Johan M.G. van der Dennen

Aggression in Freudian Psychoanalytic Theory

In his early psychological papers, Freud (1894; 1896) frequently described the aggressive thoughts and feelings his patients reported to him in the course of their free associations. In his first published discussion of the Oedipus complex (Freud, 1900), he described the little boy's aggressive jealousy of his father, and the same work details many dreams with aggressive features. In 'Three Essays on Sexuality' (1905), the existence of aggression is again recognized in the phenomenon of sadism, although in this context it is considered a perverse manifestation of libido, the sexual instinct. Freud's later clinical observations forced him to deal increasingly with the aggressive impulse as an important factor in the formation of symptoms; this was evident in his discussion of the role of Oedipal rivalry with the father in the production of phobic symptoms in a young boy (1909), and in his recognition of the central position that the conflict between love and hate (ambivalence) occupied in the formation of symptoms and in the motivation for behavior of an obsessional patient who underwent psychoanalysis (1909).
In view of these numerous observations, it is curious that aggression had no place in the basic framework of Freud's first major formulation of the nature of psychic structure - the so-called topographical theory. In this scheme there were two fundamental instinctual drives: the sexual drive (or libido) and the ego instinct for self-preservation. Whenever the libido and its associated affects and fantasies were unacceptable to the ego, the latter repressed the former, forcing them into the unconscious, whence they could achieve discharge only in the disguised and modified form of neurotic symptoms, dreams, or sublimations. In the topographical model, the focus of interest was almost entirely on the libido. Freud's clinical investigations led to an extensive description of the nature of the libido and its component parts; little attempt was made to define further the self-preservative instincts of the ego (Nemiah, 1966).
In 1908, Alfred Adler first proposed the idea that aggression was an innate, primary instinctual drive. The concept of an aggressive drive provided Adler with the cornerstone of a new theory according to which all behavior stemmed from an aggressive 'masculine protest' against feelings of inferiority, sexuality being reduced to the man's aggressive attempt to master the woman. Adler's failure to take into account the discoveries of the unconscious and childhood sexuality led to a sharp disagreement and an irrevocable break between him and Freud and his followers, who failed to see that, despite the narrowness of his theoretical formulations, Adler's introduction of the concept of aggression as an innate drive was a significant addition to the growing body of theory.
The publication in 1914 of Freud's paper 'On Narcissism' was, as his editors comment in their introduction to his essay, "One of the pivots in the evolution of his views", for it marked the first turning of Freud's serious attention to the nature and structure of the ego. From his observations concerning the phenomenon of megalomania, Freud became aware that libido could be directed toward the self as well as toward external objects, and he proposed that this self-directed narcissistic libido formed the "libidinal component to the egoism of the self-preservative instincts". A year later, a second element was added to the ego when, in dealing with the problem of hate in 'Instincts and Their Vicissitudes' (1915), Freud suggested that aggressiveness, too, was a component of the ego instincts - aggression, in other words, was at last given a formal place in the theoretical scheme, though not yet as a full-fledged, separate instinctual drive in its own right (Nemiah, 1966).
It was not until five years later, with the publication in 1922 of 'Beyond the Pleasure Principle', that Freud finally abandoned altogether the concept of the ego instincts for self-preservation, when he became aware that these instincts were really synonymous with narcissistic libido. Simultaneously, for the first time he raised aggression to the level of a distinct and independent drive - or, more accurately, the external manifestation of a more basic drive, the death instinct. In trying to understand the phenomenon of aggression turned against the self in suicide (which he had initially attempted to explain in 'Mourning and Melancholia' (1917)), and of the compulsion to repeat past emotional traumatic experiences (the 'repitition compulsion'), Freud proposed that there was 'an urge inherent in organic life to restore an earlier state of things', that is, through death the organism reverts to an inorganic state. Despite this inherent biological impetus toward death, organisms continue to live for a period of time because the aggression stemming from the basically self-destructive death instinct is partially redirected outward onto external objects, and because the death instinct is itself opposed by the forces of Eros, or the life instinct. It is important to emphasize that the concept of primary life and death instincts is a second-order metapsychological construct that goes way beyond the clinical observations concerning sexuality and aggression and the concepts of a sexual and an aggressive drive derived directly from the clinical phenomena. For Freud, the theory of opposing life and death instincts became a central feature in his final 'structural model' of personality organization; the earlier dualism of a conflict between libido and ego instincts for self-preservation was replaced by the dualism of the conflicting life and death instincts (Nemiah, 1966).
In the further development of the theory, there has been no unanimous agreement concerning the life and death instincts. Some - for example, Federn (1932; 1952), Klein (1948; 1949), and Menninger (1938), the latter two viewing it in a more clinical than biological light - have followed Freud in his postulate of a primary death instinct and have elaborated the concept further. Others, for example, Fenichel (1945; 1953) and Jones (1957), consider such a high order of metapsychological abstraction unnecessary and believe that the clinical observations are adequately dealt with by a simpler concept of an aggressive instinctual drive - a concept inferred directly from observations - without implying that the drive is inherently self-destructive or based on biological characteristics of living cells. In the current literature there is still little consensus, either as to the definition of the word 'aggression' or concerning its ultimate nature and origins (Lipton, 1951; Ostow, 1957).

In 1915, Freud had suggested that the frustration of behavior aimed at gaining pleasure or avoiding pain led to aggression. In 1939, this hypothesis was further developed by Dollard, Doob, Miller, Mowrer & Sears, who were interested in integrating the concepts arising from learning theory and psychoanalysis. They proposed that "the occurrence of aggressive behavior always presupposes the existence of frustration and, contrariwise, ...the existence of frustration always leads to some form of aggression".
A year later, Miller (1941) pointed out that the latter half of the proposition was inaccurate, since frustration does not, in fact, always lead to aggressive behavior; the 'frustration-aggression hypothesis', stated thus, failed to distinguish between an 'instigation to aggression' and aggressive behavior. Miller suggested as a modification of the hypothesis that "frustration produces instigations to a number of different types of response, one of which is an instigation to some form of aggression".
In a recent work, Mowrer (1960) has commented that, because of their behavioristic bias, learning theorists have generally ignored the variables intervening between the stimulus of frustration and the response of aggression; as a consequence they have not paid attention to the inner, subjectively experienced, affective and ideational components of aggression, which form a central focus of interest in the psychoanalytic approach. Their orientation has led them away from the conception of aggression as an innate, biologically determined, instinctual drive and toward the idea that aggression is a response to frustration - and what is more, a response that is learned by the organism. Their attention, therefore, has been drawn to the various types of responses to frustration that may be learned and to the variables determining them, and many of their experimental studies concerning aggression have been designed to investigate the factors determining the strength of the instigations to aggression, the direction and form that aggression takes, and the factors leading to an inhibition of it.
Nemiah (1966) attempts to reconcile these theoretical developments by pointing out that in many ways the two conceptual schemes are complementary rather than conflicting.
On the one hand, psychoanalysis, with its interest in the inner, subjective events of mental life, provides a wealth of material about fantasies and feelings that permits detailed investigation of many aspects of aggression that are not carried over into overt behavior. This, in addition to its fundamental concept of unconscious mental processes, provides the basis for a scientific explanation of many clinical phenomena that would be inexplicable in the purely stimulus-response framework of learning theory; for the same reasons, psychoanalysis points the way toward effective therapeutic measures and supplies a rationale for their use. Learning theory, on the other hand, precisely because of its concern with the stimulus and the response segments of the larger spectrum of aggressive phenomena, has focused interest on the events that provoke aggression. An understanding of the nature and structure of such precipitating events and an awareness of the fact that these may be dangerous situations for the individual who has major conflicts over aggression form an important dimension of crisis theory and the techniques of preventive intervention (Lindemann, 1962).
An emphasis on the learned aspect of aggressive behavior calls attention to the various ways in which aggression manifests itself. Learning theory's concern with the varieties of possible and permissible patterns of behavior available to the individual in his environment underscores the importance of cultural factors in shaping aggressive behavior and in determining where and when it may be expressed. At the same time, the psychoanalytic investigation of inner psychological defense mechanisms provides an understanding of the vicissitudes of the aggressive impulse that underlie its numerous different external manifestations - in other words, it sheds light on the still incompletely solved riddle of symptom choice (Nemiah, 1966).

