COUNTERTRANSFERENCE

Comments on Its Varying Role in the Analytic Situation

RENE A. SPITZ, M.D.

The subject of countertransference has been widely debated;

the concept itself is still ill defined. Since Freud first coined the
termJ1 despite frequent attempts at its definition both from the
pragmatic and from the theoretical viewpoint, no final agreement
has been reached on its formulation. We have made no effort to
duplicate the excellent historical survey on countertransference by
Om (12). Since Orr’s review appeared, Benedek (1, 2) and Racker
(13) published further important papers on the subject. For the
purposes of the present paper we will give a working definition
covering the range of the phenomena of which we intend to speak.
Furthermore, we will consider countertransference as something
which takes place between two persons, the analyst and his patient.
We will define countertransference as one part of the analyst’s
relation to his patient; it is one of the determinants of the emo-
tional climate of a given analytic relationship; it usually originates
in the analyst; its manifestations are varied. T h e particular shape
it takes is due to the way in which the given patient’s personality,
his behavior, and the manifestations of his transference act on and
are responded to by the given analyst’s personality. The response
will begin with a dynamic process in the analyst’s unconscious.
This will translate itself into derivatives, expressed in the attitude
of the analyst. When the patient becomes aware of the analyst’s
1 To my knowledge the term “countertransference”was used €or the Erst time by
Freud in his letter to Ferenczi dated October 6. 1910 (8, p. 83).
256
 COUNTERTUNSFERENCE 255
4. Benedek, T. Adaptation to reality i n early infancy. Psychoanal. Quart. 7:200-215,
1938.
5. Benedek, T. Countertransference i n the training analyst. Bull. Afenninger Clin.,
18~12-16,1954.
6. Berman, L. Countertransference and attitudes of the analyst in the therapeutic
- process. Psychiatry, 12159-166,1949.
7. Cohen, hf. B. Countertransference and anxiety. Psychiatry, 15:231-243, 1952.
8. Flies, R. Countertransference and counteridentifiation. This Journal, I:268-
284, 1953.
9. Freud, S. (1910) The future prospects of psycho-analytic therapy. Collected
Papers, 2285-296. London: Hogarth Press, 1946.
10. Frosch, J. et al. (Eds.) Annual Survey of psychoanalysis, 1:237-248. New York:
International Universities Press, 1950.
11. Gitelson, hi. T h e emotional position of the analyst i n the psycho-analytic situa-
tion. Internat. J. Psychoanal., 33: 1-10, 1952.
12. Glover, E. Technique of Psycho-Analysis. New York: International universities
Press, 1955.
13. Greenson. R. Panel discussion on “Sublimation.” This Journal, 3:525-527. 1994.
14. Heimann, P. On counter-tramference. Internat. J . Psychoanal., 31:81-84, 1950.
15. Johnson, A. Transference and copntertransference i n late analysis of the
oedipus. (To be published.)
16. Little, &I. Counter-transference and the patient’s response to it. Internat. J.
Psychoanal., 32:3240, 1951.
17. Lorand, S. Technique of Psychoanalytic Therapy. New York: International Uni-
versities Press, 1916, pp. 209-222.
18. Orr. D. IV. Transference and countertransference: a historical survey. This Jour-
nal, 2621-650, 1954.
19. Payne, S. M. Notes on develbpments in the theory and practice of psycho-ana-
lytic technique. Internat. J . Psychoanal., 27: 12-18, 1916.
20. Racker, H. A contribution to the problem of counter-transference. Internat. J.
Psychoanal., 3 ~ 1 3 - 3 2 4 1953.
,
21. Reich, A. On counter-transference. Internat. J. Psychoanal., 3225-31, 1951.
22. Rioch, J. T h e transference phenomenon in psychoanalytic therapy. Psychiatry,
6:147-156, 1943.
23. Sharpe, E. F. The psycho-analyst. Internat. J . Psychoanal., 28:l-6,1947.
24. Tauber, E. S. Exploring the therapeutic use of countertransference data. Psy-
chiatry, 17:331-336, 1954.
29. IVinnicott, D. IV. Hate i n the counter-transference. Internat. I. Psychoanal.,
30~69-74.1949.
26. Round-table Discussion: Doctor-patient relationship i n therapy. Am. J. Psycho-
anal., I5:3-21, 1955.
 COUNTERTRANSFERENCE 257
attitude, changes in the nature of the patient’s transference take
place. Thus a circular process between analyst and patient is set
in motion which determines the analytic climate.
T h e analyst’s countertransference may be manifested either in
