What I call 'anxiety-neurosis' manifests itself in a partial and in a complete form, and may be met with either as an isolated state or combined with other neuroses. The cases which are more or less complete and at the same time isolated are naturally those which in particular give support to the idea that anxiety-neurosis is a clinical entity. In other cases where there is a complex of symptoms corresponding to a mixed neurosis, we have to distinguish and separate from it those symptoms which belong neither to neurasthenia nor to hysteria and so on, but to anxiety-neurosis.

The clinical picture of anxiety-neurosis comprises the following symptoms:

1. _General irritability_. This is a common nervous symptom and as such belongs to many nervous conditions. I include it here because it invariably appears with anxiety-neurosis and is important theoretically. An increase of irritability always signifies an accumulation of excitation or an inability to tolerate such an accumulation, that is, an absolute or a relative accumulation of excitation. One form of expression of this increase of irritability---auditory hyperæsthesia---seems to me worthy of special mention; this undue sensitiveness to noise is undoubtedly explicable on the basis of the close inborn connection between auditory impressions and fright. Auditory hyperæsthesia is frequently a cause of sleeplessness, more than one form of which belongs to the {79} anxiety-neurosis.

2. _Anxious expectation_. I cannot better describe the condition I have in mind than by this name and by appending a few examples. A woman who suffers from anxious expectation will imagine every time her husband coughs, when he has a cold, that he is going to have influenzal pneumonia, and will at once see his funeral in her mind's eye. If when she is coming towards the house she sees two people standing by her front door, she cannot avoid the thought that one of her children has fallen out of the window; if the bell rings, then someone is bringing news of a death, and so on; whereas on all these occasions there is no particular ground for exaggerating a mere possibility.

Anxious expectation of course fades off imperceptibly into normal anxiousness. It comprises all that is covered by the word 'nervousness'---apprehensiveness, the tendency to look on the dark side of things; but at every opportunity it exceeds the limits of this plausible form of nervousness and is frequently recognized by the patient himself as a kind of compulsion. For one form of anxious expectation---that relating to one's own health---we may reserve the old term _hypochondria_. Hypochondria does not always coincide with a high degree of anxious expectation; in general it requires as a preliminary condition the presence of paræsthesias and disagreeable bodily sensations, and therefore hypochondria is the form favoured by true neurasthenics when they fall victims to anxiety-neurosis, as they often do.

The tendency to _pangs of conscience_, scrupulousness and pedantry may be a further expression of anxious expectation, a tendency which is especially frequent among morally sensitive people and likewise ranges {80} from the normal to an exaggeration known as _folie du doute_.

Anxious expectation is the nuclear symptom of this neurosis; it clearly reveals, too, something of the theory of it. We may perhaps say that there is here _a quantum of anxiety in a free-floating condition_, which in any state of expectation controls the selection of ideas, and is ever ready to attach itself to any suitable ideational content.

3. This is not the only way in which apprehensiveness---which is not usually present in consciousness but is ever lying in wait---can express itself. It can, on the contrary, erupt suddenly into consciousness without being called forth by any train of thought, and thus bring about an _anxiety-attack_. An anxiety-attack of this kind either consists of a feeling of anxiety alone without any associated idea, or associated with the nearest interpretation, such as sudden death, a stroke, or approaching insanity; or else the feeling of anxiety is combined with paræsthesias (similar to the hysterical aura); or finally, together with the feeling of anxiety there is an accompanying disturbance of any one or more of the bodily functions, such as respiration, heart's action, vasomotor innervation, or glandular activity. The patient lays stress on one or other of these symptoms, and complains of 'heart-spasms', 'difficulty in breathing', 'drenching sweats', 'ravenous hunger' and the like; and in his description the feeling of anxiety frequently recedes into the background or is described quite vaguely as a 'feeling of illness', of 'distress', and so on.

4. It is interesting, and important diagnostically, that the degree to which these elements are combined in anxiety-attacks varies extraordinarily, and that almost every accompanying symptom can alone constitute the attack just as well as the anxiety itself can. There {81} are consequently _rudimentary anxiety-attacks_ and _equivalents of an anxiety-attack_ (all probably having the same meaning), showing a manifold and hitherto little appreciated variety of forms. A closer study of these larval anxiety-states (Hecker) and their diagnostic differentiation from other attacks should soon become a necessary piece of work in neuropathology.