Freud on Aggression

From the early days of psychoanalysis Freud recognized the clinical importance of the individual's aggressive impulses towards external objects. But at the time when he classified instincts in the two categories, sexual and self-preservative, aggressive impulses were regarded as sadistic components of the sexual instinct. Even in 'Three Essays on the Theory of Sexuality' (1905) Freud was not satisfied with this explanation of the origin of aggression, and in 1909 he suggested that both the sexual and self-preservative instincts had the power of becoming aggressive. The aggressive component of instincts was seen as being in some way related to the need to obtain mastery over the external world.
In 'Instincts and their Vicissitudes' (1915), Freud reached the conclusion that aggression could not be classified as a libidinal impulse, and suggested that it originated in the self-preservative instincts.
In 'Beyond the Pleasure Principle' (1920), when Freud reclassified the instincts, both sexual and self-preservative instincts were subsumed under the life instinct, and aggression was no longer considered to originate in the self-preservative instincts, but was now seen as the outward directed manifestation of the death instinct. Freud emphasized that aggression could only be studied as it occurred mixed with libido, so that the concepts of 'fusion' and 'defusion' became important.
The essential importance of his postulation of the 'Death Instinct' was not, therefore, the recognition of the part played by aggression in human psychopathology, but the new metapsychological understanding it gave of the origins and vicissitudes of the aggressive or destructive tendencies. This led, of course, to new insights into clinical problems. Freud was led to reverse his former opinion that sadism was primary to masochism.
After 'The Ego and the Id' (1923), with the introduction of structural concepts the new conception of aggression was particularly important for the understanding of obsessional neurosis and melancholia, since it threw new light on the process of super-ego formation (Edgcumbe, 1970).
Freud was already giving passing thought to the problem of sexual aggression before the turn of the century. He presumably concurred with the view expressed by Breuer (1895) in 'Studies on Hysteria', that an increase in sexual excitation in male animals leads to an intensification of the aggressive instinct; and that in normal and healthy young men sexuality is "an unmixed aggressive instinct". It seems that at this time Freud was considering only manifest aggression directed outwards, and that he linked aggression with activity and masculinity, sexual aggressiveness being a male prerogative, while feminine development was considered to be passive from the start.
He gave a more detailed exposition of his views in 'Three Essays on the Theory of Sexuality' (1905), though his main concern here was not to understand aggression 'per se', but to understand sadism.
He said that male sexuality contains an element of aggressiveness - the desire to subjugate - because of the need to overcome the resistance of the sexual object. The perversion of sadism "would correspond to an aggressive component of the sexual instinct which has become independent and exaggerated and, by displacement, has usurped the leading position".
Freud was not satisfied with the available explanations of sadism or aggressiveness at this time. He said that nothing had been done towards explaining the intimate connection between cruelty (sadism) and the sexual instinct "apart from laying emphasis on the aggressive factor in the libido. According to some authorities this aggressive element of the sexual instinct is in reality a relic of cannibalistic desires - that is, it is a contribution derived from the apparatus for obtaining mastery, which is concerned with the satisfaction of the other and, ontogenetically, the older of the great instinctual needs". In 'Jokes and their Relation to the Unconscious' (1905), there is a passage in which Freud appears to differentiate between aggressiveness, in the sense of the desire to subjugate or master, and sadism, though both are considered to be libidinal impulses. He is speaking of what happens when the man's sexual aggressiveness (as expressed in sexually exciting speech) is faced with an obstacle, and therefore alters its character. "It becomes positively hostile and cruel, and it thus summons to its help against the obstacle the sadistic components of the sexual instinct".
In the passages mentioned above, aggression is considered to be a libidinal impulse. But in the 1905 edition of the 'Three Essays' there was a passage, deleted in later editions, in which Freud suggested that 'the impulse of cruelty' (referring to the sadistic impulse) might arise from sources independent of sexuality, though this would seem to foreshadow the final development of Freud's theory of instincte the two become united at an early age.
Freud then remained for many years unwilling to accept the existence of an independent instinct of aggression. In 1909, in the course of his discussion of the hostile and aggressive propensities of 'Little Hans', he stated strongly his disagreement with Adler's concept of an aggressive instinct, saying that he preferred to adhere to the view that each instinct (sexual and self-preservative) has its own power of becoming aggressive. Thus here, aggression is no longer thought of simply as a libidinal impulse.
The next step in Freud's investigation of the origins of aggression came in 1915, when he rejected the notion that aggression is a libidinal impulse. In 'Instincts and their Vicissitudes' he discusses the polarity of love (affection) and hate (aggressiveness) which he was later to compare with the polarity between the life and death instincts, and he recognizes the complexity of the relationship between love and hate. He states that hate is older than love, it arises from a different, i.e. non-libidinal source, and he suggests that this source is the self-preservative instincts.
Hate was originally the ego's reaction to the external world with its unwelcome influx of stimuli, so that in the course of their development "the sexual and ego-instincts can readily develop an antithesis which repeats that of love and hate".
However, as a result of his work on narcissism, Freud was already questioning in this paper whether the distinction between the libidinal (object) instincts and the self-preservative (ego) instincts could be upheld any longer. Indeed, in an addition to the 'Three Essays' made in the same year, he remarked that the origins of aggression were not yet understood.
The new understanding of the way in which hostile, destructive tendencies, originally directed to external objects, are turned back on the self in melancholia ('Mourning and Melancholia', 1915), made it difficult to uphold for long the view that aggression has its origins in the self-preservative instincts, for aggression could no longer be explained simply as the impulse to mastery over the external world.
The next step came in 1920, when, with a drastic revision of this theory, Freud subsumed both sexual and self-preservative instincts under Eros - the life instinct - and postulated the existence of the death instinct, in opposition to Eros. Aggression was now no longer considered to have its origins in the self-preservative instincts, but in the death instinct, and Freud compared the polarity of love (affection) and hate (aggressiveness) with the polarity of the life and death instincts.
The death instinct aims at restoring the organism to the state of inorganic matter that existed before life arose. "The dominating tendency of mental life the effort to reduce, to keep constant or to remove internal tension due to stimuli (the nirvana principle...) - a tendency which finds expression in the pleasure principle; and our recognition of that fact is one of our strongest reasons for believing in the existence of death instincts".
In 'Beyond the Pleasure Principle' Freud was mainly concerned with the workings of the death instinct within the organism, but in later papers he turns his attention to the way in which the death instincts are directed outwards, becoming manifest as destructive or aggressive impulses. He stated that the erotic and death instincts are present only in various mixtures or fusions though defusion can occur under certain circumstances. This topic is pursued in 'The Ego and the Id' (1923) where he states that the concept of fusion is an indispensable assumption. The death instinct expresses itself partly as an instinct of destruction directed against the external world and other organisms, and for purposes of discharge is habitually brought into the service of Eros. "The sadistic component of the sexual instinct would be a classical example of a serviceable instinctual fusion; and the sadism which has made itself independent as a perversion would be typical of a defusion, though not of one carried to extremes".
Shortly after his 1920 formulation of instinct theory, Freud introduced his structural concepts in 1923 and the formulations concerning aggression allowed of new understanding of the process of super-ego formation. When the super-ego is set up it is endowed with that part of the child's aggressiveness which, because of his love for them, he cannot direct against the parents, who frustrate his instinctual wishes. The internalized parents become the super-ego, but the severity of the super-ego corresponds not with the severity of the parents, but with the amount of the child's aggressiveness towards them, now taken into the super-ego. The identifications and introjections occurring in super-ego formation involve an instinctual defusion, so that the released destructive impulses add to the severity of the super-ego.
Libidinal regression also results in the setting free of the previously bound instinct of destruction, directed towards the object. This occurs in obsessional neurosis, and the ego has to defend against the anal-sadistic impulses. In melancholia, where the ego identifies with the lost object and the libido regresses to narcissism, all the destructive impulses, now entrenched in the super-ego, are turned against the own ego. "A pure culture of the death instinct holds sway in the super-ego", hence the tendency to suicide.
In 'Civilization and its Discontents' (1930), Freud noted that the analytic literature shows a predilection for the idea that any kind of frustration results in a heightening of the sense of guilt. But he says that this should be regarded as applying only to the aggressive instincts, since we cannot account on dynamic and economic grounds for an increase in the sense of guilt appearing in place of unfulfilled erotic demands. "When an instinctual trend undergoes repression, its libidinal elements are turned into symptoms, and its aggressive component into a sense of guilt".

Freud's 'Death Instinct' ('Death Drive', 'Todestrieb')
In the first general formulation of his new position, Freud (Beyond the Pleasure Principle, 1920) called upon the science of biology for support of the idea that there is a tendency in all living things to revert to the inorganic state - i.e. to die. This is an instance of a general formula"'to the effect that instincts tend toward a return to an earlier state". Since all living things arose out of the inanimate, the trend toward death may be considered inborn, a death instinct, Thanatos. The striving here is no longer for a 'pleasurable' reduction of tension among organ systems which constantly build up new energies. The striving of the death instinct is toward absolute zero. Freud accepted for this process a term suggested by an early follower, Barbara Low: the Nirvana principle (Munroe, 1957).
With this formulation the concept of the sexual instincts was broadened still further. 'Eros' came to represent all the trends in the organism which seek to unify, bind together, preserve, and build up. The term 'libido' was applied not only to the more obviously sexual instincts but to "the whole available energy of Eros... present in the as yet undifferentiated ego-id (the newborn organism) and (serving) to neutralize the destructive impulses which are simultaneously present". Freud did not hesitate at this point to subsume the instincts of self-preservation and the reality principle under Eros, along with instincts leading to preservation of the species. Similarly, love of self and love of others (ego love and object love) belong in the same instinct grouping.
There is no special term like libido for the energy of the death instinct (Mortido and Destrudo have been suggested as terms analogous to libido, but have not been widely accepted), and Freud considered this instinct very difficult to study. So long as it operates internally, it remains 'silent'. 'We only come across it after it has become diverted outward as an instinct of destruction'. The diversion outward seems to be mediated by the skeletal musculature - and in this manner the development of 'aggression' becomes a normal and even necessary phenomenon. Otherwise it remains too far internalized and can lead only to self-destruction. Normally there occurs a fusion of the death instinct and Eros - that is, the outwardly directed aggressiveness develops in close relationship to experiences of libidinal gratification, with a consequent increasing tenderness toward the object of love and constructive care for its preservation. Where such fusion fails to occur, the outwardly directed aggression may remain relatively unmodified or may not appear at all - as in the pitiful infants observed by Spitz, who simply die for want of any loving stimulus to turn their instincts toward the outside world. In later life defusion may occur, resulting in aggression either toward the object or toward the self when the energy attached to the object (object cathexis) is for some reason withdrawn and the psychic drama becomes again internalized.
For most Freudians, the highly complex concept of the death instinct became an inborn aggressivity which requires consideration as such. The young child 'naturally' destroys and takes pleasure in destruction. This urge is thought of not as merely a by-product of thwarted libido but as an instinctual trend in its own right which requires expression and may undergo repression in much the same manner as the sexual instincts. The idea that fusion and defusion with the sexual instincts may occur is generally accepted. Many Freudians consider 'aggression' highly constructive as well as destructive and at times seem almost to equate it with the energetic effort characteristic of the healthy child to investigate and conquer the outside world (Munroe, 1957).

