15

V

It was our desire to study the way in which symptoms are
created, and the secondary struggle of the ego against them,
but in selecting phobias for this purpose we have evidently not
made a very fortunate choice. Anxiety, which is the predomi-
nating characteristic of these disorders, now appears as a com-
plication which conceals the true state of affairs. There is a
plenitude of neuroses in which no anxiety is manifested. True
conversion hysteria belongs in this category; its most severe
symptoms may be free from any admixture of anxiety. This
fact alone ought to serve as a warning to us not to connect
anxiety with symptom formation too rigidly. The phobias
are in other respects so closely related to conversion hysteria
that I have considered it justifiable to align the former with
the latter as “anxiety hysteria”. But no one has yet been able
to formulate what it really is that determines whether a case
will assume the form of a conversion hysteria or of a phobia;
in other words, no one has established what constitutes the
prerequisite for the development of anxiety in hysteria.

The symptoms most frequent in conversion hysteria–a motor
paralysis, a contracture, or an involuntary movement or motor
discharge; a pain; an hallucination–are cathexes either per-
manently sustained or intermittent; but to say this only intro-
duces new difficulties into their explanation. One really does
not know much that can be said about these symptoms. Through
analysis one may learn for what particular excitation whose dis-
charge is interfered with they are the substitute; in most instances
we find that the symptoms themselves participate in this dis-
charge of excitation, as if the total energy of the excitation had
been concentrated upon this one fraction. Pain was present in
the situation in which repression took place; hallucinations were
the perception of, motor paralysis the defense against, an action
which should have been carried out in that situation but was
inhibited; contracture is usually a displacement elsewhere of a
muscle innervation intended at the time in question; the con-
vulsive seizure the expression of an outbreak of affect which has

16

escaped from the normal control of the ego. The sensation of
unpleasure (Unlust) which accompanies the appearance of
symptoms varies to an extraordinary degree. In the case of
the permanent symptoms where a displacement upon motility
has occurred, such as paralyses and contractures, it is usually
absent; the ego behaves towards them as if it were not involved;
in the case of the intermittent symptoms and those in the
sensory sphere, definite feelings of unpleasure are experienced
as a rule, which may be increased to an excessive degree in the
case of the symptom of pain. It is extremely difficult to isolate
out of this complexity the factor which makes such differences
possible and which would permit of their explanation on a
single all-inclusive basis. Furthermore, there is little to note
in conversion hysteria regarding the struggle of the ego against
the symptom, once the latter has been created. It is only when
the sensibility to pain on the part of some portion of the body
has become a symptom that the part of the body in question is
placed in the position of playing a double rôle. The painful
symptom makes its appearance just as certainly if this part of
the body is stimulated from without as when the pathogenic
situation which the symptom represents is associatively acti-
vated from within, and the ego adopts precautionary measures
to forestall the activation of the symptom through external per-
ception. Whence arises the particular incomprehensibility
which characterizes symptom formation in conversion hysteria
we are unable to guess, but it supplies a reason for quickly
leaving this unfruitful field behind.

We turn to the compulsion neurosis in the expectation that
we may learn more in this disorder of the way in which symp-
toms are created. The symptoms of compulsion neurosis are,
broadly speaking, of two kinds and of contradictory purport.
They are either prohibitions or prophylactic measures or atone-
ments, and are thus of a negative character, or they are on the
other hand substitute gratifications, very often in symbolic guise.
Of these two groups the negative, defensive, punitive variety is
the earlier; with the progress of the illness, however, the grati-
fications, proof against all measures of defense, gain the upper

17

hand. It represents a triumph of symptom creation when this
process succeeds in amalgamating prohibition with gratifica-
tion, so that the originally defensive command or prohibition
acquires the significance of a gratification, for which purpose
highly artificial associative links are often utilized. In this
performance is manifest that tendency to synthesis which we
have already attributed to the ego. In extreme cases the
patient succeeds in bringing it about that the majority of his
symptoms acquire a meaning the exact opposite of that which
they originally possessed—a testimonial to the power of that
ambivalence which plays so large a rôle, though we do not know
why, in compulsion neurosis. In the extremest case the symp-
tom is dichronous; that is to say, upon a piece of behavior
which carries out a given prescription there follows imme-
diately a second one which abolishes or nullifies it, even though
not yet venturing to institute its opposite.

Two impressions emerge immediately from this cursory sur-
vey of compulsive symptoms. The first is that in these symp-
toms a continuous struggle against the repressed is being
maintained, in which the tide of battle turns increasingly
against the repressing forces; the second, that ego and superego
here participate to a particularly large degree in the formation
of symptoms.