I will here append a list including only those types of anxiety-attack known to me:

_a_. Attacks accompanied by disturbances of the _heart's action_, such as palpitation, either with transitory arhythmia, or with tachycardia of longer duration that may end in grave weakness of the heart, the differentiation from organic morbus cordis never being easy; or pseudo-angina pectoris---diagnostically a delicate problem!

_b_. Attacks accompanied by disturbances of _respiration_, several forms of nervous dyspnœa, attacks similar to asthma, and the like. I would emphasize that even these attacks are not always accompanied by recognizable anxiety.

_c_. Attacks of _sweating_, often nocturnal.

_d_. Attacks of _tremor_ and _shuddering_, which are only too easily confounded with hysteria.

_e_. Attacks of _ravenous hunger_, often accompanied by giddiness.

_f_. Attacks of sudden _diarrhœa_.

_g_. Attacks of locomotor _vertigo_.

_h_. Attacks of so-called _congestion_, practically embracing all that has been called vasomotor neurasthenia.

_i_. Attacks of _paræsthesias_ (but these seldom occur without anxiety or some similar feeling of distress). {82}

5. _Awakening in fright_ (the _pavor nocturnus_ of adults), which is usually combined with morbid anxiety, dyspnœa, sweating and the like, is very frequently nothing but a variety of the anxiety attack. This disturbance conditions a second type of sleeplessness which also lies within the compass of anxiety-neurosis.---I have no doubt in my own mind that the _pavor nocturnus_ of children also includes a type which belongs to anxiety-neurosis. The hysterical tinge, i.e. the coupling of morbid anxiety with the reproduction of some appropriate experience of dream, gives the _pavor nocturnus_ of children the appearance of being something distinct; but it also appears in a pure form without dream or recurring hallucination.

6. _Vertigo_ takes a prominent place in the group of symptoms of the anxiety-neurosis. Its mildest form is better described as giddiness; and its graver manifestations, 'attacks of vertigo' (with or without anxiety), constitute one of the most momentous symptoms of this disease. The dizziness of the anxiety-neurosis is neither rotatory nor is it confined to particular planes or directions as is Menière's vertigo. It belongs to the class of locomotor or co-ordinatory dizziness, as do the cases of oculo-motor paralysis; it consists of a specific discomfort accompanied by the feeling that the ground is rocking, the legs giving way, that one can't keep upright because one's legs are as heavy as lead and are shaking and wobbling. This dizziness never leads to a fall. I venture, however, to assert that a profound fainting fit may supervene in the place of an attack of vertigo of this kind. Other swoon-like conditions in anxiety-neurosis appear to result from cardiac collapse.

An attack of vertigo is not seldom accompanied by the worst kind of morbid anxiety, and is frequently attended by cardiac and respiratory {83} disturbances. According to my observations the vertigo produced by heights, mountains and precipices also frequently belongs to the anxiety-neurosis; further, I do not know whether one would not be justified in recognizing alongside this the existence of a vertigo of gastric origin.

7. On the basis of chronic apprehensiveness (anxious expectation) on the one hand, and a tendency to attacks of vertigo with anxiety on the other, two groups of typical phobias develop, the first relating to common physiological dangers, the other to locomotion. To the first group belongs the fear of snakes, thunderstorms, darkness, vermin, and so on, as well as the typical moral over-sensitiveness, and the forms of '_folie du doute_'; the available anxiety is here used simply to exaggerate the aversions which are implanted instinctively in everyone. Usually, however, a phobia with obsessive strength arises only when, added to such an instinctive aversion, a reminiscence of an experience in which the anxiety could come to expression supervenes---for example, after the patient has actually experienced a thunderstorm in the open air. To attempt to explain such cases as mere continuations of strong impressions would be incorrect; what makes these experiences significant and their retention in memory of long duration is indeed simply the anxiety, which both originally and subsequently thus found a means of expression. In other words, such impressions remain potent only in cases where 'anxious expectation' is present.

