The Psychoanalytic Study of the Child

ISSN: 0079-7308 (Print) 2474-3356 (Online) Journal homepage: http://www.tandfonline.com/loi/upsc20

The Mutual Influences in the Development of Ego

and Id

Heinz Hartmann

To cite this article: Heinz Hartmann (1952) The Mutual Influences in the Development of Ego and
Id, The Psychoanalytic Study of the Child, 7:1, 9-30, DOI: 10.1080/00797308.1952.11823149

To link to this article: https://doi.org/10.1080/00797308.1952.11823149

Published online: 13 Feb 2017.

Submit your article to this journal

Article views: 2

View related articles

Citing articles: 13 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=upsc20
THE MUTUAL INFLUENCES IN THE DEVELOPMENT
OF EGO AND ID
Symposium held at the Seventeenth Congress of
the International Psycho-Analytical Association,
Amsterdam, Holland, August 8, 1951.
THE MUTUAL INFLUENCES IN THE DEVELOPMENT
OF EGO AND ID

By HEINZ HARTMANN, M.D. (New York)

I cannot say that I feel too much at ease in introducing this sym-
posium on The Mutual Influences in the Development of Ego and Id.
There is in analysis hardly a topic that is more comprehensive. Whatever
I could tell you would hardly add up to an integrated picture. The time
allotted would not even suffice for a catalogue of the problems involved.
But I do hope that this very difficulty, of which you are no doubt as well
aware as I am, will prevent you from accusing me of any sins of omission;
and that you will extend to me the privilege of a personal approach: the
right to accentuate freely and, above all, to select for my presentation
only certain aspects of our problem, while discarding many others, though
they may be of equal importance for an integrated psychoanalytic theory
of development.
I shall submit to you for discussion some possible avenues of approach,
trying to place the problem, as it were; make a few suggestions for clari-
fying, evolving and integrating some of its aspects; and will start, as is
customary, with some historical remarks, which, however, I shall try to
limit to a minimum.
The concept of an ego you find already in Freud's (1950) physiological
psychology of 1895 and in some clinical papers dating from the same pe-
riod. These first formulations were followed by years of great discoveries:
the psychological foundation of analysis in The Interpretation of Dreams;
the libido theory; the insight into the etiology of neurosis; the genetic
turn-that is, the discovery of the decisive relevance of early life history;
and the development of psychoanalytic technique. During these years the
role of the ego is little emphasized and at times even completely sub-
merged under the impact of the theory of instincts. Only in the twenties
was ego psychology explicitly defined as a legitimate chapter of analysis.
The ego evolves as one system of personality, clearly set apart from the
functions of the id and of the superego. This renaissance of the ego con-
cept encompasses Freud's insights into the unconscious and the instinc-
tual drives, the disregard of which had been a deadly limitation of the
9
10 HEINZ HARTMANN

usefulness of other, preanalytic concepts of the ego. Freud outlined an

ego which is infinitely richer in importance, dimensions, and specificity
of functions, in comparison with his earlier formulations. On this later
level Freud's ego concept, though elements of early formulations have
been integrated, appears as something essentially new, also as to its
effects, because of the revolutionizing impact it had on the development
of many aspects of psychoanalysis, including the theory of instinctual
drives. This development, by the way, always struck me as one rather
clear-cut example for a tenet of the philosophy of Hegel, who saw the
evolution of concepts in terms of thesis, antithesis and synthesis.
To approach more closely the ego-id problems under discussion today,
we may say that this growing in stature of the ego's role in Freud's think-
ing can be seen: structurally, in its description as a partly independent
unit of personality; dynamically, in Freud's warnings against a simplify-
ing generalization he had noticed in the work of some analysts, which
tended to underrate the ego's strength vis-a-vis the id (see also A. Freud,
1936); economically, in the hypothesis of its being fed by a mode of
energy different from that of the drives. The independent aspect of the
ego is even more conspicuously stressed in one of Freud's later proposi-
tions, suggesting the hereditary nature of some of its elements.
In developing his ideas on ego-id relationship, Freud followed the
lead of technical and clinical as well as theoretical insights. The interest
in these problems extends from technological detail to the most abstract
level of theory formation. However, we should not forget that the aspects
of the ego we see, viewing it from the angle of resistances, are not neces-
sarily the same as those which are in the foreground in the study of, let
us say, psychosis, and neither the one nor the other of these groups of
aspects will fully coincide with that part of the ego which becomes visible
in the direct observation of children. Thus, partial ego concepts devel-
oped which Freud succeeded in integrating in his more general propo-
sitions. Different facets of Freud's thinking on ego and id have been
worked out by different analysts in different directions. Besides the nature
of the data used, theoretical preferences have an obvious influence on
an analyst's centering his research on one rather than on another of these
partial concepts of the ego. To emphasize only one partial concept of the
ego, at the expense of other aspects, may be a question of expediency
vis-a-vis specific problems. But we shall remember that the reality ego,
the defensive ego, the organizing, the rational, the social ego; the ego
that leads a shadowy existence between the great powers, the id and the
superego; the ego evolving under the pressure of anxiety situations, are
not "the ego" in the sense of analytic psychology. These are partial con-
cepts to be distinguished from Freud's general ego concept.
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 11

Freud knew that the reliability of our statements and particularly of

our predictions depends, in analysis, as in other sciences, among other
things, on how comprehensive and consistent a general theory has been
developed. He wanted to get insight "into the entirety of mental func-
tions," as he wrote very early. That is, he aimed, as he repeatedly said,
beyond his clinical research at what one could call general psychology,
encompassing normal phenomena as well as pathological ones. This has
remained one trend in his work through all the years. Its outline in
Freud's work is considerably more comprehensive than what has yet been
systematically elaborated in psychoanalysis. He often said that his not
yet having dealt with some problem did not imply his negating its
relevance.
I mention this here because what we are to discuss today is actually
one aspect-perhaps the most important one in the present situation of
analysis-of such an analytical approach to general psychology. It obvi-
ously transcends a narrower concept of analysis that would limit it to the
understanding and therapy of neurosis. It aims at normal as well as
pathological development. Secondly, the dealing with these develop-
mental problems also often transcends what is directly accessible to the
psychoanalytic method. I am speaking of the child's growth and develop-
ment up to the end of the preverbal stage. Still, this trend in analytic
research is relevant also for a better understanding of clinical and tech-
nical problems, and it will become important particularly in regard to
questions of mental prophylaxis.
Guarded extrapolations from what we know about later stages of
development to earlier stages are widely used in the genetic hypotheses
of psychoanalysis. It is amazing how much analytical reconstruction has
taught us even about those primordial stages. Still, a host of questions as
to the relative relevance of our various constructs, about the chronology
in the development of different functions, and so on, remain contro-
versial. In this situation the most auspicious development is the recent
introduction into analytic child psychology of direct observation of the
growing infant and child by analysts or at least analytically trained
observers (see A. Freud and others). This can be helpful in checking our
genetic hypotheses against observational data; and it can be decisive in
giving us positive clues for the formation of hypotheses. We can learn
from the correlation of reconstructive data with data of direct child
observation how the latter can be used as indicators of structurally central
developments, etc. This trend has already given our knowledge of early
ego-id development an incomparably greater concreteness, especially as
to its reality aspects. Here, then, not only the "negative" aspect of the
ego, its role as adversary of the drives, but also many other specific ego
12 HEINZ HARTMANN

