), in which the patient unconsciously
induces feelings in the analyst and nudges the analyst into acting in concert
with the projection. Sandler (1993) regarded this form of identification with the
fantasised object as more or less the same as Racker's (1968) notion of
complementary countertransference. He distinguished it from a process of
primary identification, an automatic mirroring process that underlies analytic
empathy. Sandler underscored that any intense emotional reaction by the
analyst to the patient's words or behaviour is not projective identification
‘unless it is unconsciously intended to evoke such a reaction in the analyst’
(1993, p.1105). In this regard he sharpened the definition to avoid a tendency in
the literature to ascribe any intense countertransference feeling to a state that
is induced by the patient.

Countertransference enactment

Among classical analysts associated with the ego-psychological perspective, the

concept of enactment has stirred a great deal of interest over the past decade.
However, usage of the term varies, and full agreement on a specific definition
has not been reached (Panel, 1992). Jacobs (1986) was instrumental in
introducing the term as a way of understanding subtle instances of interlocking
transference-countertransference

478

dimensions that operate outside of conscious awareness, often through

nonverbal means, such as body postures. In an elegant review of the term,
McLaughlin (1991) noted the roots of the word in the notion of playing a part or
simulating, and in the notion of persuading or influencing someone else in the
interpersonal field. McLaughlin defined enactment broadly as

all behaviors of both parties in the analytic relationship, even

verbal, in consequence of the intensification of the action intent of our
words created by the constraints and regressive push induced by the
analytic rules and frame (1991, p. 595).

He also offered a more specific definition: ‘Those regressive (defensive)

interactions between the pair experienced by either as a consequence of the
behavior of the other’ (p. 595).

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
When enactments are narrowed further to a focus on countertransference
enactments, the connection to projective identification becomes clear, as in
Chused's (1991) definition: ‘Enactments occur when an attempt to actualize a
transference fantasy elicits a countertransference reaction’ (pp. 629). Boesky
(Panel, 1992) noted the similarities between enactment and projective
identification, and he suggested that detailed study of enactments might allow
for a better understanding of how projective identification works. Chused
(Panel, 1992) stressed that implicit in the notion of projective identification is
that any analyst would respond in approximately the same manner to specific
behaviour or material in the patient. Countertransference enactments, on the
other hand, assume that the intrapsychic meaning of an interaction in the
analysis could be entirely different for different analysts, who might then
behave differently when presented with the same material by the same patient.
McLaughlin (1991; Panel, 1992) suggested that in projective identification the
analyst is viewed as virtually empty and is simply a receptacle or container for
what the patient is projecting.

The distinctions made by Chused and McLaughlin may be more apparent than
real. As noted previously, modern Kleinians such as Spillius (1992) and Joseph
(1989) share the same concern that it would be inappropriate to assume that all
of the analyst's feelings derive from the patient. They would agree with
Chused's perspective that individual variations in the analyst might result in
different countertransference enactments or different variants of projective
identification.

It is true that more classical analysts, when writing about enactments, often
focus to a greater extent on countertransference in the narrow sense, i.e.
experiences from their own past that are revived in the interaction with the
patient (Jacobs, 1986). However, most would agree with the Kleinian notion that
the analyst's countertransference may convey important information about the
patient (Abend, 1989). As Jacobs has noted,

The inner experiences of the analyst often provide a valuable

pathway to understanding the inner experiences of the patient …
(1993a, p. 7).

Similarly, Renik (1993) described a countertransference enactment in which he

felt immobilised and emphasised that the enactment was partly determined by
his own childhood wish to save his mother and partly by his patient's need to
elicit a rescue response in him.

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
Roughton also regarded countertransference enactments and projective
identification as strikingly similar. He made a distinction, however, between an
enactment, which simply involves putting an experience into behaviour, and
actualisation, which he sees as

subtle forms of manipulation on the part of the analysand that

induces the analyst, often unknowingly, to act or to communicate in a
slightly special way or to assume a particular role with the analysand
that silently gratifies a transference wish or, conversely, defends
against such a wish. This interactive aspect might also be called an
enactment which has an actualizing effect (1993, p. 459).

He acknowledged that this view of enactment as actualising a

countertransference response in the analyst is virtually the same as Sandler's
(1976) role-responsiveness and Ogden's (1979) understanding of projective
identification. He noted that the principal difference may be that the common
usage of projective identification is in the context of more primitive patients

479

who are in somewhat regressed states during analytic treatment.