The other group includes _agoraphobia_ with all its accessory forms, collectively characterized by their relation to movement. We frequently find a precursory attack of vertigo as the foundation of the phobia; but I do not believe that one can postulate this every {84} time. Occasionally we see that after a first attack of giddiness without anxiety locomotion still continues possible without hindrance, although henceforth constantly accompanied by the sensation of giddiness; but that under certain conditions, such as being alone, in narrow streets and so on, locomotion becomes impossible when once anxiety has become combined with an attack of vertigo.

The relation of these phobias to those of the obsessional neurosis, the mechanisms of which I have discussed in an earlier paper[22] in this Journal, is of the following kind: the correspondence between them is that in both an idea becomes obsessive in consequence of its being connected with an unattached affect. The mechanism of _transposition of affect_ holds good therefore for both kinds of phobia; but in the phobias of the anxiety-neurosis (1) this affect is always the same, always that of anxiety; (2) it does not originate in a repressed idea, proves not reducible further by psychological analysis, and is also not amenable to psychotherapy. The mechanism of substitution does not therefore hold good for the phobias of the anxiety-neurosis.

Both kinds of phobias (or obsessional ideas) frequently appear side by side, although the atypical phobias which are based on obsessional ideas need not necessarily develop on the basis of an anxiety-neurosis. A very frequent, apparently complicated mechanism makes its appearance when the content of an originally simple phobia of the anxiety-neurosis type is replaced by another idea, the substitution being then subsequently added to the phobia. The 'protective measures' originally directed towards combating the phobia are the ideas most frequently employed as substitutions. Thus, for {85} example, 'brooding mania' arises from the patient's effort to disprove that he is crazy, as his hypochondriacal phobia maintains: the hesitations and doubts, and still more the repetitions, of _folie du doute_, arise from a justifiable doubt about his own powers of correct reasoning, since he is aware of the persistent disturbance of the obsessional idea; and so on. We can therefore assert that many syndromes of the obsessional neurosis, such as _folie du doute_ and the like, are also clinically, though not conceptually, to be reckoned as belonging to the anxiety-neurosis.[23]

8. Digestive processes are subject to only a few disturbances in anxiety-neurosis, but these are characteristic. Sensations such as nausea and biliousness are not at all rare, and the symptom of ravenous hunger can, by itself or in combination with others (congestions), constitute a rudimentary anxiety-attack; as a chronic condition analogous to anxious expectation we find a tendency to diarrhœa which has given rise to the queerest diagnostic mistakes. If I am not mistaken, it is this diarrhœa to which Möbius[24] has recently called attention in a short paper. I conjecture further that Peyer's reflex diarrhœa, which he derives from the disorders of the prostate,[25] is nothing but this diarrhœa of anxiety-neurosis. The deception of a reflex relationship comes about because the same factors which are active in the origin of such prostatic affections also come into play in the ætiology of anxiety-neurosis.

The behaviour of the gastro-intestinal tract in anxiety-neurosis presents a sharp contrast to the influence of neurasthenia on those functions. Mixed cases often show the well-known 'alternation of {86} diarrhœa and constipation'. The urgent need to micturate that occurs in the anxiety-neurosis is analogous to this diarrhœa.

9. The _paræsthesias_ which may accompany attacks of vertigo or anxiety are interesting in that they (as also the sensations of the hysterical aura) become associated in a definite sequence; but I find that, in contrast to those of hysteria, these associated sensations are atypical and changing. A further similarity to hysteria ensues because a kind of _conversion_[26], which may otherwise easily be overlooked, to bodily sensations takes place in anxiety-neurosis, e.g. a conversion which takes effect in rheumatic muscles. Quite a number of rheumatic persons, so-called, who moreover are demonstrable as such, in reality suffer from anxiety-neurosis. Along with this increased sensitiveness to pain I have observed in a series of cases of anxiety-neurosis a tendency to _hallucinations_ which could not be explained as hysterical.

10. Several of the symptoms mentioned which accompany or take the place of an anxiety-attack appear also in a chronic form. They are then still less easy to recognize, since the accompanying anxious sensation is less clearly recognizable than in anxiety-attacks. This is particularly true of diarrhœa, vertigo, paræsthesias. Just as an attack of vertigo may be replaced by a fainting-fit, so may chronic giddiness be replaced by a constant feeling of sinking to the ground, exhaustion, etc.