functions and their interrelatedness become of necessity a legitimate con-

cern of the analyst. This is a decisive step toward a general analytic
theory of motivation.
It has also become apparent, I think, that to speak of the ego in a
summarizing way as, let us say, threatened by the id or helpless vis-a-vis
the id, as is often done, is no longer a sufficient description of develop-
mental reality even in those early stages. It is not always advisable to
conceive of these relations between ego and id as if they were just two
opposing camps (Freud, 1926). The object of research is the great variety
of developing ego functions, in their antagonistic but often also syner-
gistic interdependence with the id, and their differential consideration
(intrasystemic approach; Hartmann, 1951).

In speaking of the mutual influences in the development of ego and

id, we are used to considering the former, more often, the dependent,
the latter, more often, the independent variable. We are impressed by
the flexibility, by the learning capacities of at least parts of the ego, and,
on the other hand, by the stubborn opposition to change of the instinc-
tual drives. Still, there are those changes in the id that are brought about
by the growth or development of the instinctual drives through all their
subsequent phases; also, the ego can take a measure of influence by
draining the instinctual energies of the id or damming them up; there
are those modifications that, via the ego, analysis can induce in the id;
there is, although it may not be fully understood yet, the id aspect of the
outcome of repression (see also E. Bibring, 1937). Freud (1926) felt that
his originally general assumption that repressed impulses remain un-
changed in the id might be in need of revision. This might not be the
only possible outcome of repression. Two cases would have to be con-
sidered: "mere repression and the true disappearance of an old desire or
impulse." Repressed instinctual tendencies may lose their cathexes, which
could then be used in different ways. In the case of the breakdown of
the oedipus complex, according to Freud, they are sublimated and used
in the resulting identifications. In other cases one may think of a kind
of displacement of these energies that might help to promote the next
step in instinctual development, an important proposition which has
been suggested by A. Karan-Angel (1951).
The strength of the ego in its relationships with the id lies in finding
ways that make discharge possible; or, in other cases, in imposing changes
of aims, or of the modes of energy involved; in the capacity to build
countercathexes; in its control of perception and motility, and in its use
of the danger signal and access to the pleasure-uri pleasure principle. One
aspect of ego development can be described as following, in several re-
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 13

spects, the lead of the drives. We are used to speaking of an oral and
anal ego, and so on, and trace specific ego attitudes to specific libidinal
characteristics of the correlated phase. This aspect shows the phases of ego
development in close connection with the sequence of libidinal phases.
However, while rich clinical material and also data of direct observation
testify to the importance of this relation, the ways in which ego attitudes
are formed by the characteristics of the libidinal phase are not always
clear. I think that in some cases the characteristics both of the instinctual
tendencies and of the attitudes of the ego may have a common origin in
the undifferentiated phase. Of giving, getting, etc., we can assume that they
are modeled after instinctual patterns. A partial modeling after instinc-
tual patterns we may also assume in the case of some defense mechanisms,
as for instance in identification and projection (Hartmann, 1939a). But
to describe ego formation only in terms of its dependence on instinctual
development is to give only part of the picture. This is only one of its
facets, among several; a point to which I shall return in more detail later.
While describing the development of the child in terms of libidinal
phases, we are today very much aware of the fact that cross sections of
development cannot be completely described in referring only to libidinal
aims-not even if we include the corresponding object relationships in
our description. We have to describe them also with respect to the involve-
ment of two other series of factors: the vicissitudes of aggressive drives
and the partly independent elements in the ego. It might well be that
even the timing and the individual formation of the typical phases could.
to some extent, be traced to individual variations of ego development,
e.g., to the precocity of certain of its functions, which might become
relevant also for pathology (Hartmann, 1950b).
Some aspects of earliest ego-id interrelations could be partly clarified
through the study of regressive phenomena in psychosis." and also, for
instance, of the phenomena occurring during the process of falling asleep
(Isakower, 1938). For the understanding of the same problems, in some
instances, and of different ones in others, the approach through the study
of the body ego and of object relations has proved essential. The body
being the mediator between the inner and outer world, and what we call
objects being the emotionally most relevant representatives of the latter,
the approach through the body ego and the object relations is also the
preferred access to studying how ego-id relations develop in the individ-
ual's interaction with the environment. The development of the body
1 I may add that today, as one consequence of progress in analytic child psy-
chology, a clarification also in the opposite direction, from the knowledge about in-
fancy and early childhood to a better understanding of psychosis, seems to be well
under way.
14 HEINZ HARTMANN

ego will be discussed by Hoffer.s I shall at this point say a few words
about object relations; or, rather, about only a few facets of object rela-
tions that seem relevant to our discussion. Freud (1926) found that "the
influence of the environment is intensified, the differentiation of the ego
from the id is promoted very early, the dangers which the environment
presents are increased in importance, and the value attached to the object
who alone can offer protection against these dangers . . . is enormously
augmented," as a consequence of the protracted helplessness and depend-
ence of the human child. We may also say that in the human the pleasure
principle being a frequently unreliable guide to self-preservation, and
the id, as Freud once said, neglecting it, the development of a specific
organ of learning and adaptation, the ego, has become of vital impor-
tance. This we could call a circular process. The ego-id differentiation
complicates the relations between pleasure and preservation of the self.
The id, in obvious contrast to the instincts of the animals, neglects the
latter. But this very fact probably acts as a stimulus for further ego-id
differentiation (Hartmann, 1948). I am emphasizing here the specifically
human side of these problems, the distinction between ego-id structures
of man and the instincts of lower animals, as a basis for later discussion
of ego-id differentiation.
It is in approaching the problem of the child's interaction with his
objects, of his indulgences and frustrations, that the study of the "reality
factor" and the interest in ever more specific situations in the child's life
became particularly meaningful-what Kris (1950b) called the "new con-
sideration for the environment." On the side of theory, one aspect of this
trend is clearly based on that part of Freud's reformulations which traces
internal danger situations to external ones, and on the subsequent work of
A. Freud and others. For the time being it is this trend in analysis, above
others, which quite naturally leads to a development that was briefly
mentioned before: the integration of the reconstructive data of analysis
with data gained from the systematic, not merely occasional, use of direct
observation of children, and to an increased concern for a more inclusive
view of child development. Some of these studies, as you know, also
include an investigation of the most important objects in the child's life
(mostly the mother, who is studied together with the child). Thus, for
instance, the relevance of the mother's conflicts in the shaping of the
child's attitudes and defenses can sometimes be traced (E. Jackson and
E. Klatskin, 1950).
Such newer studies show in detail the participation of instinctual and
ego tendencies in the development of the child's object relations. What
2 See this volume, pp. 111-41.
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 15