To Freud, then, aggression was always negative or destructive. It was antilife or pathological. And behavior was a manifestation either of eros, of the desire for death, or of a combination of these. The striving for identity, for self-assertion, for social interest, had no role in Freud's perspective. Few have accepted the idea of a death instinct, even though some of Freud's successors, such as Melanie Klein (1950), have adopted and developed this view. A recent work on aggression by the psychoanalyst Fromm (1973) revises the notion of the death instinct in a direction consistent with current ethological findings and psychological research. Freud saw aggression as rooted in a death instinct, although its manifestations may blend with eros. Fromm, however, recognizes two independent sources, only one of which is instinctual. Instinctual aggression is benign and defensive; uninstinctual aggression, rooted in man's character, is malignant and destructive.
According to Fromm, man instinctually protects himself against threats to his survival, his freedom, and other basic values. Harm or destructiveness that results from this defense is unintended or purely instrumental. The aim is to overcome threat, the activity ends when the threat does. Thus, benign aggression is reactive, not appetitive. It is aroused by stimuli, not internally generated by an increase in tension. In this Fromm's instinctual aggression differs fundamentally from that of Lorenz and Freud.
Moreover, where human aggression for Lorenz and Freud is largely negative, hostile, and destructive, for Fromm instinctual aggression is positive, contributing to man's growth, self-assertion, and independence, and thus to the survival of the species.
Malignant aggression, however, results from specifically human passions seated in our character. Man's organic needs and emotions are integrated and organized according to his major goals. This structure of organization is his character. It is a 'human phenomenon', enabling man to adapt to multiple environments and challenges.
Character structures differ, and malignant aggression, organized as it is within specific structures, may never be manifested. Moreover, it takes different forms. It can be vengeful or ecstatic hate and destructiveness. It can be sadistic, with the desire to have absolute control over others, or masochistic in wanting to be completely under another's control. It can be a passion to destroy and tear apart living things, such forms are social categories resulting from man's history an institutions. For Fromm, the way to reduce malignant aggression is to radically alter 'techno-cybernetic society', to create new forms of decentralization in which man would be freer to assert his self and live the good life (Rummel, 1977).
Not all psychoanalysts fall within the schools mentioned. Some, like Storr (1968) adopt almost completely Lorenz's instinctual view and hydraulic model of aggression adding to it various psychological mechanisms that inhibit or channel its expression. Storr feels that aggression is an essential element in society, encouraging competition for food and sex, and ensuring peace and order through status hierarchies. Since it is instinctual, it is 'impossible to believe that there could ever be a society without strife and competition". Moreover, aggression is not all negative. In childhood it is a drive teward the eventual independence and separation from the parents. Indeed, as an adult, the more dependent the person, the more latent the aggression.
Aggression is also a means by which people establish their identity. Identity requires opposition, which is manifested through aggression. The negative effect of aggression is due to confusion between it and paranoid hostility or hatred (in which frustration plays a large role). Such hostility is reduced by encouraging competition (to drain it off), diminishing overpopulation, and preventing aggression from turning into hate (Rummel, 1977).

Freud's Concept of Group Behavior
After 1920 Freud gave considerable attention to the role of the destructive instinct in group formation and civilization. The most detailed exposition is to be found in 'Civilization and its Discontents' (1930), where he states that a powerful share of aggressiveness must be reckoned among the instinctual endowments of man.
The truth is, he writes "that men are not gentle, friendly creatures wishing for love, who simply defend themselves if they are attacked, but that a powerful measure of desire for aggression has to be reckoned as part of their instinctual endowment. The result is that their neighbor is to them not only a possible helper or sexual object, but also a temptation to them to gratify their aggressiveness on him, to exploit his capacity for work without recompense, to use him sexually without his consent, to seize his possessions, to humiliate him, to cause him pain, to torture and to kill him. Homo homini lupus; who has the courage to dispute it in the face of all the evidence in his own life and in history? This aggressive cruelty usually lies in wait for some provocation, or else it steps into the service of some other purpose, the aim of which might as well have been achieved by milder measures. In circumstances that favor it, when those forces in the mind which ordinarily inhibit it cease to operate, it also manifests itself spontaneously and reveals men as savage beasts to whom the thought of sparing their own kind is alien... The existence of this tendency to aggression which we can detect in ourselves and rightly presume to be present in others is the factor that disturbs our relations with our neighbors and makes it necessary for culture to institute its high demands. Civilized society is perpetually menaced with disintegration through this primary hostility of men towards one another. Their interests in their common work would not hold them together; the passions of instinct are stronger than reasoned interests. Culture has to call up every possible reinforcement in order to erect barriers against the aggressive instincts of men and hold their manifestations in check by reaction-formations in men's minds. Hence its system of methods by which mankind is to be driven to identifications and aim-inhibited love-relationships; hence the restrictions on sexual life; and hence, too, its ideal command to love one's neighbor as oneself, which is really justified by the fact that nothing is so completely at variance with original human nature as this...".
In his 'Why War' (1952) Freud states:
"Thus, when a nation is summoned to engage in war, a whole gamut of human motives may respond to this appeal; high and low motives, some openly avowed, others slurred over. The lust for aggression and destruction is certainly included; the innumerable cruelties of history and man's daily life confirm its prevalence and strength. The stimulation of these destructive impulses by appeals to idealism and the erotic instinct naturally facilitates their release. Musing on the atrocities recorded on history's page, we feel that the ideal motive has often served as a camouflage for the lust of destruction; sometimes, as with the cruelties of the Inquisition, it seems that, while the ideal motives occupied the foreground of consciousness, they drew their strength from the destructive instincts submerged in the unconscious. Both interpretations are feasible". (In his descriptions of instinctual urges, impulsions, needs, or drives, Freud invariably used the term 'Trieb'. In the 'Standard Edition' Freud's term 'Trieb' has been translated by 'instinct' throughout).
Freud goes on: "Yet I would like to dwell a little longer on this destructive instinct which is seldom given the attention that its importance warrants. With the least of speculative efforts we are led to conclude that this instinct functions in every living being, striving to work its ruin and reduce life to its primal state of inert matter. Indeed, it might well be called the 'death-instinct'; whereas the erotic instincts vouch for the struggle to live on. The death instinct becomes an impulse to destruction when, with the aid of certain organs, it directs its action outwards, against external objects. The living being, that is to say, defends its own existence by destroying foreign bodies. But, in one of its activities, the death instinct is operative within the living being and we have sought to trace back a number of normal and pathological phenomena to this introversion of the destructive instinct. We have even committed the heresy of explaining the origin of human conscience by some such 'turning inward' of the aggressive impulse. Obviously when this internal tendency operates on too large a scale, it is no trivial matter, rather a positively morbid state of things; whereas the diversion of the destructive impulse towards the external world must have beneficial effects. Here is then the biological justification for all those vile, pernicious propensities which we now are combating...The upshot of these observations, as bearing on the subject in hand, is that there is no likelihood of our being able to suppress humanity's aggressive tendencies...what we may try is to divert it into a channel other than that of warfare".

In 'Totem and Taboo' (1913) Freud followed Darwin's suggestion that the primordial human society consisted of hordes of brothers led by a powerful father. Under the influence of the universal and most focal of all human conflicts, the Oedipus conflict, the sons rebelled against and killed the chief; the horde thereupon changed into a disorganized fraternal society, a leaderless community of brothers. The need of the brothers for a powerful leader eventually led to totemism and later to religious systems - the totem and the deity being the reincarnation of the murdered father. Freud was convinced that humans have a profound emotional need for strong leadership, and this conviction is the cornerstone of all his sociological speculations. In fact, Freud had a general distrust of democratic institutions. The essence of Freud's social theory is that a stable human society becomes possible only when the universal patricidal tendencies of the sons are overcome, so that the family - the 'cell of society' is preserved. Since the taboo against incest, which is also a component of the Oedipus conflict, makes extrafamilial marriage mandatory, different families are bound through marriage into clans, tribes, and eventually nations. The psychological nucleus of cultural development, in this view, thus lies in overcoming Oedipal strivings.
Freud supported his theory by extensive anthropological evidence, most of which he drew from The Golden Bough (1890) by Sir James George Frazer. Frazer's work supported his idea of the universality of both the incest taboo and the taboo against killing the totem animal, which symbolizes the father of the tribe. This sociological theory aroused a great deal of controversy among anthropologists, yet it was the first psychodynamic explanation for the ubiquitous marital laws prohibiting the different forms of incest, as well as for the great diversity of religious taboos in primitive societies (Alexander & Selesnick, 1966).

Freud never questioned the idea that girls go through an Oedipal phase, sometimes called the 'Electra complex', from a Greek myth in which the daughter connives at the murder of a faithless mother.
The little girl must come to 'hate' her mother and become her rival with respect to the father.
Hostility toward the mother is latent in children of both sexes because of the role she must play in weaning and toilet training; that is, the necessary disappointments and frustrations of the infant's acculturation prepare a pattern of 'hatred' available to the child when the positive aspects of normal developmental ambivalence are overwhelmed. The shift toward hatred of the mother, following upon tender attachment, occurs when the little girl first becomes aware that she lacks a penis. Freud writes: "She is wounded in her self-love by the unfavorable comparison with the boy who is much better equipped, and therefore gives up the masturbatory satisfaction which she obtained from her clitoris, repudiates her love towards her mother, and at the same time often represses a good deal of her sexual impulses in general" (New Introductory Lectures on Psycho-Analysis, 1933).
At first her lack doubtless appears to her a personal misfortune, but as she learns that all females, including her mother, are afflicted in the same way, she comes to depreciate all women. "The incentives to hostility (toward the mother), which have been so long accumulating, get the upper hand". The little girl obscurely feels that her mother is responsible for this new, special deprivation. Also, the active and passive genital wishes toward the mother that she, like the boy, has entertained must undergo a serious setback (Munroe, 1957).