a sublimated form or in the form of id derivatives or as a crude
expression of a drive. The function of countertransference in the
given analytic relationship will be determined on one hand by the
form in which it is manifested, on the other by its content. I n the
further course of this paper we will speak of some aspects of
countertransference as well as of its, genetic origin and we will
attempt to investigate its metapsychology.
T h e interpersonal aspects of this definition of countertransfer-
ence coincide to a large extent with the one given by Annie Reich
(14). We have for the moment omitted making the distinction
which she has clearly established between countertransference
proper, and the more general concept of countertransference
worked out by her and Fenichel (3) in their discussion on this
topic, and which includes the analyst using the analysis for acting-
out purposes.
It follows from our definition of countertransference that we
believe that it is constantly present in analytic work and that it is
a normal phenomenon. As Adolph Stern (16) remarked, counter-
transference in the analyst is exactly the same phenomenon as
transference in the patient. Gitelson (5) expressed the same idea
when he suggested abolishing the term of countertransference and
calling it instead “the analyst’s transference to the patient.”
One of the most important single contributions on the subject
is Annie Reich‘s article (14); it has influenced much of my own
thinking. She states, “Countertransference is a necessary pre-
requisite of analysis. If it does not-exist, the necessary talent and
interest is lacking.” In this statement she indicates clearly the
constructive function of countertransference. She explains that
the psychological interest of the analyst is based on a very com-
plicated countertransference which is desexualized and subli-
mated in character. I n contrast with this she cites pathological
examples in which the analyst’s conflict persists in its original
form and in which the analytic situation is used by him for one of
258 RENe A. SPITZ
three purposes: (a) for the living out of the underlying impulses;
(b) for defending against these impulses; (c) for proving that n o
damage has occurred i n consequence of these impulses.
Racker (13), and i n a recent paper Lucia Tower (17), have
introduced the concept of “countertransference neurosis.” Racker
defines it as an independent entity, as the pathological part of
countertransference and the expression of neurosis. H e states that
the countertransference neurosis, like any other neurosis, and also
like the transference neurosis, is centered i n the oedipus complex.
I n view of the recent conceptualization of this particular aspect
of countertransference when it becomes pathological, it seems
worth while to discuss it i n some detail, particularly since the
question has been raised whether t h e countertransference neurosis
might not have its uses i n treatment, just as the transference
neurosis has.
I n countertransference neurosis, like i n any other neurosis,
affect is released. Only in countertransference neurosis it is the
analyst who releases his own affects, not the patient. We will now
examine the conditions under which countertransference affects
are released by the analyst, and the influence this has on the
patient.
We attribute a goodly part of the therapeutic effectiveness of
the analytic treatment to the release of affect by the patient and
to his affective re-experience of repressed memories, This results
i n a modification of his personality. T h a t is not necessarily the
patient’s intention; his intention, when he releases affects, is to
modify the relationship between himself and the analyst. This
may involve foisting on the analyst a role which is not justified by
the reality situation. But the analyst refuses to change either the
relationship or his own role i n the relationship; instead of this he
attempts to understand the patient’s affective behavior and to
transmit this understanding to him.
One may then ask how analytic therapy can be benefited if the
analyst releases affects as a consequence of a countertransference
neurosis. If we pursue the analogy between the patient’s releasing
affects and the same process i n the analyst, it would seem that such
a release by the analyst is an attempt either to modify his relation
 COUNTERTRANSFERENCE 259