we call "satisfactory object relations" has an id- but obviously also an

ego-aspect. In recent years the impact of incomplete or empty relation-
ships with the mother on ego development has been emphasized repeat-
edly (Durfee-Wolf, 1933; Ribble, 1943; Spitz, 1945; and others). While
these findings are valuable and no doubt valid, the danger of overempha-
sizing and oversimplifying this side has not always been avoided. The
fact that the mother has "rejected" her child in one way or another is
frequently, in unilinear causal relation and rather indiscriminately, made
responsible for nearly all varieties of later pathological developments and
particularly of ego disturbances. That the ego needs, in order properly
to function and to develop, a secure relation not only to the drives but
also to the objects, is obviously true. But ego development and object
relationships are correlated in more complex ways than some recent
works would let us believe-which we could already expect on theoretical
grounds. We do not know much about corrections of very early unsatis-
factory situations through later maturational processes," It might also be
that not only can "poor" early object relations be sometimes made up
for by later ego development; but also that so-called "good" object rela-
tions may become a developmental handicap-probably, I should think,
if and in so far as the child has not succeeded in utilizing them for the
strengthening of his ego. Also there is a long way from the object that
exists only as long as it is need-satisfying, to that form of satisfactory
object relations that includes object constancy. The work done by A.
Freud and her co-workers has an immediate bearing on this subject.
This constancy probably presupposes on the side of the ego a certain
degree of neutralization of aggressive as well as libidinous energy (a con-
cept we shall discuss later); and on the other hand it might well be that
it promotes neutralization.s That is, "satisfactory object relation" can
only be assessed if we also consider what it means in terms of ego
development.

Of all the manifold relationships between ego and id, the one of con-
flict, the one in which the instinctual drives come to be considered as a
danger-in which case the anxiety signal induces defense of the ego-is
the one by far most familiar in analysis. It is the one most immediately
relevant for our clinical work and at the same time, because of specific
features of our technique, the one best accessible to our method. Thus
most of our clinical knowledge on the interaction of ego and id we owe
to the study of conflict.
S See, however, Lois Murphy (1944); also Beres and Obers (1950); and now the
important paper by A. Freud and S. Dann (1951).
4 For this second aspect see also A. Freud (1949) and E. Kris (1950b).
16 HEINZ HARTMANN

However, we also speak of collaboration of ego and id and in doing

so seem to point to a variety of processes: The ego may serve the aims of
the id; or the energy of the id is available for the aims of the ego; there
may be substitution of ego aims for id aims, or neutralization of instinc-
tual energy. The two last-mentioned processes often go together but may
also vary partly independently, as is the case in sexualization.
What the methods used by the ego and its defensive actions are, and
what these mechanisms mean in terms of the ego and of the id, has been
stated with great precision in the classical contributions of Freud (1926),
Anna Freud (1936), Nunberg (1932), and others. Freud's ideas about
countercathexis brought us a metapsychological grasp of the ego aspect.
This subject of fundamental significance, conflict and defense, is today
among the best known chapters of analytic theory, clinic and technique-
though some aspects, as for instance the chronology of defense mecha-
nisms, stilI pose a number of unsolved problems.
At this point, I want to discuss only some aspects, developmentally
relevant but more or less at the periphery of defense itself. It has proved
useful to isolate for specific purposes the setup "defensive action-warded-
off impulse." But, of course, for the developmental approach-and, for
that matter, even sometimes for the clinical or technical aspects-it
becomes relevant and indeed necessary also to ask how in a developmental
cross section or in a longitudinal section, considering the predisposition
to or the precursors (and also the aftermaths) of defense, this setup is
interrelated to other functions of the ego. This is what I had in mind in
speaking of the interrelation of the conflictual and nonconflictual spheres
of the ego (1939a). Factors in the nonconflictual sphere codetermine the
methods by which instinctual stimuli are dealt with, or, more specifically,
the ways of conflict solution, and are in turn influenced by the latter. To
study these processes seems particularly relevant in the early stages in
which not only the use but the development of the defense mechanisms
is in question.
There is a factor of another order that may have an influence on
conflict, a factor whose origin also transcends the factors immediately
involved in the conflict situation. I am thinking of a proposition formu-
lated by Freud in one of his last papers; it has so far been given little
attention. Freud (1937) suggests that there may exist an individually
varying tendency toward conflict which, independent of the conflict
situation itself, could be correlated with the presence, or the amount, of
free aggression. He suggests that we might "... review all of our knowl-
edge of physical conflicts from this new angle." I tried some time ago
(1950a) to develop Freud's suggestion in a specific direction, about which
I shall say a few words later.
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 17

In the context of today's discussion we may ask what the antecedents

of the ego's turning against the id are. This direction of the interest of
some analysts is somewhat analogous to the turning toward the preoedipal
phases also, after the main aspects of the oedipal situation had been
explored. In what follows, I shall merely touch on a few points of the
earliest phases of this development. In the last part of my paper I shall
then turn to a later phase: to how some of the more remote consequences
for the ego of the ways in which conflicts with the id have been dealt
with can be evaluated.

