5

IV

The first case which we shall consider is that of an infantile
hysterical zoöphobia, taking as an example a case, typical in
all essential respects, of a phobia of horses, the case of “little
Hans”¹⁾. The very first impression we derive makes it evident
that the conditions which obtain in an actual case of neurotic
illness are far more complicated than we should be led to
expect as long as we work with abstractions only. It is some-
thing of a task to get one’s bearings with regard to what the
repressed impulse is and what its substitute in the form of
symptoms, and wherein the motive for repression becomes
recognizable.

Little Hans refused to go out on the street because he was
afraid of horses. This is the raw material of the case. Now,
what is it in this which is the symptom: the development of
anxiety, or the choice of the object of anxiety, or the renounc-
ing of freedom of movement, or several of these combined?
What is the gratification which he renounces? Why does he
have to renounce it?

It would be easy enough to reply that there is nothing so
very puzzling about this case. The unintelligible fear of
horses is the symptom; the inability to go out on the street is
an inhibition, a restriction, which the ego imposes on itself in
order not to arouse anxiety. The correctness of the explana-
tion of this last point is sufficiently evident, and this inhibition
may therefore be left out of account in our subsequent discus-
sion. But this cursory initial acquaintance with the case does
not even give us the true text of the manner in which the
alleged symptom is expressed. For as we learn on closer
inquiry, it is not at all a question of an indefinite fear of horses,
but of a definite anxious expectation–namely, that of being
bitten by a horse. This content, it is true, seeks to avoid
becoming conscious and to be replaced by the indefinite phobia

¹⁾See The Analysis of a Phobia in a Five-Year-Old Boy (Collected Papers III).

6

in which only the anxiety and its object still remain. Is this
content perhaps the nucleus of the symptom?

We shall make no further progress so long as we fail to take
into account the entire psychic situation of the little patient as
this is revealed to us during analysis. He finds himself in the
jealous and hostile œdipus attitude to his father, whom, how-
ever, in so far as his mother does not enter into the picture as
the cause of dissension, he loves devotedly. Thus we have a
conflict springing from ambivalence–a firmly founded love and
a not less justified hatred, both directed to the same person.
His phobia must be an attempt to resolve this conflict. Such
conflicts due to ambivalence are very common; we are
acquainted with another typical outcome of them in which one
of the two contending trends, usually the tender one, becomes
enormously augmented, while the other disappears. Only the
excessiveness of the tenderness and its compulsive character
betray the fact that this attitude is not the only one present, and
that it is ever on its guard to keep the contrary attitude sup-
pressed, making it possible to construe a train of events which
we describe as repression through reaction formation (in the
ego). Cases like that of little Hans show nothing of such
reaction formation; evidently there are various paths of escape
from an ambivalency conflict.

Meanwhile there is something else which we have perceived
with certainty. The instinctual impulse which succumbs to
repression is a hostile impulse against the father. The analysis
furnished the evidence of this in tracing the origin of the idea
of the biting horse. Hans saw a horse fall down; he saw too a
playmate with whom he had played “horsie” fall and hurt
himself. The analytic material has given us the right to con-
strue a wish on Hans’s part to the effect that his father should
fall and hurt himself as had the horse and Hans’s playmate.
References of his to a departure which he saw his father make
lead us to suppose that the wish for his father’s destruction has
also found less timid expression. Such a wish, however, is
equivalent to the intention of doing away with him himself,
to the murderous impulse of the œdipus complex.

7

From this repressed instinctual impulse there exists as yet
no pathway to the substitute for it which we assume the phobia
of horses to be. Let us now simplify little Hans’s emotional
situation by leaving out the infantile factor and the ambiva-
lence; let us call him a younger servant in a household who is
in love with the mistress of the house and enjoys certain favors
at her hands. We will retain too the element that he hates the
master of the house, who is of course stronger than he, and
wishes he had some means of getting him out of the way; the
most natural upshot of this situation then is that he would fear
revenge or retaliation on the part of this master, that a state
of anxiety in relation to him would be engendered–quite like
little Hans’s phobia of horses. But this means that we cannot
term his phobic anxiety a symptom; if little Hans, in love with
his mother, were to betray fear of his father, we would have no
right to ascribe a neurosis, a phobia, to him. We should,
rather, be confronted merely with a wholly intelligible affective
reaction. That which makes this affective reaction into a
neurosis is singly and solely an altogether different feature,
namely, the substituting of the horse for the father. It is this
displacement, therefore, which produces something deserving
of the term symptom; and this is that other mechanism we have
spoken of which permits the resolving of the ambivalency con-
flict without recourse to reaction formation. This displace-
ment, moreover, is rendered possible or facilitated by the
circumstance that inborn traces of a totemistic mode of thinking
can still be easily activated at this tender age; the gulf between
man and animal is not yet recognized, certainly not so over-
emphasized as later. The adult male, admired but also feared,
still belongs in the same category with large animals, which
one envies for many things but against which one has also been
warned because they can be dangerous. Thus the conflict due
to ambivalence is not worked out in the same person, but is,
so to speak, detoured, in that one of its component impulses is
foisted upon another person as substitute.