To summarise my major thesis up to this point: the modern usage of projective

identification among those analysts influenced by Klein (and by the British
School of object relations) and the usage of countertransference enactment by
classical or ego-psychological analysts both involve an understanding of the
analyst's countertransference as a joint creation by patient and analyst
(Gabbard, 1994a, b, c). The analysand evokes certain responses in the analyst,
while the analyst's own conflicts and internal self- and object-representations
determine the final shape of the countertransference response.

A consensus is emerging that such countertransference enactments are

inevitable in the course of psychoanalytic treatment. What is less consensual is
the extent to which such enactments are useful to the process (Chused, 1991).
Eagle (1993) presented a case vignette in which a transference-
countertransference enactment in and of itself appeared to cure a symptom.
He invoked the mastery-control theory of Weiss & Sampson (1986) as an
explanatory framework, assuming that the patient disconfirmed a core
unconscious pathogenic belief, which in turn led to symptom remission without
insight.

Chused (1991) has noted the value of enacting certain impulses within the
analytic frame, only to catch oneself and retrospectively examine what

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
happened. She stressed, however, that the value for the analysis is not in the
enactment itself but rather in the observations and eventual understanding
that derive from those enactments. Jacobs (1993b) has taken a middle course in
which he has said that both experience and insight operate together and
cannot truly be separated from one another.

Renik (1993) argued that countertransference awareness always emerges after

countertransference enactment. He shares Boesky's (1990) view that analysis
may not proceed unless the analyst gets emotionally involved in ways that he
or she had not intended. Renik has embraced a technique that allows for
spontaneity of the analyst even though a certain degree of the analyst's
subjectivity inevitably works its way into interventions. In this regard, he has
aligned himself with constructivists such as Hoffman (1983, 1992, 1994) who
recognise the inevitability of bringing subjectivity to bear in understanding the
analytic interaction. The constructivist view also acknowledges that to some
extent the analyst's behaviour is shaped by influences from the patient. Both
transference and countertransference would be regarded as joint creations
within this view.

Central to the constructivist (or socialconstructivist) perspective is the notion

that enactments are going on continuously in the analytic setting, and analysts
must continually monitor themselves for the possibility that they are
unconsciously participating in an internal scenario scripted by the patient (Gill,
1991; Hoffman, 1992). Moreover, the process goes both ways in the sense that
the analyst's actual behaviours influence the patient's transference to the
analyst. Another implication of the constructivist understanding of
transference-countertransference enactments is that the intrapsychic and
interpersonal realms cannot be divorced from one another in the analytic dyad
(Hoffman, 1991), a view also stressed by Coen (1992), who has approached the
issue from a more classical orientation.

Relational theorists, such as Mitchell (1988, 1993), Aron (1991), Hirsch (1993,
1994), and Tansey (1994) have arrived at similar conclusions about the
inevitability and usefulness of countertransference enactments. Mitchell, for
example, in pointing out the similarities between his view and those of Sandler,
Gill, Racker, and Levinson, has made the following observation:

The analyst is regarded as, at least to some degree, embedded

within the analysand's relational matrix. There is no way for the
analyst to avoid his assigned roles and configurations within the
analysand's relational world. The analyst's experience is necessarily
shaped by the analysand's relational structures; he plays assigned

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
roles even if he desperately tries to stand outside the patient's system
and play no role at all (1988, p. 292).

He went on to emphasise that unless the analyst enters into the patient's
relational world, the analytic experience will not be optimal.

480

Intersubjectivity

The demise of the analyst-as-objective-surgeon construct has led to a

widespread recognition that two subjectivities are operating in the analytic
enterprise. Bollas, influenced by thinkers associated with the British School of
object relations, principally Winnicott, also regarded the analytic process as
involving two subjectivities. However, he regarded the countertransference as
offering a unique window into the patient's internal world: ‘In order to find the
patient we must look for him within ourselves. This process inevitably points to
the fact that there are ‘two patients’ within the session and therefore two
complementary sources of free association’ (1987, p. 202).

