This article was downloaded by: [University of New Hampshire]

On: 13 February 2015, At: 17:33

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954
Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH,
UK

Contemporary Psychoanalysis
Publication details, including instructions for
authors and subscription information:
http://www.tandfonline.com/loi/uucp20

Countertransference
and Projective
Counteridentification
León Grinberg M.D.
Published online: 24 Oct 2013.

To cite this article: León Grinberg M.D. (1979) Countertransference and Projective
Counteridentification, Contemporary Psychoanalysis, 15:2, 226-247, DOI:
10.1080/00107530.1979.10745579

To link to this article: http://dx.doi.org/10.1080/00107530.1979.10745579

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the
information (the “Content”) contained in the publications on our platform.
However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness,
or suitability for any purpose of the Content. Any opinions and views
expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the
Content should not be relied upon and should be independently verified with
primary sources of information. Taylor and Francis shall not be liable for any
losses, actions, claims, proceedings, demands, costs, expenses, damages,
and other liabilities whatsoever or howsoever caused arising directly or
indirectly in connection with, in relation to or arising out of the use of the
Content.

This article may be used for research, teaching, and private study purposes.
Any substantial or systematic reproduction, redistribution, reselling, loan,
sub-licensing, systematic supply, or distribution in any form to anyone is
expressly forbidden. Terms & Conditions of access and use can be found at
http://www.tandfonline.com/page/terms-and-conditions
Downloaded by [University of New Hampshire] at 17:33 13 February 2015
LEON GRINBERG, M.D.

Countertransference
And Projective
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

Counteridentification
INTHIS PAPER* I SHALL PRESENT a synthesis of the ideas which I
have developed in various articles and books regarding the concept
of “projective counteridentification”. I first coined this term in
1957 to refer to a specific and differential aspect of counter-
transference, based on the unconscious analytic interaction be-
tween the patient and the analyst, and which is brought about by the
particularly intense use of and psychopathic modality of the
mechanism of projective identification of the patient. As a result of
the pathological quality of this mechanism, the patient is able to
induce different roles, affects and fantasies in the analyst, who
unconsciously and passively feels himself “carried along” to play
and experience them. (Grinberg, 1957).

I. Projective Identification

I should like to refer briefly to the mechanism of projective iden-

tification.
This was described by Melanie Klein in her paper “Some Notes
on Schizoid Mechanisms” (1946), and included in her hypotheses
about emotional development in the first months of life. It consists
in an omnipotent fantasy that unwanted parts of the personality,
and parts of internal objects can be split, projected and controlled
in the external object into which they have been projected.
Under normal conditions projective identification determines
the empathic relationship with the object, not only because it allows

* This paper will appear in a forthcoming volume on countertransference co-

edited by L. Epstein, Ph.D. and A. H. Feiner, Ph.D.; New York:Jason Aronson. I t is
published here with permission.

226
Copyright 0 I879 \C.A.\V. Institute. Neu York
All rights of repniduction in an! form reserved.
Contemporary Psychoanalysis. Vol. 15. No. 2 (IY79)
COUNTERTRANSFERENCE

one to put oneself in the place of the other and therefore under-
stand his feelings better, but also for what it brings out in that
person. The subject always produces some eniouonal reaction in
the object. His attitude, the way he looks at the object, the way he
speaks, what he says, or the gestures he makes, etc., means that
there are always projective identifications at work. They stem from
the various sources which bring them about which arouse the emo-
tional responses related to the situation, i.e., sympathy, anger, sor-
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

row, hostility, boredom, etc. This usually happens, within certain

limits, in all human relationships and forms the basis of communica-
tion. The object, in turn, also functions with his respective identifica-
tions, therefore producing an interchange.
Projective identification also plays a fundamental part in symbol
formation.
The concept of the normal and the pathological are closely re-
lated to the greater or lesser predominance of aggressive impulses,
to the degree of tolerance or intolerance of frustration, to the kind
of contact with external and psychic reality, to the state of the ego
functions, to the quality of the defence mechanisms and to the
profound dynamics of the object links. Different modalities of pro-
jective identification will be at work in accordance with the pre-
dominance of one or other of the psychic states. (Grinberg, 1965).
The normal quality of later projective identification will also de-
pend, in great measure, on the quality of the projective identifica-
tions with the first object relationships. It is not only important to
know how the subject’s projective identifications worked, con-
ditioned by various fantasies and impulses, but also how the projec-
tive identifications from the first objects worked and the type of
reactions that emerged in the subject. Another factor important to
bear in mind is that of the use (at a more organised level) of the
mechanism of “adaptive or realistic control” which allows one
control over the split parts, and those which have been projected
away. The breakdown of this mechanism could bring about distur-

I differentiated between two levels of functioning of the mechanismsof obsessive

control: one includes the more regressive aspects and corresponds to the widely
known “omnipotentcontrol”, which is chiefly related to primitive states or psychotic
conditions; the other comprises the most highly developed aspects and corresponds
to what I termed “mechanism of adaptive or realistic control” to stress that they
feature better adaptation and closer contact with reality and external objects. (Grin-
berg, 1966).