Compulsion neurosis is indeed the most interesting and most
grateful subject of analytic investigation, but a problem still
unsolved. If we wish to penetrate more deeply into its nature,
we shall have to confess that dubious assumptions and unproven
conjectures cannot yet be dispensed with. The situation of
origin in the case of compulsion neurosis is in fact none other
than in hysteria, namely, the defense necessary against the
libidinal demands of the œdipus complex. Furthermore there
seems to be present in every compulsion neurosis a lowermost
layer of hysterical symptoms of very early formation. Its sub-
sequent form, however, is then decisively altered by a consti-
tutional factor. The genital organization of the libido mani-
fests itself as too weakly and too little resistant. When the ego
undertakes the measures of defense just referred to, the first

18

result it achieves is that the genital organization (of the phallic
phase) is wholly or partially thrown back to the earlier anal-
sadistic stage. This fact of regression is crucial for all that
follows.

One may take still another possibility into consideration. Perhaps the regression is not the consequence of a constitu-
tional but of a time factor. It is made possible, not because
the genital organization of the libido has proved too feeble,
but because the resistance of the ego has been initiated too
early, during the height of the sadistic phase. I do not venture
a definite decision on this point, but analytic observation does
not favor the latter supposition. It indicates, rather, that the
phallic stage has already been reached when the deviation to
compulsion neurosis occurs. The age period, furthermore, at
which this neurosis has its onset is a later one than is the case
with hysteria (the second period of childhood, after the latency
period); in one case which I had the opportunity to study, in
which this disorder developed very late, it was clear that a dis-
turbance in actual life of his genitahty, which had been intact
until then, supplied the situation prerequisite for regression
and for the genesis of the compulsion neurosis.¹⁾

The metapsychological explanation of regression I have
thought to find in an “instinct defusion”, in the segregation of
the erotic components which with the onset of the genital phase
were joined to the destructive cathexes of the sadistic stage.

The enforcing of regression constitutes the ego’s first success
in its struggle of defense against the demands of the libido. It
is convenient to distinguish here the general tendency to
“defense” from “repression”, which is only one of the mecha-
nisms utilized in the service of the defense. Perhaps in com-
pulsion neurosis, even more clearly than in normal and hysteri-
cal cases, we may recognize the castration complex as the
motivating force of defense, the strivings of the œdipus com-
plex as that which is defended against. We now find ourselves
at the beginning of the latency period, which is characterized
by the breaking down of the œdipus complex, the creating or

¹⁾See The Predisposition to Obsessional Neurosis. Collected Papers II.

19

consolidating of the superego, and the erecting of ethical and
æsthetic barriers in the ego. In compulsion neurosis these
phenomena exceed the normal degree; to the destruction of the
œdipus complex is joined the regressive degradation of the
libido, the superego becomes particularly strict and hard-
hearted, and in obedience to the superego the ego develops
intense reaction formations of conscientiousness, pity, cleanli-
ness. With inexorable and hence not always successful severity
the temptation to continue the masturbation of early infancy
is tabooed–a temptation now based upon regressive (anal-
sadistic) ideas, but at the same time representing the uncon-
trolled part of the phallic organization. An inner contradic-
tion consists in the fact that it is precisely in the interest of the
maintenance of masculinity (castration anxiety) that every
manifestation of this masculinity is prevented, but in compul-
sion neurosis this contradiction also is only exaggerated, merely
an intensification of the normal manner of overcoming the
œdipus complex. Just as every excess carries within itself the
seeds of its own surcease, so this holds true too of compulsion
neurosis, in that it is precisely the suppressed onanism which
in the form of compulsive acts compels an ever furthered
approximation to gratification.
The reaction formations in the ego of the compulsion neu-
rotic which we recognize as exaggerations of normal character
traits we may put down as a new mechanism of defense, along
with regression and repression. They seem to be lacking, or
to be much weaker, in hysteria. In retrospect we thus arrive
at a conjecture as to what it is that distinguishes the defense
process in hysteria. It appears that this is limited to repression,
in that the ego turns away from the objectionable instinctual
impulse, leaves it to work out its destiny in the unconscious,
and takes no further interest in its fate. This cannot be cor-
rect to the last detail, for certainly we know cases in which the
hysterical symptom signifies at the same time the fulfilment of
a demand for punishment on the part of the superego, but it
will describe a general characteristic of the behavior of the ego
in hysteria.

20

It may simply be taken as a fact that in compulsion neurosis
so strict a superego is formed, or one may give consideration to
the fact that the fundamental characteristic of this disorder is
regression of the libido, and attempt to connect with this the
character of the superego. As a matter of fact, the superego,
which derives from the id, is quite unable to evade the regres-
sion and instinct defusion which take place there. It can
hardly be surprising if the superego, on its part, should be
more rigid, more sadistic, more hardhearted than in normal
development.