Ogden (1994) has extended our understanding of intersubjectivity further. He

noted that contemporary psychoanalysis has moved beyond a positivist frame
where analyst and analysand can be regarded as separate subjects. The core of
the analytic process, in Ogden's view, is the dialectical movement of subjectivity
and intersubjectivity. Just as Winnicott (1960) noted that an infant cannot be
conceptualised apart from a maternal environment, Ogden has made a similar
point regarding analysis:

There is no such thing as an analysand apart from the relationship

with the analyst, and no such thing as an analyst apart from the
relationship with the analysand (1994, p. 63).

Projective identification, then, in Ogden's view, serves to create an

interpersonally decentred subject, an analytic third, as he has called it, and the
analysis takes shape in the interpretive space between the analysand and the
analyst. In this context, Ogden has argued against the totalistic view that
countertransference refers to everything the analyst thinks or feels. For the
concept of countertransference to be meaningful, it must be viewed as
constituting a dialectic between the analyst as a separate entity and the analyst
as a joint creation of the intersubjectivity of the analytic process. Ogden has
stressed that there are actually three subjectivities involved in psychoanalytic
work: the subjectivity of the analysand, that of the analyst, and that of the

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
analytic third. Projective identification negates the subjectivity of both
analysand and analyst while simultaneously reappropriating both subjectivities
to create a newly integrated ‘third’, a new ‘subject’ of the projective
identification process. A clear implication of this view is that a mutual projective
identification process is going on in both parties. Another implication is that the
portion of the analyst's psychic reality occupied by the countertransference is to
a large extent a new creation.

Discussion

In recent years there has been a growing awareness of a convergence between

classical analysts and contemporary Kleinians. Practitioners from both schools
of thought pay serious attention to the role of unconscious fantasy, share a
common understanding of the organisation of unconscious mental life, and
devise interpretive strategies that include aggressive themes (Richards &
Richards, in press).

Another area of convergence, the subject of this communication, is the manner

in which countertransference has come to be regarded. While analysts
associated with the more classical position (Abend, 1989; Chused, 1991; Coen,
1992; Jacobs 1993a; McLaughlin, 1991; Porder, 1987; Renik, 1993; Roughton,
1993; Sandler, 1976) have moved away from a strict adherence to the Freudian
view of countertransference as only the analyst's transference to the patient,
those associated with Kleinian and object-relations views (Joseph, 1989; Ogden,
1994; Scharf, 1992; Spillius, 1992) have moved away from the totalistic or broad
view of projective identification in which the analyst makes no contribution to
the emotional reaction induced by the patient.

The perspective that the countertransference represents a joint creation that

involves contributions from both analyst and analysand is now endorsed by
classical analysts, modern Kleinians, relational theorists, and social
constructivists. Although differences do exist, most contemporary analysts
would agree that at times the patient actualises an internal scenario within the
analytic relationship that results in the analyst's being drawn into playing a role

481

scripted by the patient's internal world. The exact dimensions of this role,
however, will be coloured by the analyst's own subjectivity and the ‘goodness of
fit’ between the patient's projected contents and the analyst's internal
representational world.

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
The similarities between projective identification as used in contemporary
psychoanalytic writing, role-responsiveness, and countertransference
enactment have been observed by a number of authors (Gabbard, 1994a, b, c;
Mc-Laughlin, 1991; Roughton, 1993; Spillius, 1992). Even Kernberg, who has
objected to Ogden's view of projective identification nevertheless has
recognised the induction of countertransference responses by the patient:

As part of my approach to interpretations of projections and

projective identification I see the need for the analyst to diagnose in
himself the characteristics of the self- or object-representation
projected onto him' (1992, p. 172).

To be sure, differences between the concepts may be identified. For example,

projective identification is a term usually associated with more primitive
patients with severe personality disorders and psychoses. This usage reflects
the analyst's experience of being overcome by an ego-alien force that feels
highly unfamiliar. To be more precise, repressed aspects of the analyst's internal
world are brought into conscious awareness that clash with the analyst's usual
self-experience. Hence, projective identification generally connotes aspects of
the patient being activated in a powerfully coercive way in the analyst. By
contrast, classical analysts writing about enactment tend to imply greater
contributions from the analyst's unconscious conflicts, although most would
acknowledge that these may be evoked by the patient's behaviour. Certainly
both groups would agree that the experience of the analyst is not an exact
replica of the patient's projected internal selfor object-representation. The
analyst's subjectivity lends a new element to the recreation of the past and the
present, what Ogden (1994) would call the ‘analytic third’.