227
LEON GRINBERG, M.D.

bances of varying degrees in the functioning of the ego vis a vis the
object. (Grinberg, 1966).
The tendencies and fantasies which correspond to each of the
libidinal stages will condition the appearance of projective identifi-
cations with oral, anal, urethral and genital contents, they will also
bring about specific modalities towards the respective object rela-
tionships. Whichever stage predominates (because of fixation or
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

regression) will naturally influence the contents of the fantasies

included in the projective identifications. Those fantasies projected
onto objects at the oral level are used for eating, sucking, biting or
devouring, poison or destroying with flatus or excreta at the anal
level; scalding and destroying with urine or its equivalents at the
urethral level, etc.
It is essential to evaluate the quantity and the quality with which
projective identifications intervene so as to determine the serious-
ness of the different clinical pictures. The situation is more serious
if it coincides with an increase in sadism operating in each of the
phases. It can also happen that, on certain occasions, projective
identifications function with particular violence, as happens in the
narcissistic, psychotic and psychopathic personalities.
I should now like to present a classification of projective identifi-
cations (Grinberg, 1965).
According to thkr qualities: Normal or realistic projective identifica-
tions, and pathological projective identifications (omnipotent,
hypertrophic, with multiple splitting and “bizarre” objects, “explo-
sive”, etc.).
According to their orientations: Projective identifications directed
towards the interior of an external object, towards the surface of
the external object, towards an internal object, towards the body
and different organs, etc.
According to their aims: Communicative, reparative, evacuative,
controlling, destructive projective identifications, etc.
According to their contents: Projective identifications of aspects of
the self (ego, superego, ego ideal, organs of perception, ego func-
tions, ideas, impulses, affects, etc). Projective identifications of in-
ternal objects.
According to their clinical modalities: Hysterical, phobic, perverse,
psychopathic, obsessive, manic, paranoid, melancholoid, etc.
According to their effects on the subject: Empathy, relief, confusion,
dependence, omnipotence, claustrophobia, etc.

228
COUNTERTRANSFERENCE

According to their gfects on the object: Empathy, reactivation of

countertransferential feelings, reactions of projective counteriden-
tifications.

Melanie Klein’s papers, especially the one dealing with her con-
cept of projective identification (Klein, 1946), are sufficiently famil-
iar to require no further commentary. Her paper “On identifica-
tion” (Klein, 1955), constitutes, at present, one of the most complete
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

studies of the contents and functioning of projective identification.

According to her description, it implies a combination of the split-
ting mechanism, the subsequent projection of the split parts onto
another person, with the ensuing loss of those parts to the subject,
and an alteration of the object-perception process. This process is
bound up with those processes that take place during the first three
or four months of life (the paranoid-schizoid position), when the
splitting mechanism is at its maximum height, with a predominance
of persecutory anxieties.*
In pathology, projective identification consists in an omnipotent
fantasy through which unwanted parts of the personality and in-
ternalised objects, with their attendant emotions, are split off, pro-
jected and controlled in the object towards which the projection is
directed. As a result, the object is equated with what was projected
onto him. This operates with utmost intensity during the earliest
periods of life. It is important to scrutinise not only the different
modalities of the subject’s projective identifications as conditioned
by his varied fantasies and impulses, but also those projective iden-
tifications of his parents and the kind of impact they made upon
the subject.
11. Projective Counteridentification Reaction
As I have pointed out projective counteridentification has to do
with a very specific aspect of the countertransference (1959). It is
important to stress this to demonstrate the difference between the
response I have in mind and those countertransference reactions
resulting from the analyst’s own emotional attitudes, his neurotic

* Freud, in Group Psychology and the Analysis of the Ego (1921), describes a type
of projection very similar to the projective identificationmechanism,when he points
out the projection of the ego ideal of each of the members of the army on their
commander.

229
LEON GRINBERG, M.D.

remnants, reactivated by the patient’s conflicts. I would like to de-

scribe, in a schematic form, the two processes which. co-exist in the
analyst’s mind so the difference is made clear.
A. In one the analyst is the active subject of the patient’s introjec-
tive and projective mechanisms. In this process, three important
phases can be described: i) the analyst selectively introjects the dif-
ferent aspects of the patient’s verbal and non-verbal material, with
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

their corresponding emotional charges; ii) the analyst works

through and assimilates the identifications resulting from the iden-
tification of the patient’s inner world; and iii) the analyst (re)pro-
jects the results of this assimilation by means of interpretations
(Fliess, 1942).
B. On the other hand, the analyst can also be the “passive object”
of the patient’s projections and introjections. And two further situ-
ations may develop: i) the analyst’s emotional response may be due
to his own conflicts or anxieties, intensified or reactivated by the
patient’s conflicting material; ii) the emotional response may be
quite independent from the analyst’s own emotions and appear
mainly as a reaction to the patient’s projections on him.
The second process presents for us considerable interest, espe-
cially in connection with the problem raised in this paper. In one
phase, it is the analysand who, in an active though unconscious way,
projects his inner conflicts upon the analyst, who in turn acts as a
passive recipient of such projections.
The emphasis should be attributed to the extreme violence of the
projective identifications of the analysand. I will show, later in clini-
cal examples, how the particular intensity of this mechanism is
usually related to traumatic infantile experiences, during which the
patient suffered the effect of violent projective identifications.
Whenever the analyst has to meet such violent projective iden-
tifications, he may react in a normal way, i.e. by properly interpret-
ing the material brought by the patient and by showing him that
the violence of the mechanism has in no way shocked him. Some-
times, however, the analyst may be unable to tolerate it, and he may
then react in several different ways: a) by an immediate and equally
violent rejection of the material which the patient tries to project
into him; b) by ignoring or denying this rejection through severe
control or some other defensive mechanism (sooner or later, how-
ever, the reaction will become manifest); c) by postponing and dis-
placing his reaction, which will then become manifest with another