Earliest stages of ego development can be described as a process of

differentiation that leads to a more complete demarcation of ego and id
and of self and outer reality; as a process that leads from the pleasure
to the reality ego; as the development of the reality principle; as the way
leading from primary narcissism to object relationships; from the point
of view of the sequence of danger situations; as the development of the
secondary process, etc. The important thing for a systematic study of
the subject, which, as I said, is not intended here, would be to clarify the
interrelatedness of all these aspects of ego development.
In the earliest postnatal stage it is difficult to disentangle the nuclei
of functions that will later serve the ego from those that we shall attribute
to the id. Also, it is often hard to decide what part of it could already
be described in terms of mental functioning. Neither is there at that
stage any differentiation of the self from the world outside. That there is
no ego in the sense we use the term for later stages, seems clear; what the
state of the id is at that level is unknown. This stage we may term the
undifferentiated stage (Hartmann, 1939a; Hartmann, Kris and Loewen-
stein, 1946). This conception of the earliest postnatal stage seems to be in
agreement with Freud's later thoughts. At least once, in the "Outline"
(1939), he speaks of "... the id, or rather, the undifferentiated ego-id."
In speaking of ego-id differentiation, Freud introduces hypotheses
some of which clearly follow anatomical or physiological models; and he
uses not only ontogenetic but also phylogenetic hypotheses. I am not
concerned here with studying the interrelation of ontogenetic and phylo-
genetic propositions, interesting as such an attempt might be. Also, our
acceptance of phylogenetic hypotheses depends more on our adherence
to this or that school of evolutionism than on our analytic experience
and thinking. Anyway, in the present context it becomes important
clearly to demarcate the two sets of hypotheses.
In ontogenesis, the id-ego differentiation follows the leads of outer
and inner perception, of motility, and of the systems of preconscious
memory traces, of experience and learning. The replacement of hallu-
18 HEINZ HARTMANN

cination by thinking, of direct motor discharge by action, are essential

elements in Freud's theory of ego development. The body in its double
position as part of the inner and also of the outer world plays a decisive
role in this process-above all, as Freud said, its surface, but also those
stimuli that reach the mental apparatus from the inside of the body, and,
in a specific way, pain. After Freud, Schilder (1938), Bychowski (1943),
Scott (1948), Hoffer (1949, 1950) have helped us to get some glimpses into
these many-faceted developments. Hoffer's recent studies of the early
connections between oral functions and the use of the hand, and of the
role they play in the development of the primitive ego, have clarified
one of the earliest and most consequential steps.
I shall discuss one aspect of these differentiating processes in greater
detail; not necessarily because it is the one that seems most important,
but because its role in ego development has not always been clearly
realized. Generally speaking, the apparatus serving perception, motility,
and others that underlie ego functions, seem, in the infant, to be ac-
tivated by instinctual needs. Their use independent of immediate needs,
and in a more differentiated relation with external stimuli, is already
part of the development of the reality ego. But they are not created by
the needs. These apparatus, as well as those that account for the phenom-
ena of memory, are partly inborn; they cannot be traced, in the indi-
vidual, to the influence of the instincts and of reality, and their matura-
tion follows certain laws which are also part of our inheritance." They
will gradually come under the control of the ego; on the other hand,
they act on the ego and its subsequent phases of development (Hartmann,
1939a, 1950a). They can also be considered as one factor among those to
which the ego-id differentiation can be traced. Here, then, is one of the
points where phylogenetic hypotheses have to be clearly set apart from
ontogenetic ones, if we want to avoid misunderstandings. The differen-
tiation of ego and id, developed by whatever process of evolution through
hundreds of thousands of years, is, in the form of a disposition, in part
an innate character in man. That is, this differentiation does not start
from scratch in every newborn child.
It is tempting to view this aspect of ego development in a way analo-
gous in principle, though not in extent, to that which we have long
since accepted in accounting for the libidinal phases. In tracing their
significance we are used to considering the anatomical and physiological
5 It is rather generally accepted in biology that part of what we call maturation is
developed without the guidance of function as such, and that it may have adaptive
significance only in reference to its future function (Weiss, 1949). However, the same
author adds that there is, of course, no rigid preadaptedness "to fit precisely one par-
ticular detailed course of life." For the difference between adaptedness and adapta-
tion see also Hartmann (1939a).
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 19

growth processes underlying them. Freud mentions the importance of the

appearance of the teeth, the development of the anal sphincters, etc. I
think something similar holds good for the development of the ego-
maturational processes in the motor apparatus and the interaction with
specific ego functions we may consider one case in point. A detailed
knowledge of the stages of development on the side of the ego will be
our most valuable guide in extrapolating reconstruction-in deciding
what degrees of differentiation and integration of function, what degree
of mechanism formation, can be assumed to exist on a given develop-
mental level. Again, here as elsewhere, the fact that I emphasize in this
context one facet-in this case, maturation-should not be misconstrued
as any underrating of the specific importance of learning processes for the
development of the ego.
This consideration of maturational processes also on the side of ego
development seems natural enough if we keep in mind that the ego
aspect of development is no less "biological" than its id aspect. It seems
hard to call nonbiological the functions of adaptation and of synthesis,
or integration, or organization (that is, the centralization of functional
control), both of which we attribute to the ego. In a late paper Freud
(1939) even attributes to the ego, and not to the instinctual drives, the
function of self-preservation in man: "The ego has set itself the task of
self-preservation which the id appears to neglect." As to the physiological
aspect of the problem, Freud always maintained that in some future time
physiological data and concepts would be substituted for the psycho-
logical ones, referring to all mental functions and not only to those of
the id, I may add that analysts as well as physiologists have, I think
correctly, emphasized that it is particularly the study of the ego functions
which might facilitate a meeting between the psychoanalytic and the
physiological, especially the brain-physiological, approach.
In the ego's relationship with the body, we can now describe three
aspects: the postulated physiological processes underlying activities of the
ego; the apparatus that gradually come under the control of the ego and
which in turn influence the timing, intensity, and direction of ego devel-
opment; and, third, but not necessarily independent of the two others,
those special structures that underlie what we call the body ego.
In his last years Freud thought that some aspects of the defense mech-
anisms may have a hereditary core. At the time he wrote The Ego and
the ld he did not think that ego functions could be inherited the same
way as he assumed that certain characteristics of the instinctual drives
were. However, he states in "Analysis Terminable and Interminable"
(1937): "We have no reason to dispute the existence and importance of
primal, congenital ego-variations," and "It does not imply a mystical over-
20 HEINZ HARTMANN

estimation of heredity if we think it creditable that, even before the ego

exists, its subsequent lines of development, tendencies and reactions are
already determined." I think that these formulations of Freud's we should
not fail to consider in this discussion. The role of analysis, vis-a-vis this
aspect of development, can be based on what Freud (1924) once wrote in
discussing hereditary versus environmental influences: "After all, it is of
interest to follow up the way in which the innate schedule is worked out,
the way in which accidental noxiae exploit the disposition."
Those inborn characteristics of the ego, and their maturation, would,
then, be a third force that acts upon ego development, besides the impact
of reality and of the instinctual drives. Of the elements on the side of the
ego which originated in this hereditary core, whose development is of
course not independent from the development of other elements, but
which enter this development as an independent variable, we may speak
as of autonomous factors in ego development (primary autonomy) (Hart-
mann, 1939a, 1950a).
It may be that very early processes in the autonomous area-cathectic
organizations, but also physiological mechanisms that develop in inter-
dependence with them, factors like postponement of discharge and also
what Freud calls the protective barrier against stimuli (see also Bergman
and Escalona, 1949), and even reflectory defenses against unpleasant
stimuli-are genetically speaking precursors of what at a later stage we
call defense mechanisms (Hartmann, 1950a).
In summarizing this part of my presentation, may I say that certain
aspects of the choice and of the chronology of defense mechanisms might
become better accessible to our understanding once we possess a closer
insight into the development of their precursors. From the point of view
of method, I may mention that at least some of them could be approached
by direct observation. Little is known so far about what the role of such
factors might be in what has been called "primary" disturbance of the
ego (see, for instance, Hendrick, 1951).