II. THE INCIDENCE AND ÆTIOLOGY OF ANXIETY-NEUROSIS

In many cases of anxiety-neurosis no ætiology is recognizable at all. It is remarkable that in such cases evidence of a grave {87} hereditary taint is seldom difficult to establish.

But where there are grounds for regarding the neurosis as an acquired one, careful enquiry to that end reveals a series of injurious conditions (_noxiæ_) and influences within the sexual life as important factors in the ætiology. At first sight these appear to be very various in their nature but they soon disclose their common character, which explains the fact that they always have the same effect on the nervous system; further, they are either present alone or together with other 'ordinary' injurious factors, which latter may be regarded as having a contributory effect. As this sexual ætiology of the anxiety-neurosis is so very commonly demonstrable I feel justified _for the purpose of this short paper_ in disregarding cases with a doubtful or a different ætiology.

In setting forth in greater detail the ætiological conditions under which anxiety-neurosis makes its appearance it will be advisable to treat of men and women separately. Anxiety-neurosis appears in female persons---disregarding for the moment their predisposition---in the following cases:

_a_. as _virginal anxiety or anxiety in adolescents_. A number of unambiguous observations has shown me that anxiety-neurosis can be evoked in maturing girls by their first meeting with the sexual problem, that is, by any more or less sudden revelation of what had hitherto been hidden, for example, seeing the sexual act, or hearing or reading something of that nature; in these cases anxiety-neurosis is typically combined with hysteria;

_b_. as _anxiety in the newly-married_. Young married women who remain anæsthetic during the first acts of intercourse often fall ill of an anxiety-neurosis which again disappears as soon as the anæsthesia {88} gives way to normal sensitivity. Since the majority of young women remain healthy during a temporary anæsthesia of this kind, there are conditions necessary to the outbreak of this anxiety, which I will mention later;

_c_. as anxiety in women whose husbands suffer from ejaculatio præcox or from impaired potency; and

_d_. whose husbands practise coitus interruptus or reservatus. These cases belong together, for on analysing a large number of examples it is easy to convince oneself that they depend simply on whether the wife obtains satisfaction in coitus or not. The latter case provides the condition for the genesis of an anxiety-neurosis. On the other hand, the wife is saved from neurosis if the husband who is afflicted with ejaculatio præcox can immediately repeat the congress with a better result. Congressus reservatus by means of condoms is not injurious to the wife if she is very quickly roused and the husband very potent; if not, this kind of contraception is no less injurious than the others. Coitus interruptus is almost always harmful; though only for the wife when the husband practises it regardlessly, that is to say, when he interrupts coitus as soon as _he_ is near to ejaculation without troubling himself about the stage of his wife's excitement. If on the other hand the husband waits for his wife's satisfaction then the coitus will be equivalent to the normal for her---but the husband will become a sufferer from anxiety-neurosis. I have collected and analysed a great number of observations which have provided the material for these conclusions;

_e_. as _anxiety in widows and voluntarily abstinent persons_, often found in typical combination with obsessional ideas; {89}

_f_. as anxiety in the _climacteric_ during the last great increase of sexual need.

Cases _c_., _d_., and _e_. embrace the conditions under which anxiety-neurosis in the female sex most frequently arises and is least dependent on hereditary predisposition. On the basis of these cases of anxiety-neurosis---curable, acquired cases---I shall try to prove that the injurious sexual condition (_noxia_) discovered in them really represents the ætiological factor of this neurosis. Before doing so, however, I will go on to discuss the sexual conditions for anxiety-neurosis in men. I propose to set up the following groups, which all have their analogies among women:

_a_. anxiety of the voluntarily _abstinent_, frequently combined with _defence_ symptoms (obsessional ideas, hysteria). The motives which are decisive for intentional abstinence bring it about that a number of hereditarily-disposed or eccentric persons, etc. belong to this category;

_b_. anxiety in men during _frustrated excitement_ (during an engagement to marry), in persons who (from fear of the consequences of sexual intercourse) content themselves with handling or gazing at the woman. This group of conditions (which by the way applies equally to the other sex: engagement, relationships involving sexual forbearance) furnishes the purest cases of the neurosis;