230
COUNTERTRANSFERENCE

patient; d) by suffering the effects of such an intensive projective

identification, and “counteridentifying” himself, in turn.
In fact, the specific response of the analyst will depend on his
degree of tolerance.
When this counteridentification takes place, the normal com-
munication between the analysand’s and the analyst’s unconscious
will be interrupted. Then the unconscious content rejected by the
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

analysand will be violently projected onto the analyst who, as the

recipient of such projective identifications, will suffer its effects.
And he will react as if he had acquired and assimilated the parts
projected onto him in a “real and concrete way”. (Grinberg, 1958,
1959).
The analyst may have the feeling of being no longer his own self
and of unavoidably becoming transformed into the object which
the patient, unconsciously, wanted him to be (id, ego, or some
internal object), or to experience those affects (anger, depression,
anxiety, boredom, etc.) the analysand forced onto him. For this
situation, I have proposed the term “projective counteridentifica-
tion”, i.e. the analyst’s specific response to the violent projective
identification from the patient, which is not consciously perceived
by the analyst. Even if the situation prevails only for a short time
(although occasionally it may persist with ensuing danger) the
analyst will resort to all kinds of rationalizations to justify his feeling
of bewilderment. (Grinberg, 1962).
This concept has been affirmed by Bion who remarked that:
The theory of countertransference offers only a partially satisfactory ex-
planation because it deals with manifestation only as a symptom of the un-
conscious motives of the analyst and; therefore, it kaves thepatunt’s cont&u-
tion without cxphnution. . . . thanks to the beta screen the psychotic patient ir
ablc to povok nnoriotlc in the anulyst. (Bion, 1961, my italics)
Bion also accepts the emergence of emotions in the analyst pro-
duced by the patient (through his projective identifications) and
which are independent, up to a point, of the countertransference
of the analyst. He goes on to state: “the use these patients make of
words is more an action directed toward freeing the psyche from
an increase of stimuli than a language”.
To understand this last statement, it would be convenient to say a
few words about the theory of alpha function, as Bion labels that
function which allows the transformation of sensorial experiences
into alpha elements which can be stored and used to form

28 1
LEON GRINBERG, M.D.

memories, oneiric thoughts, etc. They are those which permit

dreaming, thinking and maintaining the differences between the
conscious and the unconscious. On the other hand, if the alpha
function fails, beta elements appear, but they cannot be used to
form thoughts; they contain undigested facts which can only be
evacuated through projective identification and appear in the pro-
duction of acting-out. These theories offer a new contribution for
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

the understanding of the thinking processes. (Bion, 1961).

Whenever the thinking function fails, for whatever reason, we
suggest that it is replaced by projective identification which tends to
free the psychic apparatus from the increase in tension.
111. Difference Between Countertransference
and Projective Counteridentification
Based on the complementary series of Freud, Racker (1960) de-
scribed a countertransferential disposition on the one hand, and
present and analytic experiences, on the other, which leads to the
resulting countertransference. He added that this joining of the
present and the past, of reality and fantasy, of the external and
internal, etc, makes a concept necessary which envelops the whole
of the psychological response of the analyst, and he advised that the
term countertransference should be used. Nevertheless, he made it
clear that, at times, one may speak of “total countertransference”
and differentiate and separate within that term one or other aspect.
Racker emphasised the existance of a “countertransference
neurosis” where the “Oedipal and pre-Oedipal conflicts, along with
pathological processes (paranoid, depressive, manic, masochistic,
etc.), interfere with understanding, interpretation and the be-
haviour of the analyst”.
Racker made a particularly detailed analysis to two types of iden-
tification of the analyst with parts of the patient. Based on sugges-
tions of H. Deutsch, he pointed out that the analyst, with his em-
pathic tendency toward understanding everything which happens
to the patient, is able to identify “each aspect of his own personality
with its corresponding psychological part in his patient; his Id with
the patient’s Id, his ego with the patient’s ego, his superego with the
patient’s superego, accepting in his conscience these identaaiations”. (my
italics).
These are concordant or homologous identifications based, ac-
cording to Racker, on introjection and projection, in the interaction

232
COUNTERTRANSFERENCE

of the external with the internal, with the recognition of the remote
as his own (“this (you) is me”), and the association of his own with
the remote (“that (me) is you”). “Concordant identifications”would
be a reproduction of the analyst’s own past processes which are
being relived in response to the stimulus of the patient, bringing
about a sublimated positive countertransference which determines
a greater degree of empathy.
The second type of identification, called “complementary iden-
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

tifications” are the results of the identifications of the analyst with

internal objects of the patient; the analyst feels treated like those
internal objects and he experiences them as his own.
Racker also described a “concordant countertransference” where
there is an approximate identity between parts of the subject and
parts of the object (experiences, impulses, defences); and a “com-
plementary countertransference” where “an object relationship”
can exist very similar to others, a true transference in which the
analyst “repeats” earlier experiences. The patient now represents
internal objects of the analyst.
It is here where I would like to outline the difference between
Racker’s countertransferential terms and my concept regarding
“projective counteridentification”.
To begin with, confusion only arises with regard to the differ-
ence between “projective counteridentification” and “complemen-
tary countertransference”. “Concordant countertransferences” are
related to the empathic link towards the patient, the desire to under-
stand him and deal with identifications which are accepted in the
analyst’s conscience. It is worth mentioning that they almost de-
pend on an active disposition on the analyst’s part.
Let us therefore see what is the essential difference between
“complementary countertransference” and “projective coun-
teridentification”. “Complementary countertransference” arises
when the analyst identifies himself with the internal objects of the
patient and experiences them as his own internal objects. Racker
emphasizes the fact that the analyst repeats previous experiences in
which the patient represents the internal objects of the analyst. The
last experiences (which always and continuously exist) could be
called “complementary countertransferences”.
This countertransference reaction is therefore based on an emo-
tional attitude which is due to neurotic remnants in the analyst,
reactivated by the conflicts posed by the patient. It appears in the