One should also try to describe all ego-id correlations with regard to
their energic aspects. We think with Freud that the ego habitually uses a
mode of energy different from that used by the drives. He speaks of
desexualized and also of sublimated energy. We also know that if energy
serving the functions of the ego comes too close to the state of instinctual
energy (sexualization), this results in a disturbance of function. It does
not seem too hazardous to enlarge this idea of Freud's to include neutral-
ization of aggressive energy (Hartmann, 1948); this has also been done by
K. Menninger, 1938; Jeanne Lampl-de Groot, 1947; etc.), which may
serve functions of the ego and, maybe in a somewhat different state, also
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 21

of the superego. Of the modified aggressive energy used in the ego it also
seems true that if its state comes too near to the instinctual mode, this
may interfere with ego function (Hartmann, 1950a). The term neutraliza-
tion, used here and elsewhere, is meant to cover, besides what Freud
called sublimation (which he limited to one of the vicissitudes of the
libidinal drives), also the analogous change in mode of aggressive drives.
If we assume the widest possible concept of neutralization (including
sublimation), we may say that, though it may serve defense, it is of a far
more general nature than other processes used for defensive purposes.
Neutralization in this sense may well be a more or less constant process-
if we are ready to assume that all the ego functions are continuously fed
by it. But it is this very character that gives it its specific importance for
the understanding of ego-id relations, also outside of the sphere of
conflict.s
Although the general energic features of typical ego and id functions
are no doubt as we are wont to see them, it seems not unlikely that there
exist, looked at from this angle, transitions between instinctual and fully
neutralized energy. It is probable that aggression used by the superego
against the ego is closer to the instinctual condition of energy than the
one used by the ego in some of its functions. Probably correlated with
this aspect are the degrees to which the primary process has been replaced
by the secondary process.
Neutralization of energy seems clearly to be postulated from the time
at which the ego evolves as a more or less demarcated substructure of
personality. And viewed from another angle, we might expect that the
formation of constant object relationships presupposes some degree of
neutralization. But it is not unlikely that the use of this form of energy
starts much earlier and that already the primordial forms of postpone-
ment and of inhibition of discharge are fed by energy that is partly
neutralized as to some of its aspects. Some countercathectic energy dis-
tributions arise probably in infancy. Again, these and related phenomena
seem easier to understand if one accepts the hypothesis of gradations of
neutralization as just outlined.
A further complication is added by the fact that we know a rather
wide field of phenomena that we could describe as Janus-faced in the
6Neutralization, even where it is used for defense, stands apart from other defen-
sive techniques of the ego in so far as it is specially defined by its energic aspect
(among others) , which means here, by the change of one mode of energy into another
one. That sublimation is not really a "mechanism" in the usual sense, Fenichel (1945)
has clearly seen, and this holds good also for neutralization in general. Also, that its
relation to countercathexis is different from the one we find in other forms of defense.
However, I cannot follow Fenichel when he simply equates sublimation with successful
defense.
22 HEINZ HARTMANN

sense that one aspect shows the primary and the other the secondary
process. To use Anna Freud's (1936) example, in displacement as a
mechanism of defense, a characteristic of the primary process is used for
the purposes of the ego. This we also clearly see in dreams. Also,
processes we describe as vicissitudes of the instinctual drives may at the
same time be used by the ego for its own purposes (see also Eidelberg,
1940). As for the case of displacement, we may add that in a way it also
is a primordial form of learning. It widens the child's experience and is
a primitive basis on which the integration and differentiation of experi-
ences may be built. I think that M. Klein (1930) thought along similar
lines in emphasizing the relevance of symbol formation for ego de-
velopment.
There are many early and important developments which we have
learned to consider as two-faced, that is, as to their ego and their id
aspects. From the point of view of developmental psychology it becomes
relevant to see whether the two aspects are co-ordinated in the way we
would expect, according to our knowledge of parallel development in
ego and id; on which side the functional accent-if I may say so-lies at
a given stage; whether one of the aspects has outdistanced the other, etc.
Cases in which the expected equilibrium between ego and id develop-
ment is lacking often give us a good opportunity for insight into the
psychological structure of the developmental phase in question. Some-
times, disorder in the typical sequence of danger situations may result.
Precipitating or retarding factors we find both on the side of the id and
of the ego; precocious ego development, for instance, may be due to
specific instinctual demands (danger situations); to early identifications;
to an unusually early development of the body ego; to autonomous ele-
ments, etc. Here again I select for discussion only one developmental
aspect of object relations. We can view them from the angle of the needs
involved, but they also have a cognitive side, a perceptual side, and so
on. "Object formation" has a somewhat different meaning in analytic
and nonanalytic child psychology. Still, I emphasized long ago that what
nonanalytic psychologists have carefully described in their experimental
work as the evolving of constant and independent objects in the child's
world (as tested for instance in the child's handling of toys, etc.), cannot
be fully understood without considering the child's object relations in
our sense (see also Spitz and Wolf, 1949). One may suggest that the ele-
ment of identity and constancy in what one calls "objects" in the general
sense is partly traceable to the element of constancy gradually developing
in what we describe as libidinal or aggressive object cathexis-though, of
course, other factors too, partly autonomous, are involved. The child
learns to recognize "things" probably only in the process of forming more
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 23