In 'Group Psychology and the Analysis of the Ego' (1920) Freud referred to a book by a French physician and social psychologist, Gustave Le Bon, La Psychologie des Foules (1895). Le Bon's main thesis is that when man becomes a part of a group he regresses to a primitive mental state. Acting as an individual, he may be cultivated and rational; acting in a group, he may behave like a barbarian, be prone to violence, abandon his critical sense, become emotional, and lose all his moral standards and inhibitions. His unique, individual features disappear, and the common ancestral heritage in man's unconscious becomes dominant. Freud explains Le Bon's description of these regressive features of mob psychology in terms of the nature of human conscience. The essence of conscience is 'social anxiety', the fear of public opinion; social anxiety is naturally diminished in the members of a mob. Because the voice of individual conscience is silent in a group, all that has been repressed, all that violates the standards of the conscience, is free to appear uninhibited.
Freud follows Le Bon in saying that the behavior of the members of the group is comparable to that of someone in a heightened state of suggestibility, as in hypnosis, but he raises a question that Le Bon does not discuss: Who is the hypnotizer?
Freud contended that the leader of the group subjects its members to a hypnotic spell and that their relationship to the leader explains the relations of the members to each other. The leader becomes each individual's ego ideal, to whom he hands over all his critical faculties, just as the hypnotized individual abandons his self-determination to the control of the hypnotizer. This shared ego ideal, which ties every member of the group to the leader, also determines their interrelationship, for through their common attachment to the leader they identify with each other.
To explain the nature of the group's attachment to the leader Freud utilizes a concept of 'aim-inhibited libido' or desexualized libido. The group members are bound by libidinous ties to their leader, but the leader has no emotional attachments to anybody but himself. It is precisely this narcissistic quality that makes him a leader. "He loves no one but himself, or other people insofar as they can serve his needs". He is 'of masterly nature', 'self-confident', and 'independent'. He thus represents qualities that the group members themselves cannot attain, and because he does so, he becomes their ego ideal.
By introducing the concept of libido, which ties the members of the group to the leader, Freud could dispense with Trotter's (1915) 'herd instinct' as the force responsible for group cohesion. Freud's explanation clarified the mutual relationships between the members of the most elementary social group, the family. The mutual attachments between the members of the family reveal themselves as libidinous ties and do not require the invention of a special new kind of instinct. The same principle can be applied to account for bonds within the family as well as for those operative in the larger extension of the family - social groups. Freud's theory lacks a complete definition of the emotional ties of the group members to the leader; in terms of modern psychoanalytic theory, this emotional relationship can be described as a return to the infant's dependent attitude toward his parents. The regressive nature of group behavior is then satisfactorily explained by the childlike dependence the members exhibit toward their leader, under whose spell they renounce the internalized parental images (their own consciences) and regress to the phase in which they blindly followed the guidance of their parents. Apparently most persons retain sufficient residues of childhood dependency and insecurity to be susceptible to such emotional regression. The blissful security of the Garden of Eden is a perennial motif in art and philosophy. Freud's theory served to clarify some of the dynamics of group behavior. The concept of dependence on the leader resolves the apparent contradiction that a group, which may become ferocious and destructive, is also capable of self-sacrific and devotion. The group attitude obviously depends upon the nature and ideals of the leader, who can influence his followers in either direction. Furthermore, by stressing the group's dependence upon the leader, we can more clearly understand the phenomenon of panic. In time of danger the group's reliance on the leader increases. Should the leader weaken, the members of the group become overwhelmed by paralyzing anxiety, for the necessary guidance is replaced by what is tantamount to desertion of helpless children by their parents. Thus, in time of danger, democratic societies are apt to sacrifice at least some of their freedoms in order to increase the authority of the government (Alexander & Selesnick, 1966).

In 'Civilization and Its Discontents' (1927) Freud focuses on man's hostile and aggressive impulses. In order to become a member of an organized social system man must renounce the unbridled expression of his individual strivings. These restrictions are the price he pays for the greater security he derives from collaboration with others. Freud explicitly states here something that he had previously always implied, namely, that the Oedipus complex is originally repressed because of destructive impulses directed toward the father and not because of its incestuous connotation. The sexual desire for the female member of the family supplies the motive, however, for hostility toward the father, but it is the destructive wish that is repressed by the incorporated parental image, the superego. This repression must be reinforced by external social institutions, of which social justice is an indispensable factor. Only if everyone renounces his asocial impulses can repression be maintained throughout society; if transgressions are allowed, this might mobilize into action repressed asocial impulses. Refraining from expressing asocial impulses through overt action does not relieve guilt feelings because these are aroused not only by overt acts but also by unconscious desires: not only criminal deeds, but criminal thoughts produce guilt feelings. This social guilt is the source of the universal discontent that, according to Freud, is an unavoidable part of social life.

This pessimistic view is essentially identical with Hobbes's (1651) social theory. Hobbes considered that human society consists of people warring against each other and that basically man is a ferocious animal who must be coerced into peaceful coexistence by a powerful tyrant. Law is essentially the law of the strong repressive leader; lacking such repressive external leadership, human beings would eventually destroy each other. Freud added to Hobbes's view the idea of an internal tyrant, the superego, which represents the originally external leader, the father. This social theory, founded on the assumption of a basic destructive instinct as an inalienable part of human nature, has been questioned by many post-Freudian theorists who do not take such a dim view of the human core (Alexander & Selecnick, 1966).