233
GRINBERG, M.D.
L E ~ N

first situation of process B which I have described above, in which

the analyst is the object of the patient’s projections, e.g. his internal
objects; but he reacts countertransferentially because of his own
anxieties and the reactivation of his own conflicts with his internal
objects.
On the other hand, “projective counteridentification” corre-
sponds to the second situation of process B. The analyst’s reaction
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

stems, for the most part, independently of his own conflicts and
corresponds in a predominant or exclusive way to the intensity and
quality of the patient’s projective identification. In this case, the
origin of the process comes from the patient and not the analyst. It
is the patient who, in an unconscious and regressive manner, and
because of the specific functional psychopathic modality of his pro-
jective identification, actively provokes a determined emotional re-
sponse in the analyst which the analyst will receive and feel in a
passive way. (Grinberg, 1963a).
In “complementary countertransference” a reaction always arises
which corresponds to the analyst’s own conflicts. On the other
hand, in “projective counteridentification” the analyst takes onto
himself a reaction or a feeling which comes from the patient.
To clarify this point, I will use one of Racker’s examples. It is the
case of a patient who threatens the analyst with commiting suicide.
Racker writes:
The anxiety which such a threat aparks off in the analyst can lead to
various reactions or defence mechanisms within him, e.g. a dislike of the
patient. These feelings, the anxiety and the loathing, would be the contents
of the “complementary countertransference”.His awareness of dislike or
loathing towards the patient can also bring about, at the same time, a guilt
feeling in the analyst which can lead to desires of reparation and to the
intensification of “concordant identification and concordant counter-
transference”.(Racker, 1960)
Now, if we analyse this extract we find both processes superim-
posed on each other or co-existing simultaneously. (This usually
happens.) The analyst experiences anxiety in the face of the suicid-
al threat. In this anxiety, two main components are evident: one
corresponds to the analyst’s own anxiety due to the feeling of re-
sponsibility which he has, when confronted with the eventual
danger of suicide of his patient which, at the same time, may repre-
sent one of the analyst’s internal objects. (It can be the patient’s
internal object which is being experienced as one of the analyst’s
own internal objects). This form of anxiety corresponds to a “com-

234
COUNTERTRANSFERENCE

plementary countertransference”. On the other hand, the analyst

takes onto himself the patient’s specific anxiety which, through
projective identification, the patient placed in him with the idea of
the analyst controlling and eventually resolving it. This response of
anxiety now forms part of “projective counteridentification”. Later
the analyst reacts with dislike (his own mechanism, belonging to
“complementary countertransference”) and guilt. If we analyze
further this kind of guilt, we find that part of it has a persecutory
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

characteristic, i.e., “persecutory guilt” (Grinberg, 1963b). This

brings about the dislike for also having embodied (although in a
partial way), the impotence and desperation of the patient and his
fear of not being able to make a reparation. Another part of this
guilt belongs to the “depressive guilt” (Grinberg, 1963b)which the
patient is still not able to perceive nor manage, and which, pro-
jected into the analyst, makes the analyst feel able to make a repara-
tion. These last considerations with regard to the patient’s projec-
tion of the two types of guilt and the analyst’s response, dem-
onstrate how “projective counteridentification” works. However,
it does not include the two qualities of guilt which the analyst may
feel, due to his own conflicts which are reactivated by the material
presented by the patient (“complementary countertransference).
Naturally, these processes are never pure nor are they isolated;
they generally coexist in different proportions.
When mentioning other examples, Racker maintains that a
transferential paranoid-depressive state of the patient corresponds
to a “manic-countertransferential state” of the analyst, in the as-
pects of “complementary countertransference”. He is implying the
co-existance of the two mechanisms. The analyst may react mani-
cally because of his own conflicts which make him feel strong or
dominant when confronted by a depressed object; or because he
has taken onto himself the manic and triumphant attitude of the
patient which, due to the special use of projective identification has
“placed” him in that position.
Through “complementary countertransference”, each analyst
identifying himself with his patient’s internal object, will react in a
personal way according to the type and nature of his own conflicts.
Different analysts will react di,fferentlr to the same situation, posed
by a hypothetical patient. On the other hand, this hypothetical
patient using his projective identification in a particularly intense
and specific way could bring about the same countertransferential

235
L E 6 N GRINBERG, M.D.

response (“projective counteridentification”) in different analysts. I

had the opportunity to confirm this through the supervision of
material of a patient who had been in analysis succesively with
various analysts.
I n the way a transferential attitude begs a countertransferential
response, a projective identification will also beg a specific projec-
tive counteridentification. Although the analyst introjects, albeit
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

passively, this projective identification, what is important to recog-

nize is that the specific reaction of the analyst is due to the way in
which the patient projected, lodged or “forced” into the analyst his
projective identification.
Furthermore, the “projective counteridentification” will have
different modalities related to the respective modalities of the pro-
jective identification, coloured by the qualitative shading which
gives it a functional specificity. Habitually, in all extra verbal com-
munication, the type of functioning (degree and quality) of projec-
tive identification on the part of the patient does not go over the
critical threshold of the analyst and the extra verbal message pro-
duces countertransferential resonance. It stimulates the response
which could be received, controlled and verbalised with relative
ease by the analyst. But, on certain occasions in which the degree
and quality of the projective identification influence its functional
modality in a special way, the result is that the extra verbal com-
munication will pass over the critical threshold, producing “projec-
tive counteridentification”, This threshold will depend on, in each
case, the personality of the analyst, on his previous analysis and the
degree of knowledge o r awareness he has regarding this
phenomenon. (Grinberg, 1976).
I also think that sometimes the analyst, when faced with an exces-
sive projective identification on the part of the patient may respond
with a paranoid attitude which will bring about a counterresistance
and which will undoubtedly affect his work.
“Projective identification and counteridentification” phenomena
are frequent in the analysis of narcissistic and borderline per-
sonalities, and give rise to a pathogenic interaction between the
analyst and patient which is not easy to resolve. One might say that
what was projected, by means of the psychopathic modality of pro-
jective identification, operates within the object as a parasitic
superego which omnipotently induces the analyst’s ego to act or
feel what the patient wanted him to act or feel in his unconscious

236
COUNTERTRANSFERENCE

fantasy. I think that, to some degree, this is similar to the hypnotic

phenomenon as described by Freud (1921) in which the hypnotist
places himself in the position of the ego ideal and a sort of paralysis
appears as a result of the influence of an omnipotent individual
upon an impotent and helpless being. I believe the same idea
applies, sometimes, to the process I am discussing. The analyst,
being unaware of what happened, may resort to all kind of
rationalizations to justify his attitude or his bewildermentjust as the
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

hypnotised person does after executing hypnotic suggestions.