Thus far matters are clear enough, but on another score the
analysis of little Hans’s phobia has been highly disappointing.

8

The distortion of which the symptom consists is carried out
not at all upon the representative (the ideational content) of
the instinctual impulse to be repressed, but upon an entirely
different one which corresponds only to a reaction against what
is objectionable per se. Our expectation would rather have
been that little Hans would have developed, instead of his fear
of horses, a tendency to maltreat them, to beat them, or would
have given clear expression to his wish to see them fall down
and injure themselves, possibly to die in convulsions (the
“making a row with the feet”¹⁾. Something of the kind really
occurred during his analysis, but it is not at all conspicuous in
his neurosis, and—curiously enough—if he had developed such
hostility as a central symptom, directed against horses only
instead of against his father, we should certainly not have
judged that he had a neurosis. There is something wrong
here somewhere, either in our conception of repression or in
our definition of a symptom. One thing of course strikes us
immediately: If little Hans had really manifested such an atti-
tude toward horses, then repression would not have altered in
the least the character of the objectionable aggressive instinc-
tual impulse but only transformed its object.

It is quite certain that there are cases of repression which do
not do more than this. In the genesis of little Hans’s phobia,
however, more has happened; how much more, we may dis-
cover from another analytic fragment.

We already know that little Hans gave as the content of his
phobia the idea of being bitten by a horse. Now we have since
obtained insight into the genesis of another case of zoöphobia
wherein it was a wolf which gave rise to anxiety but which
likewise represented a father substitute². In connection with
a dream which the analysis was able to elucidate, this boy
developed a fear of being eaten by a wolf, like one of the seven
little goats in the fairy story. The authenticated fact that little
Hans’s father had played “horsie” with him was certainly a
determining factor in his choice of the animal causing anxiety;

¹⁾Loc. cit., page 193.—TR. NOTE.

²From the History of an Infantile Neurosis (Collected Papers III).

9

in the same way, it is at least highly probable that the father of
this second patient, a Russian analyzed in his twenties, had
mimicked a wolf in playing with him as a youngster and had
jokingly threatened to eat him up. I have since come upon a
third case, that of a young American, who did not develop any
animal phobia, it is true, yet in whom its very absence helped
to an understanding of the other cases. His sexual excitement
had been kindled by a fantastic nursery tale which was read to
him of an Arab sheik who pursued a Gingerbread Man in order
to devour him. With this edible being he identified himself,
the sheik was readily recognizable as a father substitute, and
this fantasy became the first substratum of his autoerotic
activity. The idea of being eaten by the father belongs to the
typical primal stock of childhood ideas; analogies from mythol-
ogy (Kronos) and from animal life are generally familiar.

Despite such assistance, these ideas are so foreign to us that
it is only with incredulity that we are able to concede them to
the child. We also do not know whether they really mean
what they seem to express, and do not understand how they
can become the subject of a phobia. Analytic experience, of
course, gives us the required information. It tells us that the
idea of being eaten by the father is the regressively debased
expression of a tender passive impulse which craves to be the
object of the father’s love in the sense of genital erotism. The
further history of the case permits of no doubt as to the correct-
ness of this interpretation. The genital impulse, it is true, no
longer betrays anything of its tender intent when it is expressed
in the language of the stage of transition from the oral to the
sadistic phase of libido organization, a stage long since left
behind. Is it, moreover, a question only of a substitution of
the instinct representative by a regressive mode of expression,
or is it, rather, a genuine regressive debasement of the genitally
directed impulse in the id? This seems to be a matter by no
means easy to decide. The clinical history of the Russian
“wolf man” points quite definitely to the latter more serious
possibility, for from the time of the crucial dream onward he
behaved “badly”, cruelly and sadistically, and developed shortly