Freud's Concept of Group Behavior: Comments
In 'Group Psychology and the Analysis of the Ego' (1921) Freud proposed a group psychology based on relationship to the leader and fellow followers - long before the Duce and the Führer and Stalin made 'leader' a term of horror (Munroe, 1957).
The essential feature of Freud's position is the developmental nature within the individual of group 'instincts', however profound and universally operative they may appear. There are no inborn trends toward gregariousness, leading and following, imitating, or any of the long roster of traits frequently considered native to man as a social animal. Freud's position is that the observable trends grow out of the experience of the helpless infant with specific biological needs as he necessarily encounters the ministrations and frustrations of parents and - a little later, but still prior to a sophisticated appreciation of external reality -the rivalry of siblings. Such experience inevitably leads to many of the traits which social thinkers considerad 'instinctive' because of their widespread occurrence. The merit of Freud's analysis is that it provides a means of understanding the irrationality of group phenomena more precisely by aligning them with psychological dynamics as established for the individual. Socially oriented feelings are built in infancy and necessarily contain the longings, contradictions, and ambivalences of their origin. Freud points out, furthermore, their close relationship to the self, to ideals and conscience.
The individual is thus prepared to expect almost complete support and gratification from the outside power which awakens attitudes toward the father-leader. The pattern of total self-abnegation and blissful repose in the almighty reaches deep into infancy and can fulfill itself toward God or toward any person, idea, or whatever that fits well enough to serve as an adequate stimulus. All of us to some extent, some of us to a very great extent, have a nostalgia for this early peace. Expectation always runs along lines of magic rather than of rational evaluation of what the leader might accomplish, least of all rational appreciation of the assets and limitations of a specific leader. The follower tends to build up the power of any leader out of his own psychology. The demagogue typically panders to this deep will-to-believe by demonstrations of power and solidarity at a primitive level. Appeal to reason by outsiders can rarely be effective in such circumstances. Even personal hardships are not critically evaluated in relation to the leader's claims, because he has a right to demand sacrifice, just as the parents had in childhood.
The attitude toward the leader, however, is always ambivalent. Loving magnification of his power is accompanied by fear and hatred, rebellion and rivalry, as with the primal father. Hence the ease with which a group can turn against a leader formerly adored. The fickleness of the mob is the natural consequence of the psychological ingredients of group psychology - is, indeed, predictable in its very fickleness when the psychological roots of group devotion are known.
Love of the leader contains, as an integral part of its development, hatred to the point of murder. Murder may always gain the upper hand, as in Freud's concept of prehistory and as may be observed with impressive frequency throughout history. Actual murder and the more pervasive fantasy of murder then lead to a sense of guilt in the follower which makes him more than ever compliant.
This statement of the fundamental ambivalence of attitudes toward the leader is still grossly oversimplified. Love and hate are experienced toward the same object, a divided feeling difficult enough to assimilate by itself. But, further, the love and hate are magically overstated in the infantile psyche, projected upon the parental image, and then introjected as the superego. The adult does not alternately love and hate the leader as the young child can love and hate. The adult can no longer feel simple remorse for a given act, relieved by simple forgiveness or expiation. The leader's attitudes have been incorporated into his own personality, where they are subject to the revamping of complex individual dynamics. Overt action against the leader lowers the pressure of the instinctual need which prompted it - and by this very fact automatically increases the relative strength of other aspects of libidinal ties. Thus, the murder of the father does not bring only a welcome relief from his tyranny. The relief itself allows concomitant feelings of love to gain the ascendancy, along with feelings of severe deprivation of the father's powerful support. Freud here offers an explanation of the often cited phenomenon of deification of the fallen leader by the rebels themselves, the turning of the populace against the leaders who performed the execution in the name of the people, and the unfortunate tendency of revolutions against dictatorship to result in even more severe dictatorial regimes.
To be sure, the overthrow of existing institutions by sudden violence leads to a factual dislocation of services normally rendered by the state in a complex society at such periods, the state is often attacked from the outside. The individual frequently finds himself worse off than before, for reasons which might rationally be tolerated as temporary. Yet in fact moderate revolutions are always swept aside by the intolerance of the masses, who seem then to show a disconcerting love of the whip which goes far beyond rational necessity. Group mobilization of aggression against the leader is followed almost regularly by group mobilization of love and longing for him - and the guilt-laden belief that whatever unpleasant events still occur are the merited punishment of the all-powerful leader - i.e. father. (Freud discusses this attitude in detail in 'Moses and Monotheism').
The irrational submission to increasingly tyrannical leaders and the acceptance of hardships even beyond those that originally provoked the revolt have still another dynamic source, according to Freud's view. Frustration of libidinal needs naturally arouses aggression. The aggression cannot be openly expressed against the father. In part it may be displaced to other individuals - often the unlucky innocent bystander, more often a scapegoat unconsciously selected as somehow appropriate. But in large part such aggression is internalized as part of the superego. The very young child and the psychopath may be relieved of remorse if their crime goes undetected or is forgiven by the external authority. But the superego is immediately aware of aggressive impulses, whether or not they come to action or even to consciousness. Thus, every undischarged aggression tends to be experienced in the overly conscientious person as guilt, as meriting the just retribution of the all-powerful external authority which has now become part of the self. Such retribution is often directed against the self, regardless of immediate external circumstances.
The group member does not merely childishly love and hate, depend upon and rebel against, the father image, and, after periods of overt rebellion, submit blindly to punishment merited for his misdeeds. The internalized leader, the conscience, builds up aggression in itself. In the name of the gentle Christ, the Christian world has shown plentiful examples both of self-flagellation and massacre of unbelievers. In rejecting the external power of the priests, the protestant ethic did not reject these excesses. On the contrary, the Calvinist conscience became one of the most cruel and intolerant forces the world has known.
How, then, does this complex relationship with the leader function in group psychology? Generally speaking, the leader and the admonitions of the leader are substituted for the ego ideal or superego. The great man - in some circumstances, the abstract idea - takes over the individual conscience. Too often this substitution involves a lowering of the intellectual and moral capacities of the individual. By himself the Nazi guard was often as decent a human being as the rest of us, with as little relish for wholesale murder. He did not execute mass murder solely because he dared not disobey, or even because he rationally believed that Hitler knew what was best for the world. These motives and others played a role, but the main determinant of his actions was merely that the leader became his conscience. Our sturdy New England forebears rarely had qualms about persecuting poor old women as witches or dragging sickly infants to baptism in wintry weather. The Puritan father could grieve over the consequent death of his child and comfort himself with the thought of the salvation of the little soul, but it would not occur to him to feel guilty about his action. In fact, guilt was far more likely to attach itself to the hesitation with which he had taken the child to church because of his human fear.
Once the leader or leader-idea is identified with, its dictates become the dictates of conscience. Interestingly enough, people usually do not feel acute remorse for actions committed under conditions of leadership conscience. The Nazi, the participant in a lynching party, may have a morning-after headache as he reviews his actions, perhaps with shame at having allowed himself to be so misled. Yet despite the absence of external coercion in most cases, the individual usually feels - with some psychological justice - that it was not he who committed the crimes but Hitler or the mob or even the Puritan ethic. This attitude was reported as characteristic for most Germans at the end of World War II (Munroe, 1957).
When, as so often, the great man is reviled and killed, his image exerts itself the more strongly as the very personification of conscience. His precepts become the superego of the group and are even more firmly adopted by every member. Freud further asks 'Who becomes a leader and why' ? In 'Moses and Monotheism' (1939) Freud asks: What is the great man? He points out that the definition can be only partly objective. To be sure, the great man must have superior capacities, but his greatness resides mainly in the fact that he can formulate with especial clarity and cogency the trends which are already present in the group.
The members of a group determine (largely unconscious) who can be accepted as leader in the full sense of substitution for the inner ego ideal.
In Freud's group psychology 'the sons' are all banded together against the father and in love for the father. Any differentiation among members of the group in their feeling for one another is ascribed to the role played in the parent-child relationship.
The community of the sons in relationship to the father suggested to Freud the origin of justice. Originally, perhaps, there is only the need to satisfy pleasure strivings and the aggressive instinct, gradually modified in terms of reality and of the powerful dictates of the parent-child relationship. Originally, there is unbounded striving among the sons to supplant the father. The banding together of the sons in prehistory, however, led to curtailment of the potential struggle for absolute power and to the sharing of love (Munroe, 1957).
In 'Totem and Taboo' (1913) Freud formulated his theory of the 'primal event'. In the beginning men lived in family hordes in which a single, dominant, aggressive and jealous male monopolized the females and threw out his sons. The ejected sons (the 'brothers' in Freud's terminology) formed themselves into a homosexual hand, until the fatal day arrived. "One day the brothers who had been driven out came together, killed and devoured their father and so made an end of the patriarchal horde". He continues, "Some cultural advance, perhaps command over some new weapon, had given them a sense of superior strength".
However, although they hated the father, the brothers also envied him and admired him and hence at his death they felt guilty as a result of 'delayed obedience'. Consequently, they invented totemic prohibitions on the one hand and incest taboos on the other. The father was identified with a totem animal which could not be slain except in ritual when the brothers reaffirmed their solidarity, and the incest taboos meant the renunciation of the father's women who were the cause of all the trouble. From these beginnings in ritual and renunciation all that is truly human sprang. And, of course, this was the basis of the Oedipus complex, the fundamental feature of human personality (Fox, 1967).
Towards the end of his book Freud has second thoughts. Perhaps the transition was not bloody and murderous. He says, "The mere hostile impulse against the father, the mere existence of a wishful fantasy of killing and devouring, would have been enough to produce the moral reaction that created totemism and taboo". He continues, "The alteration might have been effected in a less violent fashion and none the less have been capable of determining the appearance of the moral reaction". In consequence, "No damage would then be done to the causal chain stretching from the beginning to the present day, for psychical reality would be strong enough to bear the weight of these consequences".
In the contemporary family group, the little child is typically baffled and resentful at the intrusion of siblings into its early fantasy relations with the parents. Sibling rivalry appears very frequently, despite every effort of the parent not to show partiality, not to neglect the toddler in favor of the new infant. Often, indeed, it is the favored child who is most jealous of any attention paid to the siblings, obviously because his personal expectations are higher. Freud devoted especial attention to the problem of handling guilt in the group, because for him the death wish toward the father plays a great part in group psychology. Religious ritual among primitive tribes and even in the Western church seems to concern itself quite specifically with the matter by (1) dispersing the guilt among all individuals so that the crime becomes anonymous and hence easier for each person to bear, and (2) ritualistically dedicating a scapegoat to the purpose. In religion it is easy to see the sacred character of the scapegoat - how it is hallowed before its sacrifice can have meaning. The Christian ritual (and the almost sacred hatred of rather primitive Christian peoples for the Jews, who 'killed God') lends itself readily to interpretation along these lines (Munroe, 1957).

The infant's attitude in the early oral stage is passive and receptive. When the teeth irrupt and the general maturing of the nervous system allows more active mastery of stimulation, he enters upon the 'oral-sadistic', or cannibalistic, phase, in which he bites with all his strength.
The infant still has no clear comprehension of the object as thoroughly distinct from himself. The object is still closely bound up in 'meaning' with his own instinctual demands, the mode of response being characteristically one of incorporation. Certainly not all biting at this age has a definitely sadistic coloring, but it readily becomes fused with truly aggressive impulses. The baby comes to use his teeth more and more in direct offense or defense or as a punishing response to frustration. By the mechanism of projection, he fears similar aggression in others, mainly in powerful adults. Thus, the primitive experience of devouring can take an aspect of genuine destructiveness. The more the child himself bites in anger, the more he attributes the same impulse to others. Since oral activity is still the main source of pleasure and it's object is genuinely loved, the addition of a sadistic component now makes for a real ambivalence, in contradistinction to the quasi-ambivalence of the earlier period.
Evidence for the interpretation of such 'meanings' comes from various sources: the fantasy life of little children still close to the infantile formulation but able to express themselves to some extent verbally and in the manipulation of objects (play techniques); the speech and behavior of psychotics who have regressed to the narcissistic stage; the dreams of neurotic and normal adults; enduring myths, turns of phrase, and the like found all over the world in the codification of language.
Paradoxically, excessive gratification and excessive frustration of primitive oral needs by the environment both appear to result in a tendency to fixation. Apparently some measure of deprivation, as against passive contentment, is the condition for growth of mature object relations.
On the other hand, it appears that the infant positively requires a considerable measure of gratification of oral needs and may continue to demand it long past the normal period of renunciation (he may be an 'oral character' throughout life) if too severely frustrated in infancy. With undue frustration, the normal early oral dependency is likely to turn either into extreme, effortless pessimism or into a fretful, demanding aggressivity still without constructive effort by the individual himself (Munroe, 1957).