When the analyst is able to overcome this reaction, he may take
advantage of this phenomenon so as to clarify some of the patient's
unconscious fantasies and emotions making an adequate interpre-
tation possible.
IV. Clinical Examples
Some clinical vignettes will serve as examples.
1) A male patient begins his session in the following way:
I feel very nervous today. I don't know how to describe it, but it is abso-
lutely necessary that I do. I would like to tell you what I have discovered or
what has been revealed to me (with great emotion). It was so surprising the
other day when the diahorrhea stopped as a result of what you said to
me . . . Besides I remember that something else you said gave me a physical
stitch. Diahorrhea is a physical process . . . Since you spoke, these words
seem to produce a physico-chemical reaction in some or other of my nerve
cells; but before that, when you think, there is also a transformation in
other cells to the point where the voice comes from lungs, lips, tongue, etc.,
and a string of words which are now sound, vibrations, comes out. At this
moment the receptive process begins in my ear, through various means
until it becomes conscious listening. I ask myself if all those words, instead
of being spoken by you. came from someone else, would they have the
same therapeutic meaning? I think not. It is extremely important for me
that those words came from Dr. Grinberg and no-one else.
All this material was said with force and with a resounding voice
which surprised me. It was not common for him to express himself
like that and, therefore, I felt particularly attracted as much by
what was said as the way it was said. Using as a guide my impres-
sion, I interpreted that he was trying to produce in me the same
effect that he said I had had on him, and he tried to show me that
that was his voice and no one ekes which produced this special
effect on me. That is to say that my interpretation was made show-
ing his positive transference. I did not yet realise that it was only a
defence against his deepest paranoid situation.

237
LE6N GRINBERG, M.D.

The patient goes on to say:

Now that we are talking about sound and listening, I would like to talk
about music: it is divided into three basic parts, rhythm, melody and har-
mony which are undividedly joined together. I play jazz; in that we see
rhythm and the harmony of the song we are playing. The melody is impro-
vised. In modern jazz, the rhythm and the harmony are also improvised. I
can improvise for hours on melodies with rhythm but I find it difficult to
carry on a specific harmony. A melody in 8/4 time in the chord of A for
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

four beats and the other four in A sharp is impossible for me. The same
happens with written music; I cannot give the timing correctly to each note.
On the other hand, when my music teacher played one of the pieces I was
studying, I could play it after exactly by ear. In the session, for example, I
find it difficult to adapt to the reality of time. I don't even know what time
it is. It is as if I made my own time, which is different from your time. I can
compare it to my inhibition in music; this specific harmony which we im-
provise is the kind which allows people who don't know each other to
improvise a jam session.
While he was telling me all this, I did not fully understand what it
was that was happening. I felt quite uneasy. I felt sorry that I did
not understand sufficiently the theory and technique of music
which I have always loved. I admired and envied his knowl-
edge and apparent precision with which he described and
explained i t . . . with its technical jargon, the relationship between
rhythm, melody and harmony. I felt the need to interpret it in his
own words; it was a way of showing him that I could also play in the
same field as he and which he knew so well.
My interpretation was that I represented the chord of A major
and he the chord of A natural, but that between our beats there was
no harmony and we needed to find a rhythm and a timing between
the two of us which would harmonise so that we could improvise
(free association) together in a common melody.
The interpretation was now spoiled; it only demonstrated a par-
tial aspect and in a way different from what was essential. The
important thing would have been to show him his envy and that he
was really interested to discover which was my timing and which
was my rhythm. The meaning of his deepest fantasy began to dawn
on me and I payed attention to the following material.
The patient goes on:
I don't know why I thought that one could do all kinds of tests to a patient;
encephalograms. B.M.R.'s, a tape recording, a thermometer to take a
temperature with, an oscilloscope to record sound waves; anyway the use of

238
COUNTERTRANSFERENCE

all those appliances so that you would have a better knowledge of the
patient. both inside and out.
While I was listening to him I surprised myself with a parallel and
simultaneous fantasy to have a metronome to regulate, control and
direct the time in him, that is to say, to have something which I
already knew was lacking in him. I realised exactly at that moment
all the play of his unconscious fantasy contained in his intense
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

projective identification and also “how” I “counteridentified my-

self projectively with a partial aspect of his, full of envy, and anxi-
ety. One of the major effects of my “projective counteridentifica-
tion” was the blind spot of the paranoid content of his attitude and
my having stressed instead the positive aspect of the transference.
The patient used it in a defensive way to pacify the persecutor,
which I represented. But that was only his defense because of his
anxiety and panic due to the power he thought I had. My words not
only cured him of his diahorrhea but they also gave him physical
stitch. I was in possession of a secret which he envied and feared
because I could do what I wanted with him. He wanted to take this
over so that he could limit its danger and also so that he could
dominate me at the same time. For this he needed both to know me
and to control me. It was for this that he “took” me into his own
field, acoustics and music. He made me feel, projectively, what he
had felt with me. My feeling of dislike corresponded to his feeling
of anxiety. My admiration and envy reflected similar feelings which
he had felt, and my need to use his terminology and concepts was
the equivalent of his desire to take onto himself my special ter-
minology and concepts. My fantasy of the metronome, formed the
response to his desire to use all kinds of medical apparatus so as to
get to know me completely, that is to say, to control me. As a last
resort, and as a transactional solution, he offered me his beat and
timing in exchange for knowing mine.