to the patient or to modify the role he plays in this relation, or

both. Eventually it would result in a n effort of the analyst to
modify the patient himself. As a therapeutic goal the latter
might be acceptable. As a therapeutic method it is open to ques-
tion. T h e method certainly is not in accordance with the prin-
ciples of psychoanalysis and the goals of analytic training. T h e
latter endeavors to replace in the analyst the need to release affects
i n the treatment situation by insight and understanding. This does
not imply that affects should not arise in the analyst in response
to his patient’s productions, nor that he should be in any way
rigid, inflexible or not subject to change. We will discuss further
in what way the analyst’s personality should be flexible and what
use he should make of his affects.
Continuing our parallel between countertransference neurosis
and any other neurosis, we are reminded that neurosis is character-
ized by its compelling nature-a fact also stressed by Racker (13).
This does not apply to compulsion neurosis only. Any neurotic
finds himself under the inner constraint to act i n terms of his
neurosis rather than in terms of reality. H e is under an inner
constraint to act out his neurosis. He is under the constraint of the
repetition compulsion. H e is not a free agent.
That is exactly the opposite of what we expect of the analyst,
whose activity i n the treatment situation should be only controlled
by his ego. An analyst acting under the compulsion of id impulses
in the treatment situation has relinquished his therapeutic role.
T h e degree of freedom available to the analyst is well formulated
in the witticism of a Viennese comedian, who described the nor-
mal person as somebody who may do anything, but does not have
to do it.
We would then say that in the analytic procedure countertrans-
ference neurosis in the analyst is not only not useful but highly un-
desirable. TYhat we expect of the analyst is that he achieve a
countertransference sufficiently sublimated, so that he can make
use of it i n identifications of brief duration with his patient. This
process has been aptly described by Kris (9) i n regard to the artist.
H e called it a “regression in the service of the ego.” That is ex-
actly what we expect of the analyst.
260 RENE A. SPIT2
I n my recent discussion of transference (15) I referred cursorily
to its dynamics and to some of its economic aspects, while elabo-
rating extensively its genetic aspect.
If we agree to consider countertransference as an analogue of
transference in the patient, then it follows that its genetic history
is the same as that of transference. In other terms, it is a new edi-
tion, a facsimile, of impulses and fantasies belonging to the past.
T h e past to which they belong, as I have shown in the previously
mentioned communication, is the earliest parent-child situation.
I n agreement with Greenacre (7), Macalpine (Il), Lagache (lo),
and others, I explained how the situation of the child’s helpless-
ness was re-created in the analytic setting and would inevitably
result in the reproduction of fantasies originating in that situation.
Countertransference assigns a role to the analyst which is the
obverse of that of the patient. The patient is helpless, while the
analyst’s role is to be helpful. The situational stimulus in the
analytic setting which acts on the analyst is, therefore, the patient’s
helplessness. It evokes in the analyst fantasies derived from the
ego ideal which he formed in identification with his parents.
We have postulated that the analytic setting places the patient
into an anaclitic relationship. I may be permitted to suggest a
distinctive term for the role of the analyst in this setting. Anaclitic
means leaning onto; I recommend for the analyst’s attitude the
term diatrophic? which means supporting.
T h e diatrophic attitude has its origin in a developmental stage
of the infant which emerges toward the end of the anaclitic rela-
tionship. T h e diatrophic attitude is a facsimile of the fantasies
which belong to the stage in which the young child forms his
secondary identifications with the parental figures. I am referring
here to those early make-believe games, to be seen in the first half
of the second year, when the child feeds its teddy-bear from a nurs- -
ing bottle, copies the nurse in a nursery by distributing diapers
to the other children, etc.
There is a basic countermovement in the unfolding and the
fate of the anaclitic attitude on the one hand, of the diatrophic
2 From the Greek: Gra-~pCqm=tomaintain. to support throughout.
 COUNTERTRANSFERENCE 26 1