Waelder (1939) and Freudian Group Psychology
It is Waelder's thesis that nation-state communities must be characterized as belonging to the category of masses (as opposed to associations), and hence it becomes important to analyze the group psychology of masses. First, masses tend to have less regard for, and to be more cruel towards nonmembers of the mass or disloyal members than the same individuals belonging to the mass would show where they were not acting collectively. Second, masses are more prepared to suffer hardship and make sacrifices on behalf of the collective interest than individuals in similar circumstances. Third, masses are generally speaking less subject to fear than individuals in like situations. This lack of susceptibility to fear is in proportion to the degree of collective unity of the mass concerned; but on the other hand, in situations where the unity of the mass tends to wane, the intensity of the reaction to danger is in almost geometric proportion to the degree of disintegration, so that panic is likely to occur upon dissolution of the mass. Fourth, there is a lack of balance between the emotional characteristics, on the one hand, and the intellectual nature, on the other hand. There is a diminution or, in extreme cases, a complete suspension of the critical faculty in regard to matters of collective concern. It may be said that there is a kind of 'splitting' of the conscience vis-à-vis individual and collective issues, and this contrasts with the homogeneity, in general, of the conscience of members of an association. The 'splitting' finds particular expression in the contrast between emotional allegiance given to a personal leader or some collective ideology, and the attitude towards a question of a purely private or individual character. The deeper psychological explanation of these characteristics of masses appears to lie primarily in two hypotheses: (1) The mass is a collectivity of individuals, each of whom has substituted for his conscience either his leader or a collective ideology commanding group allegiance, to the extent that this leader or such ideology now conditions what is permitted or prohibited. The internal conscience of the individual stands suspended. (2) The group loyalty to the leader or to the collective ideology derives from a process of emotional identification, produced by common membership of the mass, calling into being unconscious emotional relationships between the individuals in the mass.
With regard to both hypotheses, the further point may be made that inasmuch as the internal conscience, originally constructed by reaction to the external situation of the commands and prohibitions of parents, teachers, and others, has been re-externalized by reference to a mass leader or to a collective ideology, there has been a process of psychic regression. This regression through membership of a mass tends to reproduce the old infantile characteristics of the original development of conscience in childhood; thus, the members of the mass are impelled to suspend their critical and rational faculty, and to believe again in external authority, represented by the leader or the collective ideology. This re-creation of external authority goes far also towards explaining the diminution of fear of outside dangers through being member of a mass. In the same way as a child felt secure and free from anxiety when assured of the love and protection of parents, so there is a similar feeling of security and of lack of anxiety in adherence to a leader, and, in a lesser degree, to a common ideology. There are other factors which serve to explain the suspension of rational judgment. For one thing, the process of mutual identification between members of the mass operates to reduce them to the same common level; the members of the mass yield to the easier option of not using their full intellectual capacity, and adopt the scale of the least intelligent members of the mass. Moreover, allegiance to a leader or subscription to a common ideology tends to hinder the function of independent judgment in areas deviating from the leader's policy or the common ideology. Reason thus becomes isolated from the standards of the individual's own conscience, and cannot perform its synthetic function of assembling and reconciling the various tendencies and processes of the psyche. This failure of synthesis re-establishes the former infantile immaturity of the individual's mind in matters concerning the mass.
It follows that individuals who are ordinarily peaceful and, according to conscience opposed to violence, may regard violence and mass-destruction as permissible where this is sanctioned by the policy of the leader or the common ideology of their group.
Thus, there are innate anti-peace tendencies in the relationships of nation-states (Starke, 1968).
Some psychiatrists are of the view that, just as in the case of certain neurotic patients who lack the will to recover, the will to become mature is beyond the reach of the majority of mankind. As Waelder pointed out in his study; "The majority of people are loath to grow up. Maturity is a burden".


Freud had proposed in 1905 that the basis for neurosis was a conflict between the ego instincts (the self-preservative drives) versus the sexual instincts. Adler (1908) was searching for a principle that would unify psychological and biological phenomena and still fall within the framework of an acceptable instinct theory. The aggressive drive was introduced by Adler as a unitary-instinct principle in which the primary drives, whatever they might be, lose their autonomy and find themselves subordinated to this one drive. The aggressive instinct, then, was the biological 'Anlage', or source, of psychic energy utilized when individuals overcome their organic inferiorities through compensation: "...unstable psychological equilibrium is always reestablished by the fact that the primary drive is satisfied through excitation and discharge of the aggression drive".
If there were a 'confluence of drives' - for example, if the sexual and aggressive drives occurred together - the latter was always the superordinated one. In the 1908 paper Adler proposed that drives could be turned into opposites, for example, the instinct of voyeurism could be turned into exhibitionistic behavior. Furthermore, a drive could be turned against oneself. Freud adopted these two principles in regard to instincts, calling the former 'reaction formation' and the latter 'turning' of the instincts 'upon the subject' (Freud, 1915). Later, Anna Freud (1937) in The Ego and Mechanisms of Defense, listed these mechanisms as two basic ego defenses.
At this point Adler was not aware that he was dealing with unconscious defensive functions of the ego. Nor was he, in his 1908 paper, attempting to create another dualistic-instinct theory. Instead, he was searching for a basic and 'higher principle of motivation' (Ansbacher & Ansbacher, 1956). Even though the aggressive drive as conceived by Adler was constitutionally and biologically derived, Freud could not include it in his instinct theory. Freud considered, in the case of Little Hans, that the boy's hostile and aggressive feelings were manifestations of 'aggressive propensities', which seems 'a most striking confirmation of Adler's views'. Nevertheless, Freud believed that all instincts have the 'power of becoming aggressive'. He could see no reason at this time to include the aggressive instinct in his duality concept or even to give it a place of preeminence, as Adler had proposed. By 1923, in 'Beyond the Pleasure Principle', Freud had placed the aggressive instinct within the death instinct, which was considered antagonistic to the life instinct. By this time Adler considered that the aggressive drive was really a mode of striving by which one adapts to arduous life tasks. Adler, when freed of the necessity to reason in terms of instincts, had sarcastically remarked that he was glad to have made a present to Freud of the aggressive drive. Adler was never to ask for the return of this gift (Alexander & Selesnick, 1966).
It was not this paper on the aggressive instinct that initiated the break with Freud, but rather Adler's work in the years 1910 and 1911. In 1910 Adler wrote for the first time about 'feelings' of inferiority and thus laid the cornerstone for his theory that the child feels weak and insignificant in relationship to adults. Biological stresses and the outcome of instinctual strivings were now relegated to insignificant roles as compared to how individuals reacted to feelings of inferiority. The crucial reaction was that of the 'masculine protest'. The masculine position in our culture is one of strength; the feminine is one of weakness. Each of us has a feeling of weakness (femininity) and a masculine tendency to overcome it, and from this point of view we are psychologically 'hermaphrodites'. Freud had had a similar notion some thirteen years previously, which he felt could not be validated. However, another concept that Freud never disavowed was that of bisexuality. The latter, nonetheless, instinctually rooted and to be taken literally was dissimilar to Adler's concept of hermaphroditism, which considered sex metaphorically.
Adler was now proposing that sexuality be considered in its symbolic sense. Women in our culture do not have a tendency to become neurotic because they covet the penis but because they envy the preeminence of man in contemporary culture. To women the penis symbolizes the over-exalted position of man in society. Should they wish to become men by renouncing their femininity, they will suffer from neurotic symptoms, such as painful menses, painful intercourse, or even homosexuality, all of which are expressive of their masculine protest reactions. Men who try to become excessively masculine are not reacting to anxiety over fear of castration but are overcompensating for their feelings of inadequacy as men. Adler considered that dreams constantly demonstrated the masculine protest reaction. By 1911 Adler became bold in his criticisms of Freud's sexual theories. The Oedipal situation was not to be understood as the striving of the boy to achieve sexual pleasure with his mother but instead as a symbolic battle. Feeling weak and defenseless the boy uses overcompensation to achieve superiority over the father and dominance over the mother (Adler, 1911).

The Neo-Adlerians
The first psychotherapist to propose an aggressive drive was Alfred Adler. In 1908 he published his theory that aggression is a superordinate drive that dominates motor behavior and consciousness and is a confluence of other drives. It is innate, the organizing principle of man's activities, and (of greatest significance to the psychotherapist) can turn on the self, creating various pathological manifestations.
Adler soon reinterpreted this drive as a masculine protest (a drive to compensate for feelings of inferiority), and finally as an upward striving for completion or perfection. In this later view, man was driven, above all else, to improve himself, to overcome. Aggression then became subordinate to this drive, and indeed, when directed at society, was a pathological form of striving (Rummel, 1977; Ansbacher & Ansbacher, 1956).
Although eclipsed by the work of Freud and almost forgotten for decades, many of Adler's views have been revived and transformed. One such transformation is manifested through existential psychotherapy, as in the work of Rollo May. May's major analysis dealing with aggression is Power and Innocence (1972). He considers power to be man's basic drive and aggression as one form of this drive. Power has five ontological levels for May: first, simply the power to be, to exist, to assert oneself as a living thing, akin to what Rummel (1976, 1977) has called identive power; second, the power of self-affirmation, to be recognized and to become significant; third, the power of self-assertion, of pushing against opposition; fourth, aggression, the application of power to overcome blocked self-assertion; and fifth, violence, to which man resorts when nonviolent aggression is fruitless.
Aggression is basic to man, but culturally formed. Not all bad, it is a way the individual affirms and asserts himself. It is manifested, for example, in initiating a relationship, in trying to penetrate another's consciousness, in warding off threatening powers, and in love-making. "The truth is that practically everything we do is a mixture of positive and negative forms of aggression" (May, 1972). The expression of power in its aggressive, constructive forms is healthy. It is when such expression is inhibited or blocked that violence occurs. Violence expresses impotence (see also Hannah Arendt, 1969, who makes this one of her major observations).
The views of Adler and modern neo-Adlerians stand in marked contrast to psychoanalytic thought. Adler was a member of Freud's psychoanalytic school when he proposed his aggressive drive in 1908. Freud initially rejected this view, believing that aggression did not constitute any special instinct or drive (Ansbacher & Ansbacher, 1956). It was not until more than a decade later that Freud, perhaps as a result of the bitter experiences of World War I and its aftermath, recognized an aggressive instinct. This he first elaborated in 'Beyond the Pleasure Principle' (1920), developing the concept in his later works.

Melanie Klein

On the basis of her work with children, Melanie Klein (1932) concluded that Oedipal hostility and guilt existed even prior to the third to sixth year of life, when Freud presumed that the Oedipal complex emerged. She was the first to draw attention to the fact that even an infant could feel hostile and aggressive to the parent of the opposite sex. Furthermore, she believed that because the mother's breast frequently frustrated the infant, as well as fed it, the breast was not only an object of love but of hostility. Klein believed that a child's paranoid feelings - caused by the fear that he will be destroyed by the father, who is his Oedipal rival, or the mother, whom he feels has frustrated him - originate in infancy.
In essence, she felt that the young child's neurosis was dependent upon his own inner difficulties in handling his aggression; if the instinctual drives of the young child were interpreted as early as possible, she thought, his ego would be strengthened.
Klein developed an elaborate theoretical system about early psychodynamic processes that was modeled on clinical observations made on adults, even including such phenomena as paranoid and depressive reactions that subsume the preexistence of guilt feelings (Alexander & Selesnick, 1966).