2) In another case, a woman patient came to her first session

fifteen minutes late. She lay down on the couch and then remained
still and silent for a few minutes. After that she said she felt the
same as she used LO feel when passing an oral examination (which
usually caused her great anxiety). Then she associated the analytic
session with her wedding night when, even though she was feeling
extremely frightened, she was told she looked like a statue.

239
LE6N GRINBERG, M.D.

I told her that what she felt was that she was having with me the
same experience that she had had at her oral examinations and
during her wedding night, because she feared I might deflower
her, introduce myself into her to look at things and examine them.
Here, too, she was behaving like a statue; the rigidity and stillness
she showed at the beginning of her session were intended to dis-
guise her anxiety, but also to prevent the actual possibility of being
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

penetrated.
Although I realised that this interpretation of her paranoid anxi-
ety was correct, I had the feeling that there was something wrong
with it. Still, I could not understand the reason for such a feeling. I
guessed that my interpretation had been rather superficial and that
the facts I had pointed out to her were too near to her conscious-
ness. I had to find out the deeper motives of her exaggerated fear
of my going into her.
On the other hand, her initial attitude of stiffness had particu-
larly attracted my attention, and I found myself, not without con-
siderable amazement, having the fantasy of analysing a corpse. A
thought came at once into my mind, which took the form of a
popular Spanish saying: “she is trying to force the dead into me”
(which meant that she wanted to burden me with the whole respon-
sibility and guilt). This thought showed me my own paranoid reac-
tion, aroused by the feeling that she was trying to project her fears
into me, through projective identification.
Based on this countertransferential feeling, I told her that with
her rigidity and silence perhaps she wanted to mean something
else, besides the representation of a statue; perhaps she wanted to
express in this way some feeling of her own, related to death.
This interpretation was a real shock to her; she began to cry and
told me that when she was six years old her mother, who had
suffered from cancer, had committed suicide. T h e patient felt re-
sponsible for her mother’s death, because she had hanged herself
in her presence, and it had been actually on account of her delay in
warning the rest of the family that the death could not be pre-
vented, as had been done in former attempts. She remembered
having watched all the arrangements her mother made and being
greatly impressed by them. Then she went out and waited for a
long while (perhaps fifteen minutes, she said); only then did she
run for help, but when her father came it was too late.
I had the feeling that with her corpse-like rigidity the patient was

240
COUNTERTRANSFERENCE

not only trying to show that she carried inside a dead object, but
also, at the same time, to get rid of it by projective identification.
From that moment on, she wanted me unconsciously to take over
the responsibility, to bear “the dead”. As a defense against her
violent projective identification, with which she tried to introduce
into me a dead object, I reacted with my first interpretation, which
in fact inverted the situation: she was the one who was afraid of my
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

piercing her. Later on, I managed to grasp the actual meaning of

the whole situation, I had a much clearer understanding of the
deepest sources of her paranoid anxieties and gave her a correct
and more complete interpretation.

3) The following example was given to me by a colleague.

This deals with an obsessive patient who had gone to the analyst
to try to solve his great inhibition in studying and passing examina-
tions.
Among his previous experiences he referred in a special way, to
having had a very violent stepfather, who used to hit him extremely
hard, doing it, on occasions, at the mother’s request. He had, how-
ever, felt ill treated by everybody and in every way, in spite of his
being a well disposed man. He said he had always been “kicked. As
illustrative examples of his “victimization”, he told of traumatic
homosexual experiences which he had gone through with his elder
brother and school mates who had harrased him continuously.
However, the most characteristic and remarkable of this from
the first session, was the way which he told all this with a remarka-
ble and shocking contrast to its contents. His language was pomp-
ous, in an irritatingly pedantic way, and he used a pseudo technical
terminology to tell of his experiences. He said he knew everything.
He had read Adler, Jung, Freud and, according to him, he had also
mastered different philosophical doctrines.
The countertransferential experience of the analyst was increas-
ingly annoying and, at time, somewhat anxious. He felt he could
not make headway with his interpretations because there was no
room for them and, on the few occasions when he did manage to,
the reply was usually immediate and in terms of rejection, objection
or ridicule. As a clear demonstration of the absolute control with
which the patient wished to be the dominant partner in the trans-
ference relationship, he used to interrupt his speech (since it could
not be called free association what was happening), in order to

24 1
L E 6 N GRINBERG, M.D.

demand in an obligatory tone and in a self satisfied way the opinion

of the analyst: “How do you see all this?”.
This situation repeated itself intensely in the following session.
The patient commenced a new attack on the interpretations of the
previous sessions using his now classical destructive techniques in
order to nullify them. The analyst felt that his patient was getting to
the end of what he had made into an “unbearable nightmare”, as
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

he described it. There was something more than dislike and frus-
tration in his work. He felt the attitude of the patient to be one in
which the patient mercilessly had decided to ill treat or destroy . . .
one by one . . . all the interpretations.
If the analyst had given way to his own impulses, which were well
controlled, he would have gotten up and kicked the patient. This
fantasy was that which directed his subsequent interpretation ex-
pressed with decided annoyance showing in his voice, interrupting
his patient brusquely and telling him: ‘Just a moment, you are
behaving in a way which makes me want to kick you; just like your
brother and the rest did. It’s the way you seem to want the whole
world to be against you”.
After this intervention, the analyst thought the patient would not
return. However, this did not happen and the analyst was able to
see a decided change in the patient’s attitude or behaviour.
Evidently, the interpretation was disturbed by the enormous
emotional content which the analyst demonstrated and this was
especially reflected in the analyst’s tone of voice, the way he said it
and the intention; an almost conscious wanting to convert the in-
terpretation into the concrete substitute for the “kick”. This was
due, in a large part, as the interpretation indicates, to the fact that
he was the passive receptor of the persecutory objects which the
patient projected into him as a result of the strong repetitive com-
pulsion to look for aggressions. At another level of the transference
relationship, the patient identified himself with the aggressors,
placing, through projective identification, his punished self into the
analyst, in order to make him suffer what the patient himself had
suffered. The purpose of the projection was that the analyst was
not only the depository of this suffering aspect, but changed it,
giving it the quality of a reaction to counteract its masochistic
meaning. This the analyst also did through the intervention.