During the latency period the defense against the temptation
to onanism seems to be regarded as the principal task. This
struggle gives rise to a number of symptoms which recur in
typical fashion in the most different individuals and which have
in general the character of a ceremonial. It is greatly to be
deplored that these have not thus far been collected and
systematically analyzed; as the earliest manifestations of neu-
rosis, they would most easily throw light upon the mechanism
of symptom formation here employed. They already exhibit
the characteristics which will become manifest so ominously
later on in a serious illness: such as their application to acts
which should later be carried out as if they were automatic, as
for example going to bed, washing and dressing, and locomo-
tion; and such as also the tendency to repetition and to pro-
crastination. Why this should happen, we do not as yet under-
stand in the least; the sublimation of anal-erotic components
plays a definite rôle therein.

Puberty forms a definite epoch in the development of the
compulsion neurosis. The genital organization, interrupted
in childhood, is now resumed with great vigor. But we
know that the sexual development of childhood also prescribes
the direction taken by this pubertal recrudescence. Thus
on the one hand the aggressive impulses of childhood will
reawaken; on the other, a greater or smaller proportion of the
fresh libidinal impulses–in severe cases, their totality–must
take the road laid down in advance by regression and appear

21

as aggressive and destructive designs. In consequence of this
disguising of the erotic strivings and of the intense reaction
formations in the ego, the struggle against sexuality is now
continued under an ethical ægis. The ego rebels in bewilder-
ment against the cruel and outrageous demands which are sent
out from the id into consciousness, little suspecting that it is
combating erotic wishes in so doing–among them also such as
would otherwise escape its protest. The overstrict superego
insists the more actively on the suppression of sexuality since
the latter has assumed such obnoxious forms. Thus in com-
pulsion neurosis the conflict is seen to be intensified in two
directions: the defending forces have become too intolerant,
that against which the defense is instituted too intolerable–
and both of these through the influence of the single factor of
regression of the libido.

One might find a contradiction of a number of our assump-
tions in the fact that the unwelcome obsession is conscious at
all. What is alone indubitable is that it has previously under-
gone repression. In general, the specific terms of the aggres-
sive instinctual impulse are not at all known to the ego. A
good deal of analytic work is necessary to bring them to con-
sciousness. What has penetrated into consciousness is as a rule
only a distorted substitute, rendered unrecognizable either by
a hazy and dreamlike vagueness or through an absurd disguise.
If repression has not left its impress upon the content of the
aggressive impulse, it has at least done away with the affect
accompanying it. Thus it is that the aggression does not
appear to the ego as an impulse but rather, as patients put it,
as a pure “thought content” which should leave one cold. The
remarkable thing is that this is not in fact the case.

The affect not shown on perceiving the compulsive idea
does make its appearance, but elsewhere. The superego
behaves as if no repression had taken place, as if it were
fully cognizant of the aggressive impulse in its true meaning
and with its affective character undiminished; and it treats
the ego on the basis of this hypothesis. The ego which knows

22

on the one hand that it is innocent has on the other to experi-
ence a feeling of guilt and to carry a feeling of responsibility
which it is unable to account for. The riddle which this
presents for our solution is not so difficult, however, as at first
sight it appears. The attitude of the superego is entirely intel-
ligible; the contradiction in the ego shows us only that by the
medium of repression it has shut itself off from the id while
remaining wholly accessible to influence from the superego.¹⁾
The further question why the ego does not also seek to throw
off the harassing criticism of the superego is settled by the infor-
mation that this actually happens in a large number of cases.
There are also cases of compulsion neurosis quite without a
sense of guilt; to the best of our present understanding, the ego
has saved itself from the perception of guilt by means of a new
series of self-punitive symptoms, atonements, and restrictions.
These symptoms signify at the same time, however, gratifica-
tions of masochistic instinctual impulses which likewise derive
reënforcement from regression.

So enormous is the multiplicity of the manifestations which
compulsion neuroses exhibit that despite all effort no one has
yet succeeded in providing a comprehensive synthesis of all
their variations. One is tempted to lay stress upon typical
relationships, although in doing so there is the ever present fear
of overlooking others no less important.

I have already described the general trend of symptom for-
mation in compulsion neurosis. It consists in giving more and
more room to substitute gratification at the expense of renun-
ciation. The same symptoms which originally had the signifi-
cation of limitations imposed upon the ego later assume also,
thanks to the ego’s synthesizing tendency, that of gratification,
and it is evident enough that the latter significance gradually
becomes the more potent one. An ego upon which extreme
restrictions are imposed and which on this account is reduced
to seeking gratification in symptoms is the result of this process,
which increasingly approximates to a complete miscarriage of

¹⁾Cf. Reik: Geständniszwang und Strafbedürfnis, 1925, p. 51.

23

the original effort at defense. The shifting of the balance of
power in favor of gratification may lead to the ominous end-
result of a paralysis of the will on the part of the ego, which in
every decision finds almost as strong an impulsion arrayed on
one side of the question as on the other. The excessively bitter
conflict between id and superego which dominates the disorder
from the beginning may become so extended that none of the
activities of the ego, incapable as it is of mediation, can escape
involvement in this conflict.