85

IX

The Relation between Symptom
and Anxiety

It now remains to deal with the relationship between
symptom formation and the development of anxiety. Two
opinions about this seem to be prevalent. One of them
terms the anxiety itself a symptom of the neurosis, the
other conceives of a far more intimate connection be-
tween the two. According to this latter view, all symptom
formation would be brought about solely in order to
avoid anxiety; the symptoms bind the psychic energy
which otherwise would be discharged as anxiety, so that
anxiety would be the fundamental phenomenon and the
central problem of neurosis.

The at least partial justification of this second position
can be supplied by means of certain striking examples.
If an agoraphobic who has been accompanied whenever
he went out on the street is left alone there, he produces
an attack of anxiety; if a compulsion neurotic is prevented
from washing his hands after touching something, he be-
comes a prey to almost insupportable anxiety. It is clear,
therefore, that the stipulation of being accompanied and
the compulsion to wash has as their purpose, and also
their result, the averting of an outbreak of anxiety. In
this sense, every inhibition also that the ego imposes on
itself can be termed a symptom.

Since we have reduced the development of anxiety to

86

a response to situations of danger, we shall prefer to
say that the symptoms are created in order to remove or
rescue the ego from the situation of danger. If symptom
formation is prevented, then the danger actually makes
its appearance—that is to say, a situation analogous to
birth comes about, a situation in which the ego finds it-
self helpless against the ever-increasing strength of the
instinctual demand in question; in other words, we have
present the first and earliest of the determinants of anx-
iety. For our point of view the relationships between anx-
iety and symptom prove to be less close than was sup-
posed, the result of our having interposed between the
two the factor of the danger situation. We can also say,
in supplement to this, that the development of anxiety
induces symptom formation—nay more, it is a sine qua
non thereof, for if the ego did not forcibly arouse the
pleasure-pain mechanism through the development of
anxiety, it would not acquire the power to put a stop to
the danger-threatening process elaborated in the id. At
the same time there is an obvious tendency on the part
of the ego to restrict the development of anxiety to a
minimum, to employ anxiety only as a signal, for other-
wise there would merely be experienced somewhere else
the unpleasure threatened by the instinctual process—a
result which would not accord with the purpose of the
pleasure principle, although often enough coming about,
it is true, in the neuroses.

Symptom formation thus has the actual result of put-
ting an end to the danger situation. It has two aspects:
one of them, which remains concealed from us, causes in
the id that alteration by means of which the ego is pre-
served from danger; the other, visible to us, reveals what

87

SYMPTOM AND ANXIETY

it has created in place of the instinctual
modified, namely, substitute formation.

We should express ourselves more accurately, how-
ever, if we ascribed to the process of defense what we
have just said of symptom formation, and used the term
symptom formation itself as synonymous with substitute
formation. It then seems evident that the defensive proc-
ess is analogous to flight, by means of which the ego
avoids a danger threatening from without, and that it
represents, indeed, an attempt at flight from an in-
stinctual danger. The considerations which weigh against
this comparison will themselves prove illuminating. In
the first place, the objection might be raised that object
(loss of the object’s love) and the threat of castration
are just as much dangers threatening from without as is
for example a ravening beast, and are therefore not in-
stinctual dangers. Yet the case is not at all the same. The
wolf would probably attack us, regardless of how we
behaved towards it; but the beloved person would not
withdraw his love, we should not be threatened with
castration, if we did not cherish within ourselves certain
feelings and desires. Thus it is these instinctual impulses
which become the precondition of the external danger,
its conditio sine qua non, and thereby themselves a source
of danger; and we can now combat the external danger
by measures taken against dangers from within. In the
animal phobias the danger seems still to be perceived
entirely as an external one, just as in the symptom also
it undergoes an external displacement. In compulsion
neurosis the danger is to a far greater degree internalized;
that part of the fear of the superego which may be called
social anxiety still represents an internal substitute for

88

an external danger, while the other part, fear of con-
science, is entirely endopsychic.

A second objection would be to the effect that in the
attempt to escape from a threatening external danger all
that we do, in fact, is to increase the distance in space
between us and that which threatens. We do not put our
selves in an attitude of defense against the danger, we
do not try to change anything in the danger itself, as
would be the case if we attacked the wolf with a club or
shot at it with a gun. But the defensive process seems to
go beyond what would correspond merely to an attempt
at flight; it actually interferes with the threatening in-
stinctual process, suppresses it somehow, deflects it from
its aim, and thereby renders it harmless. This objection
appears to be a very cogent one, and one we shall have
to take into account. We believe it may well be that there
are defensive processes which can with justice be com-
pared to an attempt at flight, while in the case of others
the ego offers resistance of a far more active kind, under-
taking vigorous counteractive measures. But it is possible
that the comparison of defense with flight is rendered
untenable by the fact that the ego and the instinctual
drive in the id are in fact parts of the same organization,
and do not have, as the wolf and the child do, a separate
existence from each other; so that every form of behavior
on the part of the ego must have a modifying influence
upon the instinctual process.