4) In another example a student in psychoanalytic training came

242
COUNTERTRANSFERENCE

to his own analytic session after having analysed a “difficult” pa-

tient. During the session with his own patient, the student had had
the feeling of “killing himself”, owing to his very active interpreta-
tions without obtaining any satisfactory result. He was depressed by
his feeling of failure, and after communicating his experience and
mood to his training analyst, he remained silent. While listening to
his analyst’s interpretations, which momentarily did not modify his
state of mind, the student had the impression that the same situa-
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

tion he had been complaining about was being repeated, although

with inverted roles. He realised that now it was his analyst who was
“killing himself” to obtain some reaction from him, while he was
acting in the same way as his patient had done. When, with some
surprise, he communicated his impression to his analyst, the latter
showed him that his behaviour during the session had “compelled”
(his own words) him to identify himself with the patient. The in-
terpretation was then completed in this sense: the student envied
his analyst for having better and easier patients (the student him-
self). A very intense projective identification had thus taken place,
by means of which the student unconsciously wanted his analyst to
experience his own difficulties. The student sorted to splitting,
projecting his hampered and dissatisfied professional part onto the
analyst, remaining with that part of himself identified with his own
patient, “who makes one work and does not gratify”. The training
analyst had, in turn, “succumbed” to his patient’s projection, and
felt unconsciously compelled to “counteridentify” himself with the
introjected part.
When this occurs-and this process is much more frequent than is
usually believed-the analysand may have the magical unconscious
feeling of having accomplished his own fantasies, by “placing” his
parts on the object. This also may arouse in him, a manic feeling of
triumph over his analyst.
Several issues presented by Hanna &gal in her paper Depression
in the Schizophrenic (1956) are closely related with the process I have
called “projective counteridentification”. These refer, especially, to
the projection of the patient’s depressive anxieties into the object
(analyst) by means of projective identification, and to the specific
response aroused in the analyst as a result of such identification. In
Segal’s words:
. . . Then one day as she was dancing round the room, picking some imagi-

243
L E 6 N GRINBERG, M.D.

nary things from the carpet and making movements as though she were
scattering something around the room, it struck me that she must been
imagining that she was dancing in a meadow, picking flowers and scatter-
ing them, and it occurred to me that she was behaving exactly like an
actress playing the part of Shakespeare’s Ophelia. The likeness to Ophelia
was all the more remarkable in that, in some peculiar way, the more gaily
and irresponsibly she was behaving, the sadder was the cffect, as though hm
gaiety itself was designed produce sadness in hm audience, just as Ophelia’s
pseudo-gay dancing and singing is designed to make the audience in the
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

theatre sad. (my emphasis)

“Projective counteridentification” was successfully dealt with by
Segal by integrating it in an adequate interpretation of her patient’s
attitude. She pointed out that the patient had put into the analyst
all her depression and guilt, thereby transforming the analyst into
the sad part of herself and, at the same time, into a persecutor,
since she felt that the analyst was trying to push her unwanted
sadness back into her.
Bion (1955)gives a clear example of the mechanism of projective
counteridentification.
The patient had been lying on the couch, silent, for some twenty minutes.
During this time I had become aware of a growing sense of anxiety and
tension which I associated with facts about the patient which were already
known to me from work done with him in the six-months he had already
been with me. As the silence continued, I became aware of a fear that the
patient was meditating a physical attack on me, though I could see no
outward change in his posture. As the tension grew I felt increasingly sure
that this was so. Then, and only then, I said to him: “You have been
pushing into my inside your fear that you will murder me”. There was no
change in the patient’s position, but I noticed that he clenched his fists till
the skin over the knuckles became white. The silence was unbroken. At the
same time I felt that the tension in the room, presumably in the relation-
ship between him and me, had decreased. I said to him: “When 1 spoke to
you, you took your fear that you would murder me. back into yourself; you
are now feeling afraid you will make a murderous attack on me.” I followed
the same method throughout the session, waiting for impressions to pile up
until I felt I was in a position to make my interpretationslt will be noted that
my interpretations depend on the use of Melanie Klein’s theoy of prqective idmtiji-
cation,first to illuminate my countertransference, and then to frame the intmpetcr-
tion which I give the patient. (my emphasis)

V. Problems Derived From Countertransference and Projective

Counteridentifkation in the Supervisory Setting
One of the most significant problems the supervisor has to tackle
consists of the student’s difficulties due to his countertransference.

244
COUNTERTRANSFERENCE

While this has already been discussed in a large number of papers,

panels and symposia on supervision, the most controversial issue
has to do with the attitude the supervisor is supposed to assume
towards the student’s countertransference. General agreement has
been reached in the sense that the candidate’s countertransference
problems should not be interpreted by the supervisor. Neverthe-
less, owing to its recurrence and intensity in students’ work, discus-
sion still goes on as to what should be the right posture for the
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

supervisor. Again I should like to point out the existence of two

different categories of countertransference issues. One is related to
those problems concerning countertransference itself. The other is
concerned with what I have called “projective counteridentifica-
tion”.
I believe it is necessary for the supervisor to be able to differen-
tiate between the two categories when he encounters such issues
during supervision. Countertransference has to be dealt with on
the couch in the student’s training analysis, projective counteriden-
tification in supervision. When the candidate presents difficulties
that stem mainly from his own conflicts the supervisor may refrain
from making direct remarks, but point out the existance of difficul-
ties and show how to approach the dynamics of the patient. Some
authors suggest making the situation explicit, advising that the can-
didate work it out in his analysis. However, this is questioned by
others as potentially having a disturbing effect on the candidate.
The supervisor’s task should then be to take care of the candi-
date’s difficulties since it is in supervision that he could help the
candidate become aware of a conflict or misunderstanding of
which the student may not be conscious. When the candidate is
thus disturbed, the supervisor has the advantage of available mate-
rial which will disclose the cause for the problem. It is essential that
the supervisor be able to illustrate from available clinical material,
where, how and why the therapist has been liable to the projective
counteridentification reaction.
The student can make the supervisor feel an emotional reaction
of the same quality as the one the patient has aroused in him. If the
supervisor has a full understanding of the genesis of his own affec-
tive response, he can, with greater ability, objectivity and experi-
ence, show the candidate the origin of the emotional reaction he
experienced during the session with his patient.
Similar problems have been discussed by other authors, although