Klein (1937) takes her departure from the earlier works of Freud and Abraham, with rather more exclusive emphasis upon the earlier (pregenital) modes of infantile sexuality and the death instinct (aggression) than most of the authors discussed, including Freud himself.
In any event, the major dynamics of ego and superego development are pushed back to much earlier periods - the first year of life. Passive sucking gives way to active biting around the sixth month; that is, the child is able to do something about his states of 'anxiety' beyond mere endurance. Furthermore, he has developed some sense of 'objects' as different from himself, and of himself as 'object'. His appreciation of 'objects' is, however, still rudimentary. It is dominated by the organs and biological processes which his own experience has allowed him to recognize (plus dim phylogenetically determined images, according to Klein, of the penis, the vagina, and the general idea of coitus and childbirth). The breast of the mother, his own feces and urine, the genital organs, the process of incorporation and expulsion, the presence and absence of the mother - these are the essential materials the infant has available.
Arguing from the major instinctual trends described by Freud and Abraham, and from her observation of how they operate the moment her techniques can be used with little children, Klein develops the position that 'oral sadism' is of crucial importance for personality development. (1) It is a natural function of the mouth, pleasurable in itself, and it may be intensified by constitutional variations in the structure of jaws and teeth. (2) This rather incidental sadism of biting is intensified and given a definitely hostile direction by the experience of oral frustration in weaning or even in waiting for food. The longed-for breast thus becomes an object of hostility as well. (3) The rage derived from frustration "serves to strengthen the sadistic instincts of the infant". Klein accepts Freud's concept of the death instinct as an active aggression against the self. (4) The emerging anal sadism of the child and his observations (or phylogenetically determined fantasies) of coitus are woven into the prevailing oral-sadistic trend and also serve to heighten his hostile aggressive impulses. Coitus is fantasied as an act of biting, leading to oral incorporation of the penis by the mother. The penis and vagina are conceived of as dangerous weapons - the penis largely equated with the breast, the vagina with the biting mouth.
At this point, in Klein's theory, the narcissistic (pleasure-seeking) libido is necessarily at odds with the increasing and increasingly focused sadism of the child. The infant weaves into this picture its increasing awareness of external objects on the one hand and of its internal processes on the other. Therefore it conceives the object as something to be incorporated or ejected, with the polarity of 'good' and 'bad' so often cited in analytic schools as the first possible autonomous 'judgment' of the child. For the most part, the definition comes from within and is determined by the instinctual conflict itself. Klein writes: "In my judgment, reality and real objects affect (the child's) anxiety-situation's from the very earliest stages of its existence, in the sense that it regards them as so many proofs or refutations of its anxiety-situations". Owing to the interaction of the mechanisms of introjection and projection, which make the object part of the self, as it were, external factors come to influence the formation of the personality.
These mechanisms are derived from swallowing and spitting out; excreting and retaining the stool - processes which Freud (and Abraham even more systematically) pointed out as dominating in the experience of the infant. The mechanism of projection is further aided by the tendency of the organism to turn the essentially autoaggressive death instinct outward in selfprotection. Thus, the infant falls into what Klein calls the 'paranoid position' as the first phase of his active relationship to the world, attributing to the parent its own hostilities. She write's: "The idea of an infant of from six to twelve months trying to destroy its mother by every method at the disposal of its sadistic tendencies - with its teeth, nails and excreta and with the whole of its body, transformed in imagination into all kinds of dangerous weapons - presents a horrifying, not to say an unbelievable, picture to our minds".
Nevertheless, Klein feels compelled to acknowledge the truth of this picture from her studies of three-year-olds in whose lives constitutional factors or actual frustration have built up these normal hostilities to the point of making them unmanageable. These hostilities, projected outward, become the bad mother. The infant expects and fears menace from the outside world as the exact counterpart of its own sadism. It is easily frightened and easily gives vent to rage.
But, of course, there is always the good mother, who facilitates the pleasure seeking of the libido, and there are libidinal urges which find gratification, or at least strive toward gratification. Very early, the infant ego mobilizes one part of the id against another - a division which is the first step toward instinctual inhibitions and the formation of the superego. In this process, the incorporated object (the child's view of the parent) plays an important role at once. Fear of the parent and love of the parent aid the infant in mastering its dangerous hostilities, in alliance with its instinctual erotic strivings.
A little later, the child enters what Melanie Klein calls the 'depressive position'. The significant change rests upon the growing recognition that the object (mainly the mother) has an intrinsic wholeness. Mother is perceived as being in herself both bad and good. One cannot attack the bad mother without also destroying the good mother. Following in general Freud's outline of the dynamics of melancholia, Klein feels that now the 'whole object' is introjected, and the hostility projected in the paranoid position is now experienced by the child as again internal and directed against the self. This shift is essentially intrinsic to the psychic development of the child. External factors again operate to reinforce or diminish instinctual drives in a manner which may so change the balance of forces as to produce qualitatively different results. Nevertheless, the main determinant, for Klein, is the child's realization that the object must be preserved, and that the bad mother cannot be separated from the good mother. In the depressive position the child internalizes the conflict (Munroe, 1957).

Ruth Munroe (1957) on the Pattern of Rage

Munroe is using the term in the sense proposed by Cannon (1915). His early description stands, with such modification as might be expected after some forty years of careful physiological research. Among the psychoanalysts, it is Rado (1950) who has pointed most clearly to the significance of this affectomotor pattern for psychological development. Probably Darwin (1872) was essentially correct in his attempt to relate basic modes of affective expression to the experience of the human species under conditions of animal 'struggle for survival'. Darwin rode his theory too hard at times as regards specific motor acts, and little was known in his time about the extensive neurochemical patterns underlying the rage phenomena. With these corrections, however, it does seem as though Darwin was justified in supposing that the rage pattern had the evolutionary function of mobilizing the organism for active combat. Active combat involves 'the musculature', but rage also involves changes in the interosystems which seem in a general way functional as supporting the combat or escape status of the animal organism. It would be absurd to confine the rage pattern to hitting, biting, and other motor patterns. Increase in blood sugar, adrenalin, heart rate, etc., etc., inhibition of the digestive and at times the reproductive systems, etc., etc., under strain are familiar phenomena.
One may easily understand the very close relations, physiologically, between fear and rage: it is the appreciation of danger that provokes the reaction of combat - danger or frustration. This pattern (doubtless patterns would be preferable) is indeed built into the organism, and therefore has a dynamic 'drive' quality rooted in inner tensions.
It should be emphasized, however, that these 'inner tensions' do not have an inner rhythm of their own, like the sexual systems, like hunger and in a way motility. They arise in response to an external danger or frustration and are oriented toward motor expression. Although the degree and nature of involvement of the interosystems in patterned response gives 'rage' an inborn quality far beyond appropriate reaction to most actual dangers, and far beyond what is encountered in the 'interosystems' per se, it seems to me mistaken to speak of an 'instinct' of aggression in the sense of spontaneously generated 'energy' inherent in man. The formulation that the 'rage pattern' (or patterns) of the interosystems is oriented around perception of danger or frustration and toward motor expression invites investigation of how this orientation takes place. On the side of perception, we must call to mind again the extreme lack of specificity in the native equipment of the human infant as regards cognizance of objects in the outer world, and the capacity of the human mind to build up complex signals and symbols on the basis of experience - to develop substitute signals activity. The process of learning is applicable here, including the emphasis on its unconscious aspects characteristic of the psychoanalytic schools. In fact it is the unconscious (infantile) evaluation of a situation as dangerous or frustrating which makes many reactions of rage, aggression, and hostility so general and so inappropriate to the actual situation as to seem 'instinctive'. This is because they are rooted in experiences of terror and frustration against which the child cannot be wholly protected, which tend to be repressed and so dissociated from conscious learning.
We can easily train a subject to react with fear to an innocent buzzer signal by following it repeatedly with an electric shock. But Freud showed also how the person becomes afraid of his own impulses. After the important experiences of the early years, it can happen all too easily that quite natural inner feelings of sexual excitement or hate become themselves signals for further anxiety rage. Such reactions may - indeed, typically do - happen outside of clear awareness.

Karen Horney

Hostility plays a basic role in Horney's (1945; 1950) thinking different from that in both Freud's and Adler's. She repudiates any notion of an independent instinct toward death or destruction, or even a native 'aggression' beyond what she would call normal self-assertion and enterprise. None of the Freudian positions on this matter would satisfy her. But although hostility is not an 'instinct' the infant may experience his world as hostile in a rather general sense and may develop hostile feelings in response.
The realistic frustration of normal desires directly awakens feelings of hostility. More important for her view of general psychodynamics is that neurotic frustrations self-imposed in the interests of security also awaken feelings of hostility. In both instances, the hostile feelings themselves arouse anxiety - harking back to the helplessness of the infant whose hostilities can have no realistic success (Munroe, 1957).
Horney recognized three general directions of the neurotic personality: moving toward, moving against, and moving away; leading to a rough typology: the compliant (or dependent), the aggressive, and the detached personality.
The aggressive type tends to demand power and prestige and personal infallibility as its major mode of coping with a hostile world. There is exaggerated independence, ruthlessness, cynical 'realism' expressed in a dog-eat-dog philosophy of life. People are considered as exploitable possessions. If sexual prowess has become important to him, the man proves his success by conquering women. Often he wants money or social prestige in his choice of wife and is indifferent - consciously - to 'love' and to her personal merits in so far as they fail to contribute to his own status. Ruthless in business himself, he distrusts his business associates.
The compliant type typically does not bother to read contracts carefully on the psychologically important assumption that everyone is honorable and nice. The aggressive type typically assumes that he will be imposed upon unless he watches out for his interests. 'Outsmarting others' is the principal he uses for his own conduct and in his interpretation of the behavior of others.
The aggressive type, like the compliant, may often appear loving, loyal, and honest - replete with cardinal virtues. Yet his family and other associates are likely to suffer from the neurotic defensiveness of his life pattern (Munroe,1957). This orientation is close to the 'oral-aggressive' type described by Freud and Abraham (1927), and the 'exploitative character' as described by Fromm (1941).