attitude on the other. T h e anaclitic relationship is based on an

experience of which in the normal course of development the
reality aspects recede progressively and are lost, leaving behind
them only memories and wish-fulfillment fantasies. With advanc-
ing age anaclitic relations are relegated more and more to the
realm of fantasy or pathology.
T h e diatrophic relation begins with an identification fantasy,
but with progressive development will end u p in the reality situa-
tion of the subject becoming himself a parent.
I n the analytic setting, in the ideal case, both anaclitic and
diatrophic relations have to operate on the level of fantasies, con-
scious and unconscious, triggered by the conditions of the setting
itself. Neither of them should be translated into action. T h e
patient, in acting out, attempts to achieve reality fulfillment. T h e
aim of the rule of abstinence is to frustrate this fulfillment. The
analyst becomes able to impose this frustration on the patient only
if he himself does not act out the diatrophic attitude. He has to
understand the origin of his diatrophic fantasies sufficiently to be
able to accept as a matter of course that the rule of abstinence
operates for himself as much as it does for the patient. This is a
point touched upon also by Racker. I would say specifically that
this unconditional acceptance of the rule of abstinence requires
not only the working through of the analyst’s oedipal and pre-
genital development, but also his becoming able to relinquish the
archaic wish for magic omnipotence.
Acting out the diatrophic attitude, of course, is not the only pit-
fall of his own unconscious which the analyst faces. If we disregard
the well-known and extensively discussed acting-out possibilities
presented by unresolved problems of the analyst in connection
with the partial drives of the pregenital phase and those connected
with the conflicts of the oedipal phase, there still remains the
temptation for him to succumb to an unconscious wish for an
anaclitic relationship to his patient. Obviously the analytic setting
does not make acting out the latter easy. But as acting out i n a
countertransference neurosis disregards reality, it does not pre-
clude it either. Needless to say that to act out anaclitic wishes is as
undesirable for the therapeutic process as the other forms of act-
262 RENd A. SPITZ
ing out; indeed, it is one of the more dangerous ones of these
forms.
T h e early secondary identificatory fantasies which underly the
diatrophic attitude have a very great adaptive value for individual
development. They have mostly been studied in their significance
for pathology and little has been written about their importance
i n the formation of personality. They operate at first in the process
which I have called the humanization of the infant. This begins
with the acquisition of language and of the first elements of the
“Do’s and Don’t’s.” Eventually these secondary identifications will
serve the process of the socialization of the child. T h e formation
and the subsequent liquidation of the oedipus complex is but one
of the stations on this road.
I n the course of the child’s development, the progressive elabo-
ration of these identifications is ensured on one hand by the
pleasure gain of drive satisfaction; on the other, these identifica-
tions are infinitely valuable to the child by providing him with
ever-increasing mastery over the environment and with the con-
comitant narcissistic gratification. Throughout life successive and
ever more intricate elaborations of the diatrophic attitude in iden-
tification with the parents establish a genetic sequence in its devel-
opment. Various stages of these identifications will be used for the
purpose of occasional regressions.
I n the analyst, the diatrophic attitude offers two possibilities.
If ego-controlled, his brief regression to the parent ideal can be
made therapeutically effective for the patient. Such transient iden-
tifications with the parent ideal enable the analyst to empathize
with the infantile aspects of the patient’s behavior and to re-
interpret them in terms of infantile experience. On the other
hand, this same regression, if not controlled by the ego, may give
the therapist the opportunity to find a gratification of repressed
drives. Identifications with the patient, with his parents, or with
the vicissitudes of their relations which permit the surfacing of the
analyst’s repressed drives should be considered as acting out on
his part.
I have stated that I do not consider acting out desirable on the
part of the analyst. Such acting out can be an occasional one, pro-
 COUNTERTRANSFERENCE 263
voked by transference manifestations of the patient. Alternatively,
it can take place in the framework of a real countertransference
neurosis, as a consequence of the neurotic personality of the
analyst. In either case, it can only be an obstacle in what we call
the analyst’s understanding of the patient.
As we have stressed, a great deal of the analyst’s insight results
from brief temporary identification with the patient, that is, from
ego-controlled regression on the part of the analyst. If the ego-
controlled regression is replaced by acting out, then the analyst
can no longer remain aware of the derivatives of his own uncon-
scious and cannot make appropriate use of them i n therapy. TYhen
the analyst acts out in response to the patient’s provocation, an
interchange of acting out between analyst and patient takes the
place of an understanding of the patient’s productions. Therefore,
the analyst’s acting out cannot lead to a therapeutic interpretation.
When acting out replaces interpretation, the results will some-
times be spectacular. Such results are comparable to the successes
seen in cathartic therapy. T h e dynamics of the two are different.
But the successes will be haphazard and transitory at best, and it
is to be expected that the drawbacks of such methods will far out-
weigh ‘their advantages.
Acting out, as stressed by Reich (14) and Fenichel (3), is but
one of the manifestations of countertransference. It is the most
obvious and easily recognized one. There is much less unanimity
on what constitutes the other forms. I believe that many of our
disagreements on countertransference are caused by misunder-
standings provoked through the careless use of the term i n our
writings. We are prone to speak of countertransference, which is
an unconscious process, when what we really mean are its con-
scious derivatives. It is only with these that we can deal on the
conscious level: taking cognizance of these derivatives enables us
to perform what Glover (6) called “the analytical toilet.”
One of the reasons why the analytic candidate undertakes a
training analysis is to enable him to recognize the underlying
unconscious motivation of these conscious derivatives in himself.
When he performs this task, he becomes able to fulfill the dia-
trophic role of the analyst: like the parent ideal, he can tolerate
264 RENB A. SPITZ
aggression as well as the pressure of the patient’s libidinal demands
without retaliating in kind. H e can permit the patient’s initiatives
to unfold into directions, however different from his own ideals,
as long as they do not endanger the patient. Analysis has to be
carried out in abstinence, said Freud; I may add, abstinence of
the patient and abstinence of the analyst. For the analyst this does
not apply to countertransference as such, but to its acting out, as
well as to those others of its forms which are not syntonic with the
requirement of free-floating neutral attention.
Countertransference is a necessary prerequisite of analysis. Its
proper use involves three steps:
1. T h e analyst becomes aware in himself of the derivatives of his
unconscious as they arise in response to the patient’s unconscious.
2. From these derivatives he infers the underlying unconscious
processes in himself.
3. H e now has to possess sufficient freedom to perform a transi-
tory identification with those processes in the patient which had
provoked his own responses.
This, then, would be my concept of the metapsychology of what
we call “understanding the patient.”