245
LEON GRINBERG, M.D.

from a different frame of reference, with different terminology.

Sometimes, I wonder whether a good deal of disagreement be-
tween analysts is not more due to semantic differences rather than
to conceptual ones. Arlow (1963) says, in reference to what I have
termed projective counteridentification, that the therapist shifts
from a role that consists of reporting the experience he has had with
his patient to experiencing his patient’s experience; the supervisor
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

may find evidence that the candidate enacts an identification with

his patient. Searles (1955) states that the relationship between pa-
tient and therapist is often revealed by the relationship between
therapist and supervisor. He adds that the supervisor has a
spectrum of emotional phenomena-as the therapist and the patient
have theirs-which often throw light on classical countertransfer-
ence reactions and at times are highly informative reflections of the
therapist-patient relationship. He calls them part of a “reflection
process” to stress their source and points out that they can be a
decisive clue to obscure difficulties which emerge in the therapist-
patient relationship. Searles suggests that unconscious identifica-
tion is one of the processes involved in the genesis of this
phenomenon.
When discussing a case at supervision, the supervisor may have
the feeling that the anxieties and defenses of the therapist, while
reporting the session, are an unconscious way of communicating
something that occurs with the patient, which the student’s own
anxiety prevents him from describing to the supervisor. T h e
supervisor can be more easily aware of the nature of the problem
not only because his knowledge and experience are wider, but also
because he is more emotionally detached.
Hora (1957) accounts for this kind of phenomena in the follow-
ing way:
The supervisee unconsciously identifies with the patient and involuntarily
behaves in such a manner as to elicit in the supervisor those very emotions
which he himself experienced while working with the patient but was un-
able to convey verbally”. He further on adds that “the therapist orally
incorporates or introjects his patient in his endeavour to empathetically
understand him. While on the conscious level the supervisee proceeds with
the presentation of the factual data about the patient, unconsciously on a
non-verbal level he communicates the affective aspects of his experience
with the patient. This carries the dynamic aspects of the patient’s personal-
ity make-up.

246
COUNTERTRANSFERENCE

REFERENCES
Arlow, J. A. (1963). The supervisory situation. Journal A d a n Psychoanalytic As-
sociation, 11:576-594.
Bion, W. R. (1955), Language and the schizophrenic. In: New Directions in Psycho-
Anal$, ed. M. Klein, et al., London: Tavistock, 1955.
Bion, W. R. (1961). Learning from Expencnce. London: Heinemann.
Fliess, R.(1942). Metapsychologyof the analyst. P+oandyEic Quarterly, 11:211-227.
Freud, S. (1921). Group psychology and the analysis of the ego. StandardEdition, 18.
Downloaded by [University of New Hampshire] at 17:33 13 February 2015

London: Hogarth Press, 1955.

Grinberg, L. (1957) Perturbaciones en la interpretacibn motivadas por la con-
traidentificacion proyectiva. Revista & Psicoa-, 1423-30.
Grinberg, L. (1958). Aspectos migicos en la transferencia y en la contratransferen-
cia. Identificacibn y contraidentificacibn proyeaivas. Revista & Psicoanalicis, 15.
Grinberg, L. (1959). Sobre algunos problemas de ttcnica psicoanalitica deter-
monados por la contraidentificacibn proyectiva. Revista & Psicoanalisis, 16.
Grinberg, L. (1962). On a specific aspect of countertranference due to the patient's
projective identification. Intrmational J o u d Psycho-Analysis. 43436-440.
Grinberg, L. (1963a). Psicopatologia de la identificacibn y contraidentificacibn pro-
jectivas y de la contratransferencia. R&ta & Psicoanalicis, 20.
Grinberg, L. (1963b). Culpa y Dgrcsidn. Buenos Aires: Ed. Paidbs.
Grinberg, L. (1965). Contribucion a1 estudio de las modalidades de la identificacibn
proyectiva. h t a & P s i c o a d . 22:263-278.
Grinberg, L. (1966). The relationship between obsessive mechanism and a state of
self disturbance: depersonalization. International Journal of Psycho-Analysis.
47: 177-183.
Grinberg, L. (1976). Temia de la Identzjiuacwn. Buenos Aires: Ed. Paidbs.
Hora, T. (1957). Contribution to the phenomenology of the supervisory process.
A d a n J o u d of Psychotherapy. 11:769-773.
Klein, M. (1946). Notes on some schizoid mechanisms. IntmMtional Journal of
Psycho-Analysis. 27.
Klein, M. (1955). On identification. In: New Directions in Psycho-Analysic. ed. M.
Klein, et al. London: Tavistock, 1955.
Racker, H. (1960). Estudios sobre Tecnica a Psicoanalitica. Buenos Aires: Ed.
Paidbs.
Searles. H. (1955). The informational value of the supervisor's emotional experi-
ence. Psychiaby. 18:135-146.
Segal, H. (1956). Depression in the schizophrenic. InurnatMlal Journal of Psycho-
Analysis. 37.

Jose Ortega y Gasset 85, 7"

Madrid 6 , Spain

247