10

afterwards an outright compulsion neurosis. At all events we
learn that repression is not the only means at the command of
the ego whereby it defends itself against an unwelcome instinc-
tual drive. When it succeeds in bringing about the regression
of the impulse, it has inflicted more radical damage upon the
latter, after all, than would be possible through repression.
Occasionally, to be sure, repression follows upon the repression
which the ego has first compelled.
The situation in the case of the “wolf man” and the some-
what simpler one in that of little Hans suggest a number of
further considerations, but two unexpected pieces of insight
we obtain at once. There can be no doubt that the repressed
instinctual impulse which these phobias represent is a hostile
one directed against the father. One may say that this impulse
is repressed by a process of transformation into its opposite; in
place of aggression against the father, there appears the father’s
aggression–retaliation–against the individual himself. Since
such aggression is rooted in the sadistic stage of libido develop-
ment in any case, there is only necessary a certain degree of
degradation thereof to the oral level, such as is intimated in
the case of Hans in the being bitten, but in the Russian is car-
ried out in forthright fashion in the being eaten. But in
addition analysis makes it possible to establish beyond any
doubt that simultaneously still another instinctual impulse has
succumbed to repression, one of opposite signification, namely,
a tender passive impulse towards the father which had already
reached the threshold of the genital (phallic) stage of libido
organization. This latter seems even to be the more signifi-
cant as regards the end result of the repressive process; it under-
goes the more extensive regression; it obtains the crucial
influence upon the content of the phobia. Where therefore
we have traced the repression of but a single impulse, we have
now to recognize the conjunction of two such processes; the two
instinctual impulses concerned–sadistic aggression against the
father, and a tender passive attitude to him–form a contrasting
pair; nay, more: if we evaluate the history of little Hans cor-
rectly, we recognize that in addition the tender object-cathexis

11

of his mother has been abolished through the creation of his
phobia, although of this the content of the phobia betrays no
suggestion. We are dealing in the case of Hans–it is much
less definite in the case of the Russian–with a repression which
involves almost every component of the œdipus complex–the
hostile as well as the tender impulse towards the father, and the
tender towards the mother.

These are quite unwelcome complications for us whose
desire it was to study only simple cases of symptom formation
resulting from repression, and who had turned for this purpose
to the simplest and seemingly most transparent neuroses, those
of childhood. Instead of a single repression we encountered a
host of them, and in addition to this we had to deal with regres-
sion as well. Perhaps we have only increased the confusion by
attempting to fit both our analyses of zoöphobia, of little Hans
and of the “wolf man”, into the same Procrustean bed. For
certain differences between the two now strike us. It is only
of little Hans that we may declare with assurance that through
his phobia he overcomes the two major impulses of the œdipus
complex, the aggressive impulse towards the father and the
over-tender impulse towards the mother; tender feelings
towards his father are certainly present also and play their part
in the repression of their opposite, but it cannot be shown
either that they were strong enough to provoke repression or
that they were abolished subsequently. Hans seems to have
been quite a normal youngster, with a so-called positive œdipus
complex. It is possible that the elements which we fail to find
were also co-active in him, but we cannot demonstrate them;
the material even of our most thorough analyses is simply
deficient, our documentation incomplete. In the case of the
Russian the lacuna occurs at another point; his relation to the
female object was disturbed by a premature seduction, the
passive feminine component was strongly developed in him,
and the analysis of his wolf dream reveals little of intended
aggression against his father—all these provide the most
unequivocal proof that the repression has to do with the passive
tender attitude to the father. Here again it may be that other