BIBLIOGRAPHY

1. Benedek, T. Dynamics oE the countertransference. BuIZ. Afenninger CIin. 17:

201-208, 1953.
2. Benedek, T. Countertransference i n the training analyst. Bull. Afenninger Clin.,
18:12-16, 1954.
3. Fenichel, 0. Theoretical implications of the didactic analysis. Privately printed.
Topeka Institute for Psychoanalysis.
4. Freud, S. (1910) T h e future prospects of psycho-analytic therapy. CoNected
Pupers, 2:285-296. London: Hogarth Press. 1946.
5. Gitelson, M. The emotional position of the analyst in the psycho-analytic situa-
tion. Internat. J . Psychoanal., 33:l-10, 1952.
6. Glover, E. (1927) T h e Technique of Psycho-Analysis. New Sork: International
Universities Press, 1955.
7. Greenacre, P. The role of transference. Thi s Journal, 2:671-684, 1954.
8. Jones, E. T h e Life and Work of Sigmund Freud, 2. New Sork: Basic Books,
1955.
9. Kris, E. T h e psycho lo^. of caricature. Psychoanalytic Explorafions in Art. New
Sork: International Uruversities Press, 1952.
10. Lagache, D. Le problkme d u transfert. Rev. Frunr. d e Psychanal., Ik5-122, 1952.
11. Macalpine, I. T h e development of the transference. Psychoanat. Quart., 19:501-
519, 1953.
 COUNTERTRANSFERENCE 265
12. Om, D. I V . Transference and countertransference. This Journal, 2621-670,
1954.
13. Racker, H. A contribution to the problem of counter-transference. Internat. j .
Psychoanal., 34:315-324, 1953.
14. Reich, A. On counter-transference. Internat. J . Psychoanal., 32:3-7, 1951.
15. Spitz, R. A. Transference: the analytical setting and its prototype. Internat. j .
Psychonnal. (in press).
16. Stem, A. On the countertransference in psychoanalysis. Psychoanal. Rev., 11:
166-174, 1924.
17. Tower, L. E. Countertransference. This Journaf, 43224-255, 1956.