or less constant object relationships. We assume that progress in neutral-

ization is involved in both steps, and that as to this factor both steps have
a common origin. Also the development of what one calls "intention-
ality"-the child's capacity to direct himself toward something, to aim at
something, in perception, attention, action, etc., a process that according
to Freud probably presupposes hyper-cathexis-could be viewed as one
ego aspect of developing object relations. Actually, intentionality is
among the first achievements of the child we would not hesitate to char-
acterize as true ego functions. Others among the especially develop-
mentally interesting, but little explored object-directed ego tendencies
should be systematically approached in the same way.
To come back to the energic aspect: What Freud once called "the
witch metapsychology" would, by any other name, be what we have to
appeal to in questions of general psychoanalytic psychology. Actually to-
day we come to considering it not so much as something "meta," beyond
psychology, but just as the most general level of psychological concepts
in analysis. On principle we should be able to describe all the relations
we find between ego and id as to the modes of energy they use, but also
in terms of cathexis. We are far from fulfilling this demand. Some aspects,
after Freud, have been studied, for instance, by Glover and Rapaport.
Kris (1950a) recently approached the problem of preconscious mental
functions from this angle. The conscious and preconscious phenomena
are characterized by the secondary process, one aspect of which is inhi-
bition of discharge, and they are specific of the ego in contradistinction
to the id. To describe the preconscious in metapsychological terms has
become even more important since Freud (1939) no longer thought that
another characteristic upon which he had previously relied-that is, the
addition of word representations to thing representations-is typical of
all the mental processes in the preconscious. Evidently, for the questions
of ego-id differentiation and interrelation, it is essential to trace how
the secondary process originates. Glover (1935), taking his point of
departure in the earliest systems of preconscious memory traces, describes
the syntheses of such psychic elements associated with drive components
as nuclei of ego formations. Out of this stage an organization of memory
gradually evolves which has learned to consider elements of reality.
Rapaport (1950, 1951) puts particular emphasis on the assumption that
involuntary delay of instinctual discharge, due to external circumstances,
can later be converted into an ability to delay, that is, into internal
control. This hypothesis fits in rather well with what we recognize as one
characteristic of ego development: that is, the gradual active use by the
ego for its own purposes of primordial forms of dealing with stimuli.
Internal control is one aspect of the problem of countercathexis, which
24 HEINZ HARTMANN

Freud repeatedly tried to account for, and one fundamental aspect of ego-
id differentiation. But the fundamental question, which way the original
transformation of the primary energy distribution into that representing
instinct control takes place, is still in need of further clarification. It
might be, as I have already mentioned, that those inhibitory apparatus
serving postponement of discharge, which are gradually integrated into
the ego and which are probably also precursors of later defense mecha-
nisms, playa role in the change of one mode of energy into another one.
One may ask what we can say about the nature of the drive energies
whose mode is being changed in the process of the formation of counter-
cathexis. Again, it seems hazardous at present to venture an hypothesis
with respect to this aspect of the primordial or precursory steps of differ-
entiation. For a later stage, I tried to find an answer in the synthesis of
two of Freud's hypotheses: the one, mentioned above, which says that
free aggression may be an important factor in the disposition to conflict;
and the other, which assumes that the features of defense against in-
stinctual drives are modeled after defense in situations of danger from
without. Withdrawal of cathexis would correspond to flight, and counter-
cathexis to fight. On the basis of these two hypotheses we may develop
the suggestion (Hartmann, 1950a) that the ego's countercathexes against
the drives are likely to be mostly fed by some shade of neutralized
aggression, which nevertheless still retains some characteristics of the
original drives (fight).7 This assumption may well carry us a few steps
further also in the understanding of pathological development. I think
that the failure to achieve stable defenses, a failure we see in various
forms of child pathology and which is also a crucial problem in schizo-
phrenia, is to a large extent due to an impairment of the capacity to
neutralize aggressive energy.s This hypothesis also implies a double
correlation of stable defense with constant object relations, if what I said
before is true: that the development of constant object relations on
the one hand facilitates, but on the other also depends on, neutralization.
However, these and related implications referring to pathology I shall try
to present in a more detailed and systematic way elsewhere.
This hypothesis would imply that countercathexis may be a rather
general way of utilizing aggression in one of its neutralized forms-a way
different from the utilization of aggression in the service of the super-
ego, though maybe not quite independent from it. It might well be that
the aggressive superego pressure on the ego also results in the ego's
utilizing aggressive energy in its dealings with the id-a kind of turning
A somewhat similar proposition was formulated by M. Brierley (1947).
7
For another aspect of impairment of neutralization in the pathology of schizo-
8
phrenia, see Hartmann (1950a).
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 25

one aggressive intersystemic relation (superego-ego) into another (ego-id).

This may be one energic aspect, the aspect referring to the conditions and
distributions of energy, of the role of the superego in repression and of
other phenomena familiar to all of us from clinical experience. However,
I do not think that this dependence of aggressive ego defense on the
function of the superego applies to all of its forms; and it cannot apply,
of course, to the early stages of defenses.

I shall devote the last part of my paper to one aspect of ego-id rela-
tionships at developmental stages at which the ego has already evolved
as a definable psychic system with specific functions. It has acquired,
through its prehistory, the capacity to institute and utilize some methods
to avoid danger, anxiety, unpleasure. It has developed functions, such as
objectivation, anticipation, thought, action, etc., and it has achieved a
more or less reliable synthesis, or integration, or organization, of its own
functions and of the whole of psychic personality. The very complexity of
the system tends to increase its lability, as Freud has pointed out. How-
ever, we find that various functions of the ego may achieve various de-
grees of virtual independence from conflicts and from regressive tenden-
cies in various individuals. What I have in mind here is the question of
their reversibility or irreversibility, the question of their relative stability
vis-a-vis inner or outer stress. Obviously many, though not all, attitudes
of the ego can be traced to genetic determinants in the id, to the sphere of
the instincts-or also to defensive processes. We are used to see that ego
interests and other ego tendencies may originate in narcissistic, exhibi-
tionistic, aggressive, etc., drive tendencies. We also see that, for instance,
reactive character formation, originating in defense against the drives,
may gradually take over a host of other functions in the framework of
the ego. That, under certain conditions, the ego's achievements can be
reversible, we see in neurosis, psychosis, in the dream, in analysis. Beyond
this, we may say that ego functions, if activated, often tend to exert an
appeal, sometimes more, sometimes less marked, on their unconscious
genetic determinants; also that an attraction from the latter to the former
takes place; and there is no doubt that some of this we find also in the
normal waking life of what we would call healthy people. But there are
relevant differences in the degree to which ego functions maintain their
stability, their freedom from those potential regressions to their genetic
antecedents. At any rate, in the healthy adult this partial reversibility is
not incisive enough to create serious trouble. The degree of secondary
autonomy, as I (1950a) have called this resistivity of ego functions against
regression, is a problem equally relevant for our clinical, theoretical, and
technical work. It is closely linked up with what we call ego strength and
26 HEINZ HARTMANN