_c_. anxiety in men who practise coitus interruptus. As has been said already, coitus interruptus is harmful to the woman when it is practised without regard to her satisfaction; but it becomes harmful to the man if, in order to provide satisfaction for the woman, he voluntarily controls coitus and delays the ejaculation. It thus becomes intelligible that as a rule only _one_ partner of a married couple practising coitus interruptus falls ill. Incidentally, coitus {90} interruptus but rarely leads to a pure anxiety-neurosis in men; there usually results a combination of it with neurasthenia;

_d_. anxiety in _ageing men_. There are men who have a climacteric like women and who develop anxiety-neurosis at the time of their waning potency and increased libido.

Finally I must add two other cases which are valid for both sexes:

_e_. The neurasthenics who after practising masturbation fall victims to anxiety-neurosis as soon as they desist from their form of sexual gratification. These persons have rendered themselves particularly incapable of tolerating abstinence.

I observe here that in order to understand the anxiety-neurosis it is important to realize that any pronounced manifestation of it occurs only among men who are still potent and among women who are not anæsthetic. Among neurasthenics whose potency has already been seriously diminished by masturbation, the anxiety-neurosis resulting from abstinence is of a very meagre character and confines itself as a rule to hypochondria and mild chronic dizziness. The majority of women are to be regarded as 'potent'; a really impotent, i.e. really anæsthetic, woman is likewise only mildly affected by anxiety-neurosis and tolerates the injurious sexual conditions described surprisingly well.

How far beyond this we may be justified in postulating a constant relation between particular ætiological factors and particular symptoms in the complex of anxiety-neurosis, I do not yet wish to express any opinion.

_f_. The last of the ætiological conditions I have to bring forward appears at first sight not to be of a sexual nature at {91} all. Anxiety-neurosis also arises (in both sexes) as a result of the factor of over-work or exhausting exertion, for example, nights of watching, sick-nursing, or even after severe illness.

The principal objection to my proposition of a sexual ætiology for the anxiety-neurosis will probably run as follows: abnormal conditions in the sexual life of the kind mentioned are found so very frequently that they must be forthcoming wherever one looks for them. Their presence in the cases of anxiety-neurosis quoted does not therefore prove that the ætiology of this neurosis is to be found in them. The number of people, moreover, who practise coitus interruptus, etc., is incomparably greater than the number of those afflicted with anxiety-neurosis and the great majority of the former tolerate this unhealthy condition quite well.

To this I have to reply that we should certainly not be right in expecting to find in the neuroses an ætiological factor of _rare_ occurrence, seeing how very great their frequency admittedly is, especially that of anxiety-neurosis; also that it actually fulfils a postulate of pathology if in an ætiological enquiry the ætiological factor is proved to be more frequent than its effect, since for the latter other conditions are also required (disposition, summation of specific ætiological factors, reinforcement by other 'ordinary' injurious factors); and further, that detailed exploration of suitable cases of anxiety-neurosis proves beyond question the importance of the sexual factor. I will here confine myself to the ætiological factor of coitus interruptus only and to adducing certain observations which confirm it.

1. So long as an anxiety-neurosis in young married women is not yet established, but only appears sporadically and disappears again {92} spontaneously, it is possible to demonstrate that every such wave of the neurosis is traceable to a coitus lacking in satisfaction. Two days after this experience, or in persons of little resistance, on the next day, an attack of anxiety or vertigo regularly appears, bringing in its train the other symptoms of the neurosis, which all again disappear together with the attack if marital relations occur sufficiently seldom. A chance absence from home on the part of the husband, or a holiday in the mountains necessitating the separation of the couple, have a good effect; the gynæcological treatment that is usually resorted to in the first instance is beneficial because marital relations are broken off while it lasts. Strange to say, however, the success of local treatment is but transitory---and even in the mountains the neurosis reappears as soon as the husband in his turn arrives; and so forth. When a physician who understands this ætiology advises a patient in whom the neurosis is not yet established to substitute normal relations for coitus interruptus a _therapeutic_ test of the statements made here is supplied. The anxiety is removed and does not return again without a fresh cause of a similar nature.