Another difference between enactment and projective identification is that the

former implies an action. Theoretically, at least, projective identification could
involve a countertransference feeling induced in the analyst that is not carried
into action. However, if one includes the subtle shifts in tone of voice, body
posture, use of silence, and so forth described by Jacobs (1986) and McLaughlin
(1991), the line between inducing feelings and action influenced by those
feelings is a narrow one. Hence, one could conceptualise countertransference
as involving a gradient or continuum with projective identification on one end,
enactment on the other, and considerable overlap in between.

What is more controversial is whether and, if so, how, countertransference

enactments are useful. Carpy (1989) believed that the inevitable partial acting
out of the countertransference is what allows patients gradually to reintroject
aspects of themselves that were previously intolerable.

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
Other analysts (Abend, 1989; Chused, 1992; McLaughlin, 1991; Renik, 1993)
have stressed that it is the interpretive working through of the enactment that
ultimately helps the patient to change. Still others (Cooper, 1992; Jacobs, 1993b;
Ogden, 1989; Pulver, 1992) have argued that it is not an either/or proposition.
Both the events occurring in the relationship and the interpretations resulting
from those events work synergistically to produce psychic change.

In conclusion, although many differences still exist between the diverse

theoretical schools of psychoanalytic thought, a narrow area of convergence
has emerged regarding the usefulness of countertransference in
understanding the patient. There is a widespread acknowledgement that an
inevitable aspect of analysis is that the patient will try to make the analyst into
the transference object. Also, the analyst's countertransference reactions will
involve a joint creation of contributions from both patient and analyst,
suggesting that part of what the analyst experiences reflects the patient's inner
world. One of the analyst's tasks in collaboration with the patient, then,
becomes to work his or her way out of the transference-countertransference
enactment and understand interpretively with the patient what is going on. In
this regard, the intrapsychic and interpersonal realms are joined, and the
positivist perspective of the analyst as blank screen is

482

no longer viable. Another task is ongoing self-inquiry and self-analysis. As

Schafer has observed: ‘we may install the analysis of countertransference
alongside the analysis of transference and defensive operations as one of the
three emphases that define a therapy as psychoanalytic’ (1992, p. 230).

I wish to express my gratitude to Drs Stanley Coen, Lawrence Friedman, Lee

Grossman and Stephen Mitchell for their careful readings of an earlier version
of this paper.

References

Abend, S. M. (1989). Countertransference and technique. Psychoanal. Q., 48:

374-395. 

Aron, L. ( 1991). The patient's experience of the analyst's

subjectivity. Psychoanal, Dialogues, 1: 29-51. 

Bion, W.R. ( 1955). Language and the schizophrenic. In New Directions in

Psychoanalysis, ed. M. Klein. et al. London: Tavistock, pp. 220-239.

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
Bion, W.R. (1957). Differentiation of the psychotic from the non-psychotic
personalities. Int. J. Psychoanal., 38: 266-275; also in Second Thoughts.
London: Heinemann, 1967, pp. 43-64. 

Bion, W.R. (1958). On arrogance. Int. J. Psychoanal., 39: 144- 146; and in Second
Thoughts. London: Heinemann, 1967, pp. 86-92. 

Bion, W.R. (1959). Attacks on linking. Int. J. Psychoanal., 40: 308-315; and in
Second Thoughts. London: Heinemann, 1967, pp. 93-100. 

483

Bion, W.R.( 1962a). The psycho-analytic study of thinking, II. A theory of

thinking. Int. J. Psychoanal., 43: 306-310; and in Second Thoughts. London:
Heinemann, 1967, pp. 110-119. 

Bion, W.R.( 1962b). Learning from Experience. London: Heinemann. 

Bion, W.R. (1970). Attention and Interpretation. London: Tavistock. 

Boesky, D. (1990). The psychoanalytic process and its components. Psychoanal.

Q., 59: 550-584. 

Bollas, C. ( 1987). The Shadow of the Object: Psychoanalysis of the Unthought

Known. New York: Columbia Univ. Press.

Carpy, D. V. (1989). Tolerating the countertransference: a mutative process. Int.

J. Psychoanal., 70: 287-294. 

Chused, J. F. (1991). The evocative power of enactments.J. Am. Psychoanal. Assoc.,

39: 615-639. 

Coen, S. ( 1992). The Misuse of Persons: Analyzing Pathological Dependency.