probably is the best way to assess it. The problem of secondary autonomy
obviously also overlaps with the problem of mental health and has to
be studied in normal development as well as from the angle of pathology.v
The relative independence of ego functions from id pressure can be
expressed in terms of distance from ego-id conflicts, or distance from the
regressive trends exerted by the id determinants: One aspect of the latter
can, in terms of the energies involved, also be described as distance from
sexualization or aggressivization. As to the developmental aspect of the
ego functions' distance from conflict and distance from the drives, it
appears relevant that newly acquired ego functions show a high degree
of reversibility in the child and that special devices are used by him in
his effort to counteract regression (A. Freud, 1951; E. Kris, 1951).
I may add that occasional regressions in the service of the ego (Kris)
can be tolerated by the adult ego if its functions are unimpaired. We
also know that the healthy ego, for certain purposes, has to be able to
abandon itself to the id (as in sleep, as in intercourse). There are also
other less well-studied situations in which the ego itself induces a tem-
porary discarding of some of its most highly differentiated functions
(Hartmann, 1939b). To do this, not only without impairment of normal
function but even to its benefit, is an achievement that has to be learned.
The child up to a certain age is not capable of using this mechanism, or
feels threatened by its attempted use. I think that this is probably one
reason why the child fails vis-a-vis the demand of free association (Hart-
mann, 1939a).
Severe irregularities in the development of autonomy are relevant in
pathology, and some of these problems apparently belong to what B.
Rank (1949) has called the "fragmented ego." We also have ample clinical
evidence of the fact that even with the "normal adult" not all functions
of the ego achieve the same degree of stability.tv
In concluding this paper, may I remind you that in the history of
psychoanalysis modifications of concepts or new formulations of hypo-
theses often followed the opening up of new areas of research, as is the
case in other branches of scientific work. At the present time, the integra-
tion of reconstructive data with data gathered through direct observation
of early childhood represents one of the more pressing demands on our
analytic thinking. My contribution to this symposium was presented with
the aim in mind to facilitate the interrelation of these two sets of data.
Some of the concepts and hypotheses I introduced, like the concept of
9 I decided, maybe somewhat arbitrarily. to omit from my discussion those aspects
of "autonomy" that relate to superego function.
10 It seems that certain phenomena described by J. Lampl-de Groot (1947) are
relevant in this context.
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND ID 27

primary autonomy, or of secondary autonomy, will appear unfamiliar to

many of you. But I found them to be useful tools, especially in dealing
with those developmental problems we are mostly concerned with in this
symposium. Although they were not used by Freud, I think they are con-
sistently developed along the lines of his developmental theories.
When Anna Freud wrote her book The Ego and the Mechanisms of
Defense, she refuted, in her introduction, the opinion, still held by many
analysts at the time, that stigmatized the theoretical study of the ego as
something essentially nonanalytic or even antianalytic, Since then, these
studies have acquired full citizenship in analysis, on an equal level with
the study of the id. There is no reason to assume that the desire to con-
quer "no-man's land" (to use an expression of E. Kris), to extend the
reach of the analytic approach to psychological phenomena beyond its
present limits, has come to an end. In Freud's ego psychology no less than
in other parts of his work we find the kind of truths we would expect
to be rather time-resisting. But there is little doubt that he considered
his outline of ego psychology, monumental as it appears to us, as a begin-
ning rather than as a systematic presentation-in contrast to, let us say,
his psychology of the dream, or of libidinal development; and that he
considered this outline in need, but also capable, of reformulation and
elaboration.U
This symposium was, of course, meant to be a symposium on theory.12 It
deals with developmental and structural propositions. Obviously, it would be
impossible to present to you, on this occasion, also the great variety of individual
clinical material and of data of direct child observation which underlie our
hypotheses. Theory is, in a way, abbreviation; but this is evidently not its only
function. Our hypotheses help us to make of the raw material of our data a
consistent and meaningful body of knowledge; it is only the formulation of
definite propositions that makes our knowledge testable, that is to say, accessible
to verification or falsification, and which gives us the basis for valid predictions;
it also helps us to ask questions which are meaningful and fruitful, and is a use-
11 I should like to emphasize here the very comprehensive character of the concep-
tual framework of Freud's ego psychology-though not all of its aspects and implica-
tions have so far been actually developed. It proves more useful than any other we
know to serve our understanding also of those data of child development which have
previously been found and described by other schools of psychologists. There is no
reason for us not to avail ourselves of relevant observational data found by others
(though their meaning will often be different if seen in our frame of reference) , or
not to study the methods they use. However, I want to emphasize here that it appears
for the most part unnecessary, and often confusing, to borrow, as has sometimes been
done, the conceptual framework and the general theories of development from other
schools of psychology, or simply to superimpose other conceptual systems on that of
Freud.
12This last paragraph is not part of the paper the author presented, but part of
the remarks he made in concluding the discussion.
28 HEINZ HARTMANN

ful tool in directing us toward areas of promising research. All this is well known
and accepted elsewhere, and its importance in analysis is in principle not so
much different from what it is in other branches of science. It would hardly be
worth mentioning here, if it were not for the fact that the function of theory
in analysis has not always been too well understood. One occasionally meets a
tendency to limit psychoanalysis to a clinical specialty, and also a lack of aware-
ness of how much, especially in analysis, the clinical approach owes to the
highly complex structure of hypotheses developed by Freud and others. There
is, with some, the habit of disparaging theory by equating it with "speculation."
Or we hear complaints that the limitless and colorful diversity of individual
clinical experiences is reduced in the process of hypothesis formation-which,
again, disregards the fact that this reduction is one of the most general and most
necessary characteristics of every scientific endeavor. There is no analyst who is
not fully aware of the fundamental relevance of clinical observation in our field.
It is not easy to see why the relevance of theoretical thinking is not likewise
realized and why sometimes the natural emphasis on clinical data turns into a
distrust of theory. The discussion of the role of hypotheses in psychoanlysis would
no doubt deserve a special study. At this point I merely want to restate that
analysis has been from its beginnings, and is most likely to be also in the future,
more comprehensive in its outline, aims, and also means, than its clinical aspect.
I mentioned to you before that one trend in Freud's work through all the
years aimed at a general psychological theory; to exclude it from psychoanalysis
would-mutatis mutandis-be somewhat like excluding physiological theory from
physical medicine. Also, while we know how much Freud's work owes to his
supreme capacity of observation and to his unflinching obectivity vis-a-vis new
facts, we should not forget the extent to which the formation of crucial concepts
and of "good" hypotheses aided his discoveries as well as their meaningful inter-
relation. Actually, for a student of the history of analysis, Freud's work is a
classical example of the point I am trying to make. It appears, from such study,
as a constant mutual promotion of observation and hypothesis formation. We
come to understand how much poorer in dimensions and less fruitful also his
clinical, also his technical work would have been, had his power of theorizing
failed to equal the power of his clinical insight. I do not think that the neces-
sity, not only to enrich our clinical experience but also to develop the body of
hypotheses we use in dealing with it, is less obvious today than it was in Freud's
time.