2. In the history of many cases of anxiety-neurosis both among men and women we find a striking fluctuation in the intensity of the clinical symptoms, and even in the appearance and disappearance of the whole condition. One year, they will say, was pretty good, the next was frightful; on one occasion the improvement coincided with a certain treatment which however on the next attack turned out quite useless; and so on. Now if we enquire into the number and sequence of the children and compare this record with the history of the neurosis a simple solution results---the periods of improvement and well-being {93} coincide with the wife's pregnancies, during which of course the need for contraception was no longer present. The treatment that had been so beneficial to the man, however, regardless of whether it was at Pfarrer Kneipp's or at a hydrotherapeutic sanatorium, was the one after which his wife had become pregnant.

3. From the anamnesis of patients we frequently find that the symptoms of anxiety-neurosis have at some definite time supplanted the symptoms of some other neurosis, for instance, neurasthenia, and have taken their place. In such a case it can quite regularly be proved that, shortly before the change in the clinical picture, a corresponding change had taken place from one to another of the various kinds of unhealthy sexual conditions possible.

Observations of this kind can be supplemented to any extent at pleasure, positively compelling the physician to acknowledge a sexual ætiology for a certain category of cases; other cases, however, which would otherwise remain quite unintelligible, can at least be understood without difficulty and classified by employing the sexual ætiology as a key to them. These are those very numerous cases in which we find everything that is also present in the previous category---the clinical symptoms of anxiety-neurosis on the one hand, and the specific factor of coitus interruptus on the other---but where something else has interposed itself as well, namely, a long interval between the ætiology we assume and its effect, and perhaps also ætiological factors of a non-sexual nature too. Take, for example, a man who has a heart-attack after receiving news of his father's death and from that time onwards suffers from anxiety-neurosis. The case is not clear, for the man was not nervous before this event; the death of {94} the father who was well advanced in years did not occur under any special circumstances, and one must admit that the normal, expected decease of an aged father is not an experience which usually causes illness in a healthy adult. Perhaps the ætiological analysis will be clearer if I add that this man had for eleven years practised coitus interruptus and always with regard for his wife's satisfaction. The clinical symptoms at least are identical with those that appear in other persons after an unhealthy condition of the same sexual nature lasting for a short period and without the interpolation of another trauma. Other cases must be estimated similarly, that of a woman in whom anxiety-neurosis broke out after the loss of a child, or that of a student whose studies preparatory to his qualifying examination were interrupted by an anxiety-neurosis. I do not find that the effect in these cases is explained by the ostensible ætiology. One is not necessarily 'over-worked' by close study, and a healthy mother usually reacts only with normal grief to the loss of a child. More particularly, too, I should expect that the student would develop cephalasthenia through over-work, and the mother in our example hysteria. That they both develop an anxiety-neurosis induces me to lay emphasis on the fact that the mother had lived for eight years in marital coitus interruptus, and that the student had for three years had a passionate love-relationship with a 'respectable' girl whom he dared not allow to conceive.

These considerations lead to the conclusion that the specific sexual _noxia_ contained in coitus interruptus at least disposes the person concerned to acquire it in those cases where it is not in itself enough to induce anxiety-neurosis. The neurosis then breaks out as {95} soon as the influence of another ordinary injurious factor is added to the latent effect of the specific factor; the former can _reinforce the specific factor quantitatively but cannot replace it qualitatively_. The specific factor always remains decisive for the form taken by the neurosis. I hope to be able to prove this statement in regard to the ætiology of the neuroses also on a wider scale.

Some few pages back I mentioned an assumption which is not improbable in itself---namely, that a _noxia_ such as coitus interruptus attains its effect by _summation_. According to the disposition of the person concerned and the other burdens on his nervous system, a longer or shorter time will be required before the effect of this summation becomes evident. Those persons who tolerate coitus interruptus apparently without harmful results are in reality becoming thereby disposed to the disorder of anxiety-neurosis, which may break out either at any time spontaneously or after an ordinary and otherwise insufficient trauma; just as the chronic alcoholic will in the end develop a cirrhosis or other illness as an effect of summation, or under the influence of a fever will go down with a delirium.