Through the study of the situations which occasion
anxiety we have had to envisage with what might be called
rational idealization the behavior of the ego in defense.
Every danger situation corresponds to a given period of
life or stage of development of the psyche, to which it ap-

89

pears appropriate. In early infancy the organism is not
really equipped to cope psychically with large amounts
of excitation reaching it from without or within. At a
certain period of life it is in actual fact to the individual’s
greatest interest that the persons upon whom he is de-
pendent shall not withdraw their tender care. When the
boy perceives the powerful father as his rival for the
mother and becomes aware of his aggressive tendencies
against his father and his sexual desires towards his
mother, he is quite right in being afraid of him; and the
fear of being punished by him may, when reënforced
phylogenetically, be expressed as fear of castration. On
his becoming a social being, fear of the superego, con-
science, becomes a necessity, omission of this step the
source of severe conflicts and dangers. But at this point
a new problem enters.

Let us for the moment try the experiment of substitut-
ing for the affect of anxiety some other affect, for example
that of grief. We consider it entirely normal that a little
girl should weep bitterly at the age of four if her doll is
broken, at the age of six if her teacher reprimands her, at
the age of sixteen if her sweetheart neglects her, at the
age of twenty-five, perhaps, if she buries her child. Each of
these grief-occasioning situations has its proper time and
vanishes with its passing; but the later and more definite
ones remain operative throughout life. We should be
rather surprised, in fact, if this girl, after she had become
a wife and mother, should weep over some knickknack
getting broken. Yet this is how neurotics behave. Al-
though in their mental apparatus there have long since
developed all the agencies necessary for dealing with a
wide range of stimuli, although they are mature enough

90

to be able to gratify the greater part of their needs them-
selves, although they knew perfectly well that castration
is no longer practised as a punishment, they nevertheless
behave as though the old danger situation still existed,
they remain under the spell of all the old causes of anx-
iety.

The answer to all this will prove somewhat prolix, for
it will have in the first place to sift the actual facts of the
case. In a large number of instances the old causes of
anxiety have in reality become inoperative, but only after
having first brought neurotic reactions into existence.
The morbid fear of being alone, of the dark, and of
strangers, on the part of the smallest children, which is
almost to be labelled normal, disappears for the most
part at a somewhat later age; such fears are “outgrown,”
as we say of many other disturbances of childhood. The
phobias of animals so frequently met with share this
same fate; many of the conversion hysterias of childhood
are not carried over into later life. In the latency period
the practice of ceremonials is of extremely frequent oc-
currence, yet only a very small percentage of these cases
later develop a full-blown compulsion neurosis. The neu-
roses of children, so far at least as concerns upper-class
urban children of the white race, are regularly occurring
episodes in development, although too little attention is
still paid to them. In not a single adult neurotic do the
indications of a childhood neurosis fail of occurrence,
while on the other hand by no means all children who
show them become neurotic subsequently. Therefore in
the course of growing up the anxiety-determinants which
once existed must have been relinquished, the situations
originally endowed with danger have lost their signifi-
cance.

91

cance. To this must be added that certain of these danger
situations survive into a later period of life by means
of a modification, in keeping with that later period, of
the character of what gives rise to anxiety. Thus, for
example, castration anxiety persists in the guise of syphi-
lophobia, after it has been learned that castration is no
longer customary as a punishment for giving the sexual
appetites free rein but that serious diseases threaten in-
stinctual freedom instead. Certain other of the things
that occasion anxiety are destined not to disappear at
all, but are to accompany the human being throughout
life, such as, for example, the fear of the superego. The
neurotic is then distinguished from the normal person
in that his response to these dangers is disproportionately
increased. Yet against the return of the original traumatic
anxiety situation even maturity offers after all no ade-
quate protection; there may exist for every one a limit
beyond which his psyche fails in the attempt to cope with
the demands which the excitation in question makes
upon him.

These minor reservations cannot possibly be taken as
militating against the fact which we have been discussing:
the fact, namely, that in their response to danger so many
people remain infantile, continuing to react with anxiety
to situations which should have long ceased to evoke it;
to dispute this would be to deny the very fact of neurosis,
for it is exactly such persons whom we call neurotics.
But how does this situation come about? Why are not all
neuroses merely episodes in the individual’s development
which become a closed chapter when the next stage of
development is reached? Whence comes the element of
permanency in these reactions to danger? Whence springs

92

THE PROBLEM OF ANXIETY

the preference over all other affects which the affect of
anxiety seems to enjoy in alone evoking reactions which
we distinguish from others as abnormal and which in
their inexpediency obstruct the stream of life? In other
words, we find ourselves abruptly confronted once again
by the oft-repeated riddle: What is the source of neurosis,
what is its ultimate, its specific, underlying principle?
After decades of analytic effort this problem rises up be
fore us, as untouched as at the beginning.