Hillsdale, NJ: Analytic Press.

Cooper, A. M. (1992). Psychic change: development of the theory of

psychoanalytic techniques. Int. J. Psychoanal., 73: 245-250. 

Eagle, M. (1993). Enactments, transference, and symptomatic cure: a case

history. Psychoanal. Dial., 3:93-110. 

Gabbard, G. O. (1994a). Commentary on papers by Tansey, Hirsch, and Davies.

Psychoanal. Dial., 4:203-213. 

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
Gabbard, G. O.( 1994b). On love and lust in erotic transference.J. Am. Psychoanal.
Assoc., 42: 385-403. 

Gabbard, G. O.( 1994c). Sexual excitement and countertransference love in the

analyst.J. Am. Psychoanal. Assoc., 42: 1083-1106. 

Gill, M. M. ( 1991). Indirect suggestion: a response to Oremland's

'Interpretation and interaction'. In Interpretation and Interaction:
Psychoanalysis or Psychotherapy? by J. D. Oremland. Hillsdale, NJ: Analytic
Press, pp. 137-163.

Heimann, P. (1950). On counter-transference. Int. J. Psychoanal., 31: 81-84. 

Hirsch, I. (1993). Countertransference enactments and some issues related to

external factors in the analyst's life. Psychoanal. Dial., 3:343-366. 

Hirsch, I. (1994). Countertransference love and theoretical model. Psychoanal.

Dial., 4:171-192. 

Hoffman, I. Z. (1983). The patient as interpreter of the analyst's

experience. Contemp. Psychoanal., 19: 389-422. 

Hoffman, I. Z. (1991). Reply to Benjamin. Psychoanal. Dial., 1:535-544. 

Hoffman, I. Z. (1992). Some practical implications of a socialconstructivist view

of the psychoanalytic situation. Psychoanal. Dial., 2:287-304. 

Hoffman, I. Z. (1994). Dialectical thinking and therapeutic action in the

psychoanalytic process. Psychoanal. Q., 63: 187-218. 

Jacobs, T. J. (1986). On countertransference enactments.J. Am. Psychoanal. Assoc.,

34: 289-307. 

Jacobs, T. J.( 1993a). The inner experiences of the psychoanalyst: their

contribution to the analytic process. Int. J. Psychoanal., 74: 7-14. 

Jacobs, T. J.( 1993b). Insight and experience: commentary on Morris Eagle's

‘Enactments, transference, and symptomatic curing’. Psychoanal. Dial.,
3:123-127. 

Joseph, B. (1989). Psychic Equilibrium and Psychic Change: Selected Papers of Betty
Joseph, ed. M. Feldman & E. B. Spillius. London & New York: Routledge. 

Kernberg, O.F. ( 1992). Aggression and Personality Disorders and Perversions.

New Haven, CT: Yale Univ. Press.

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
Klein, M. (1946). Notes on some schizoid mechanisms. In Envy and Gratitude and
Other Works, 1946-1963. New York: Delacorte Press/Seymour Laurence, 1975,
pp. 1-24. 

Klein, M. (1955). On identification. In Envy and Gratitude and Other Works,

1946-1963. New York: Delacorte Press/Seymour Laurence, 1975. 

McLaughlin, J. T. (1991). Clinical and theoretical aspects of enactment.J. Am.

Psychoanal. Assoc., 39: 595-614. 

Mitchell, S. A. ( 1988). Relational Concepts in Psychoanalysis: An Integration.

Cambridge, MA: Harvard Univ. Press.

Mitchell, S. A. ( 1993). Hope and Dread in Psychoanalysis. New York: Basic

Books.

Money-Kyrle, R. E. (1956). Normal counter-transference and some of its

deviations. Int. J. Psychoanal., 37: 360-366. 

Ogden, T. H. (1979). On projective identification. Int. J. Psychoanal., 60:

357-373. 

Ogden, T. H. ( 1982). Projective Identification and Psychotherapeutic Technique.

New York: Jason Aronson.

Ogden, T. H. (1983). The concept of internal object relations. Int. J. Psychoanal.,

64: 227-241. 

Ogden, T. H. ( 1989). The Primitive Edge of Experience. Northvale, NJ: Jason

Aronson.

Ogden, T. H. (1992). The dialectically constituted/decentred subject of

psychoanalysis, II. The contributions of Klein and Winnicott. Int. J.
Psychoanal., 73: 613-626. 