12

factors played a part, but they are not in evidence. If, despite
these differences between the two cases, which are almost of the
nature of a complete antithesis, the end result of the phobia is
nearly the same, the explanation of this fact must come from
another source; it comes from the second result of our little
comparative study. We believe that we know the motive force
behind the repression in both cases and see its rôle substanti-
ated by the course which the development of the two children
takes. This motive force is in both cases the same—namely,
fear of a threatened castration. It is because of castration
anxiety that little Hans renounces aggression against his father;
his anxiety lest a horse should bite him can be readily supple-
mented by saying that his anxiety is lest a horse should bite off
his genital, castrate him. But it is from castration anxiety also
that the little Russian renounces the wish to be loved by his
father as a sexual object, for he has understood that such a rela-
tionship would have as its prerequisite that he sacrifice his
genital—that which distinguishes him from the female. Both
forms of the œdipus complex—the normal, active, as well as the
inverted—founder on the castration complex. The Russian’s
fear of being eaten by a wolf contains no suggestion of castra-
tion, it is true; through oral regression the idea has been
removed too far from the phallic stage; but the analysis of his
dream makes any other proof superfluous. It is furthermore
a complete triumph of repression that in the wording of the
phobia there is no longer anything which hints at castration.
It is in this, then, that our unexpected result consists: In
both cases the motor force behind the repression is castration
anxiety; and the content of the anxiety—being bitten by a
horse, or eaten by a wolf—is in each case a distortion of and
substitute for another content, that of being castrated by the
father. It is precisely this content which has undergone repres-
sion as such. In the Russian the content was the expression of
a wish which could not maintain itself against the revolt of
masculinity, in Hans the expression of a reaction which con-
verted aggression into its opposite. But the anxiety affect of
the phobia which constitutes its essence does not arise from the

13

process of repression nor out of the libidinal cathexes of the
repressed impulses, but from the repressing forces themselves;
the anxiety of zoöphobia is transformed castration anxiety,
therefore a real anxiety, a reality fear, fear of a danger actually
threatening or believed to do so. Here it is the anxiety that
causes the repression, and not, as I earlier stated, the repression
the anxiety.

It is not pleasant to think of it, but there is no use in denying
that I have repeatedly put forward the thesis that through
repression the instinctual representative is distorted, displaced
and the like, and the libido of the instinctual impulse trans-
formed into anxiety. Study of the phobias, which should be
particularly well calculated to furnish proof of this thesis, thus
fails to confirm it; rather, it seems completely to contradict it.

The fear in zoöphobia is castration anxiety on the part of the
ego, that in agoraphobia (though this has been less thoroughly
studied) seems to be a fear of temptation, which must certainly
be related genetically to castration anxiety. The majority of
phobias, so far as we can see at present, are traceable to such a
fear on the ego’s part of the demands of the libido. Always in
this situation it is the attitude of anxiety on the part of the ego
which is the motive of and the incitement to repression. Never
does the anxiety emanate from the repressed libido. If I had
been content on earlier occasions simply to say that following
upon repression a certain amount of anxiety appears in place of
the expression of libido that would be expected, I should not
have to retract anything now. The description is correct, and
there does exist the correspondence between the strength of the
impulse to be repressed and the intensity of the anxiety result-
ing which I asserted. But I confess I thought I was giving more
than a mere description; I supposed that I had recognized the
metapsychological process of a direct transformation of libido
into anxiety; this I can no longer maintain today. I was
unable, besides, to give any account at that earlier time of how
such a transformation was accomplished.

Whence did I derive at all the idea of this transformation?
At a time when we were still a long way from distinguishing

14

between processes in the ego and processes in the id, from a
study of the “actual” neuroses. I found that certain sexual
practices, such as coitus interruptus, frustrated excitement,
enforced abstinence, give rise to outbreaks of anxiety and a
general predisposition to anxiety–which may be induced
whenever, therefore, sexual excitation is inhibited, frustrated
or diverted in the course of its discharge in gratification. Since
sexual excitement is the expression of libidinal instinctual
impulses, it did not seem rash to suppose that through the
influence of such disturbances the libido became converted
into anxiety. Now this observation still holds good today; on
the other hand, it cannot be denied that the libido of processes
in the id is subjected to disturbance by the provocative influ-
ence of repression; thus it may still be correct to say that in
repression anxiety is created out of the libidinal cathexis of
instinctual impulses. But how is one to bring this finding into
accord with the other, namely, that the anxiety of a phobia is
an ego anxiety, originates in the ego, and does not result from
repression but on the contrary evokes it? This seems a contra-
diction, and one not simple to resolve. To reduce the two
sources of anxiety to a single one is not easy of accomplish-
ment. One may attempt it by supposing that in the situation
of interrupted coitus, of intermitted excitation, of abstinence,
the ego scents danger to which it reacts with anxiety; but this
does not get us very far forward. On the other hand, the
analysis of phobias which we have undertaken does not seem
to be open to correction. Non liquet!