BIBLIOGRAPHY

Beres, D. and Obers, S. J. 1950, The Effects of Extreme Deprivation in Infancy on

Psychic Structure in Adolescence. This Annual, V.
Bergman, P. and Escalona S. K. 1949, Unusual Sensitivities in Very Young Children.
This Annual, Ill/IV.
Bibring, E. 1937, On the Theory of the Therapeutic Results of Psycho-Analysis. Int.
J. Psa., XVIII.
Brierley, M. 1947, Notes on Psycho-Analysis and Integrative Living. Int. J. Psa..
XXVIII.
 MUTUAL INFLUENCES IN DEVELOPMENT OF EGO AND In 29

Bychowski, G. 1943, Diseases of the Body Image. l- Nero. Ment, Dis., LXLVII.
Durfee, H. and Wolf, K. M. 1933. Anstaltspflege und Entwicklung im ersten Lebens-
jahr. Ztsch. f. Kinderjorschung, 42/43.
Eidelberg, L. 1940, Instinctual Vicissitudes and Defense against Instincts. In Studies in
Psychoanalysis, 2nd ed. New York: International Universities Press, 1952.
Fenichel, O. 1945, Psychoanalytic Theory of Neurosis. New York: W. W. Norton.
Freud, A. 1936, The Ego and the Mechanisms of Defense. New York: International
Universities Press, 1946.
- - 1949, Aggression in Relation to Emotional Development: Normal and Patho-
logical. This Annual, III/IV.
- - 1951, Observations on Child Development. This Annual, VI.
- - in collaboration with Dann, S. 1951, An Experiment in Group Upbringing.
This Annual, VI.
Freud, S. 1924, The Passing of the Oedipus Complex. Collected Papers, II. London:
Hogarth Press, 1924.
- - 1926, The Problem of Anxiety. New York: W. W. Norton, 1936.
- - 1937, Analysis Terminable and Interminable. Collected Papers, V. London: Ho-
garth Press, 1951.
- - 1939, An Outline of Psychoanalysis. New York: W. W. Norton, 1950.
- - 1950, Aus den Anfiingen der Psychoanalyse: Brieje an Wilhelm Fliess, Abhandlun-
gen und Notizen aus den lahren 1887-1902, ed. M. Bonaparte, A. Freud and E.
Kris. London: Imago Publishing Co.
Glover, E. 1935, A Developmental Study of Obsessional Neurosis. Int. ]. Psa., XVI.
Hartmann, H. 1939 a, Ichpsychologie und Anpassungsproblem. Int. Ztsch, Psa., XXIV.
Translated in part in Organization and Pathology of Thought, ed. D. Rapaport.
New York: Columbia University Press, 1951.
- - 1939 b, Psychoanalysis and the Concept of Health. Int. l- Psa., XX.
- - 1948, Comments on the Psychoanalytic Theory of Instinctual Drives. Psa. Quart.,
XVII.
1950 a, Comments on the Psychoanalytic Theory of the Ego. This Annual, V.
- - 1950 b, Psychoanalysis and Developmental Psychology. This Annual, V.
- - 1951, Technical Implications of Ego Psychology. Psa. Quart., XX.
- - Kris, E. and Loewenstein, R. M. 1946, Comments on the Formation of Psychic
Structure. This Annual, II.
Hendrick, I. 1951, Early Development of the Ego: Identification in Infancy. Psa.
Quart., XX.
Hoffer, W. 1949, Mouth, Hand and Ego Integration. This Annual, III/IV.
- - 1950, Development of the Body Ego. This Annual, V.
Isakower, O. 1938, A Contribution to the Psychopathology of Phenomena Associated
with Falling Asleep. Int. [, Psa., XIX.
Jackson, E. B. and Klatskin, E. H. 1950, Rooming-In Research Project: Development
and Methodology of Parent-Child Relationship Study in a Clinical Setting. This
Annual, V.
Karan-Angel, A. 1951, The Role of "Displacement" in Agoraphobia. Int. l- Psa.,
XXXII.
Klein, M. 1930, The Importance of Symbol-Formation in the Development of the Ego.
Int. J. Psa. XI.
Kris, E. 1950 a, On Preconscious Mental Processes. Psa. Quart., XIX; and in Psycho-
analytic Explorations in Art. New York: International Universities Press, 1952.
- - 1950 b, Notes on the Development and on Some Current Problems of Psycho-
analytic Child Psychology. This Annual, V.
- - 1951, Opening Remarks on Psychoanalytic Child Psychology. This Annual, VI.
Lampl-de Groot, J. 1947, Development of the Ego and Superego. Int. ]. Psa., XXVIII.
Menninger, K. A. 1938, Man against Himself. New York: Harcourt, Brace.
80 HEINZ HARTMANN

Murphy, L. B. 1944, Childhood Experience in Relation to Personality Development.

In Personality and the Behavior Disorders, ed. J. MeV. Hunt. New York: Ronald
Press.
Nunberg, H. 1932, Allgemeine Neurosenlehre. Bern: Huber.
Rank, B. 1949, Aggression. This Annual, III/IV.
Rapaport, D. 1950, On the Psychoanalytic Theory of Thinking. Int. J. Psa., XXXI.
- - 1951, Organization and Pathology of Thought. New York: Columbia University
Press.
Ribble, M. 1948, The Rights of Infants. New York: Columbia University Press.
Schilder, P. 1938, The Image and Appea1'ance Of the Human Body. New York: Inter-
national Universities Press, 1950.
Scott, W. C. M. 1948, Some Embryological, Neurological, Psychiatric and Psycho.
analytic Implications of the Body Scheme. Int. l- Psa., XXIX.
Spitz, R. A. 1945, Hospitalism: An Inquiry into the Genesis of Psychiatric Conditions
in Early Childhood. This Annual,!.
- - and Wolf, K. M. 1949, Autoerotism. This Annual, III/IV.
Weiss, P. 1949, The Biological Basis of Adaptation. In Adaptation, ed. J. Romano.
Ithaca: Cornell University Press.