Ogden, T. H. ( 1994). Subjects of Analysis. Northvale, NJ: Jason Aronson.

484

Panel (1992). Enactments in psychoanalysis. M. Johan, reporter. J. Am.

Psychoanal. Assoc., 40: 827-841. 

Pulver, S. E. (1992). Psychic change: insight or relationship? Int. J. Psychoanal., 73:

199-208. 

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
Renik, O. (1993). Analytic interaction: conceptualizing technique in light of the
analyst's irreducible subjectivity. Psychoanal. Q., 62: 553-571. 

Richards, A. Richards, A. K. (In press). Notes on psychoanalytic theory and its

consequences for technique. J. Clin. Psychoanal. 

Rosenfeld, H. A. (1952). Notes on the psycho-analysis of the superego conflict of

an acute schizophrenic patient. Int. J. Psychoanal., 31: 111-131; also in New
Directions in Psycho-Analysis. London: Tavistock, 1955, pp. 180-219. 

Roughton, R. E. (1993). Useful aspects of acting out: repetition, enactment, and

actualization.J. Am. Psychoanal. Assoc., 41: 443-472. 

Sandler, J. (1976). Countertransference and roleresponsiveness. Int. J.

Psychoanal.., 3:43-47. 

Sandler, J. ( 1987). The concept of projective identification. In Projection,

Identification, Projective Identification, ed. J. Sandler. Madison, CT: Int. Univ.
Press, pp. 13-26.

Sandler, J. (1993). On communication from patient to analyst: not everything is

projective identification. Int. J. Psychoanal., 74: 1097-1107. 

Schafer, R. ( 1992). Retelling a Life. New York: Basic Books.

Scharff, J. S. ( 1992). Projective and Introjective Identification and the Use of the
Therapist's Self. Northvale, NJ: Jason Aronson.

Segal, H. ( 1964). An introduction to the Work of Melanie Klein. New York: Basic
Books.

Spillius, E. B. ( 1992). Clinical experiences of projective identification. In Clinical

Lectures on Klein and Bion, ed. R. Anderson. London New York: Tavistock/
Routledge, pp. 59-73. 

Symington, N. (1990). The possibility of human freedom and its transmission

(with particular reference to the thought of Bion). Int. J. Psychoanal., 71:
95-106. 

Tansey, M. J. (1994). Sexual attraction and phobic dread in the

countertransference. Psychoanal. Dial., 4:139-152. 

Wallerstein, R. (1990). Psychoanalysis: the common ground. Int. J. Psychoanal.,

71: 3-19. 

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).
Weiss, J., Sampson, H. the Mount Zion Psychotherapy Research Group ( 1986).
The Psychoanalytic Process: Theory, Clinical Observation and Empirical
Research. New York: Guilford Press.

Winnicott, D. W. (1960). The theory of the parentinfant relationship. In The

Maturational Processes and the Facilitating Environment. New York: Int. Univ.
Press, 1965, pp. 37-55.

Authorized Users

For use only by apm339. Reproduction prohibited. Usage subject to PEP terms
& conditions (see terms.pep-web.org).

PEP-Web Copyright

Copyright. The PEP-Web Archive is protected by United States copyright laws

and international treaty provisions.

1. All copyright (electronic and other) of the text, images, and photographs of the
publications appearing on PEP-Web is retained by the original publishers of the Journals,
Books, and Videos. Saving the exceptions noted below, no portion of any of the text,
images, photographs, or videos may be reproduced or stored in any form without prior
permission of the Copyright owners.
2. Authorized Uses. Authorized Users may make all use of the Licensed Materials as is
consistent with the Fair Use Provisions of United States and international law. Nothing in
this Agreement is intended to limit in any way whatsoever any Authorized User's rights
under the Fair Use provisions of United States or international law to use the Licensed
Materials.
3. During the term of any subscription the Licensed Materials may be used for purposes of
research, education or other non-commercial use as follows:
1. Digitally Copy. Authorized Users may download and digitally copy a
reasonable portion of the Licensed Materials for their own use only.
2. Print Copy. Authorized Users may print (one copy per user) reasonable potions
of the Licensed Materials for their own use only.

Copyrighted Material. For use only by apm339. Reproduction prohibited. Usage subject to PEP terms & conditions (see terms.pep-web.org).