Horney's concept of hostility as not inborn but as a natural reaction against frustration may be mentioned. The child fears retribution for his hostile acts, but then the problem becomes self-perpetuating or increased as the person comes to fear his own hostility - a stimulus from within (the vicious circle). This position is an improvement on positions which consider hostility as purely a reaction to exogenous factors (e.g., Adler) and closer to Freud's insight. In critique of Horney's theoretical position, Munroe (1957) repeats earlier comments to the effect that Horney merges inborn systems and learned adaptational systems with an insouciance which cannot stand the test of careful inquiry by other disciplines, which unduly limits interpretation of many clinical phenomena. This limitation in clinical understanding is less striking as regards the sexual drive systems (which she largely ignores), because of the very important role she assigns to hostility. In practice, indeed, it has often seemed that Horney's 'reactive hostility' becomes a more universal and undifferentiated concept than the Freudian instinctual aggressive drive. Everybody has it, in rather stereotyped relationship to fears of self-assertion dependency.
In our psychologizing about personality trends, we too often overlook altogether the enormous modifications in rage reactions which take place on the principle of realistic learning and which truly dissolve the rage reaction. The 'perceptual side' of the rage patterns, therefore, has components far beyond what is usually called perception.
The problem is further complicated on the motor side of the rage pattern. Our evolutionary heritage seems to be functionally oriented around an increase in physical activity which is inappropriate in most of the situations of danger or frustration encountered by civilized man. This problem has often been recognized. Indeed a popular cure for recognized anger is working it off by vigorous physical exercise. The great popularity of rough sports for spectators is often seen as a means of working off aggression vicariously by identification with the athletes.
The motor component of the rage pattern is probably almost as undetermined at birth as the perceptual component, but it is never unrelated to the stimulus. To suppose such nonspecificity in adult response is to mistake altogether the nature of inborn psychobiological systems and everything we know about how such systems are developed after birth. An inner psychological relationship must obtain before any physical expression of rage can have an effect beyond the temporary distraction suggested in the preceding paragraph.
The question arises as to whether some sort of motor expression is absolutely necessary to adequate discharge of the inner tensions mobilized by a rage-provoking situation, as might be theoretically expected from the phylogenetic origin of the rage patterns, or whether expression through verbal or physically limited devices will serve. The problem is a complex one both from the physiological and the psychological point of view.
(1) The increased incidence of hypertension, ulcers, etc. in modern society suggests that constant bodily 'preparation for combat' without adequate discharge may have secondary consequences throughout the interosystems. These conditions often have a psychogenic basis in that the excessive stimulation of the 'preparatory' inner rage pattern derives from psychological situations whose rage-provoking aspects are not consciously recognized or responded to appropriately. The organic changes are genuine and may become irreversible.
(2) The motor component not infrequently takes the form of restlessness or muscular tension shown in fatigue, clumsiness, and at times specific 'odd' motor behavior. It is not yet clear why this explicitly 'motor' form appears instead of involvement of the interosystems as in (1).
(3) By luck, good judgment, or the nature of his neurotic defense mechanisms, the individual may manage to avoid rage-provoking situations. If he can steer clear of competition, dominating employers, a difficult family situation, or whatever else he finds disturbing, his life may be limited, but the inner tensions of the rage pattern will not be aroused.
(4) The individual may be so convinced of the efficacy of his method of handling the situations that they no longer require the inner mobilization of the rage pattern. Thus, there is a shift in perception of the stimulus situation. This approach was discussed earlier as the normal reduction of anger with understanding - moving the chair instead of kicking at it. It is difficult for the layman to understand that an absurd compulsive ritual, which the patient himself considers nonsensical, may serve at the unconscious level to bolster his sense of mastery to the point at which the rage is unnecessary. The neurotic ritual is a 'defense', but the defense begins with a distortion of the situation on the perceptual side such that the ritual can handle it without the development of the excessive rage-fear pattern. It is known that ritualistic defenses may ultimately shift toward somatization (e.g. the ulcer) or toward psychosis if the patient too far exaggerates his distortions on the perceptual-evaluative side.

Aggression and Anxiety
Freud attempted to derive the prototype of anxiety from the experience of the infant during birth. Munroe (1957) suggests that the infant shows this syndrome at birth because it is part of a broad evolutionary patterning which appears spontaneously under a variety of threatening conditions. Where the 'combat' aspect of the pattern is predominant, we may fairly call it 'rage' - and 'aggression' if we wish to emphasize its motor phase and the self-confident feeling of the organism vis-à-vis its enemies. The type of behavior resulting looks aggressive and often has a kind of spontaneous cruelty beyond justifiable defensive reaction.
These are relatively uncontrolled 'combat' situations. The relation to 'fear' is clear. It is a rare bully who is not a coward underneath. The rage-combat aspects of the pattern may be built up secondarily in their own right and become predominant in most situations, but they are very closely related to the fear-withdrawal patterns. These patterns have a good deal of similarity both in the underlying biological structure and in the kind of external situation which provokes them. We all know how easily fear turns into rage and vice versa - in fact, how often it is difficult to determine whether the person is angry or frightened. This statement holds whether we speak of the physical manifestations of changes in the interosystems or of the psychological manifestations. The psychoanalyst frequently believes that the patient is essentially afraid when he feels angry, or that he masks his aggression under the feeling of fear.

In summary of the discussion of aggression, it seems a mistake to lump all of the phenomena commonly called aggression (mainly by Freudians) under the single heading of a single instinctual drive. Many of these phenomena result from the operation of what Munroe has called the nonsexual drive systems, with special emphasis on motility. 'Aggression' in any hostile sense of the term is here a matter of social interpretation of essentially neutral behavior, or a reactive hostility to situations created by the essentially neutral behavior, or a by-product of the effort to establish an effective idea of the self. Doubtless more 'or's' could be added.
Munroe further suggested that man inherits patterns of response to situations of danger or frustration which may conveniently be called rage. In order to be brought into action, these patterns require an external situation interpreted as dangerous or frustrating, but once aroused they tend to operate like drive systems with their own inner tensions. Although 'rage' is not spontaneously generated, as are the tensions of the sexual systems, hunger, motility, etc., experiences of danger and frustration are so universal in infancy that it becomes almost a theoretical quibble to deny the drive quality of aggression.
Stress was laid on the very great modifiability of the 'perception' of danger or frustration in the human species and on the development of unconscious signals - symbols. The subjective feeling of 'rage' may itself become a danger signal (instinctual conflict; the vicious circle). The expression of 'rage' is also highly modifiable. It seems likely that the evolutionary function of the patterned responses of the interosystems was to prepare the animal for physical combat. The import of this circumstance for societies in which gross motor expression is usually inappropriate was discussed at some length - inconclusively. Munroe did, however, try to suggest how some types of pathology might be understood as the consequence of inhibition or distorted expression of the motor component, and others as means of avoiding or 'mastering' rage-provoking situations.
Finally Munroe remarked that most of her discussion of the rage pattern is familiar to psychoanalytic theory under the heading of 'anxiety', and she pointed to the very close relationship between fear and rage in clinical observation. Instead of positing a quite vague concept of anxiety and an independently instinctual aggressive drive, Munroe thinks it is preferable to assume inborn reaction patterns to stress at the very deep level of integrated mobilization of the interosystems. Even at the physiological level the patterns of rage and fear seem to be partly overlapping, partly antagonistic - but so closely related that the fleeing animal at bay suddenly becomes a powerful fighter. At the psychological level the problem becomes infinitely more complex, partly because the signals arousing interosystem patterns are learned in an intricate manner, partly because stable integrating systems are developed in the course of living which very profoundly influence modes of perception and response.

Criticism of Psychoanalysis: Zillmann on Freud

The most remarkable aspect of Freud's death instinct is the fact that he considered it to be directed against the self. In contrast to the quite common view that man readily engages in destructive behaviors against rivals, Freud posited in no uncertain terms that the ultimate objective of the death instinct was the death of the self . However, since in comparison with outward-directed hostile and aggressive activities, explicit self-destructive behaviors are relatively rare occurrences, he drew upon such psychoanalytic mechanisms as displacement - which had been conceived of earlier, independent of aggression - to convert the potential attack on the self into an attack upon others. In Freud's view, then, the death instinct forces the individual to direct aggressive acts against the social and physical environment in order to safe him- of herself from self-destruction. Outward-directed aggression, interpersonal aggression in particular, is thus a derivative of self-directed aggression and not a primary force (Zillmann, 1979).
As noted by Marx & Hillix (1963), Freud's self-centered death instinct constitutes the most controversial element of psychoanalytic theory. The concept has been accepted in full by only a few followers (e.g. Klein, 1950, 1957; Nunberg, 1955). Some followers have applied minor modifications (e.g. Waelder, 1956). Other psychoanalytically inclined investigators have accepted the notion of instinctive aggressiveness but have transformed Freud's concept of primary self-aggression into an aggressive instinct that is primarily directed outward (e.g. Beres, 1952; Hartmann, Kris & Loewenstein, 1949; Loewenstein, 1940; Mitscherlich, 1963). Still others in the psychoanalytic movement have rejected the notion of instinctive aggression altogether, replacing it with the conception of aggression as primarily reactive behavior (e.g. Fenichel, 1945; Horney, 1939; Stone, 1971).
The death instinct has been criticized mainly on intuitive grounds. Stone (1971), for example, noted that the concept entirely neglects the behavioral significance of flight, which is always an alternative in conspecific and interspecific aggression, and he considered this fact in itself an a priori difficulty in the concept of a destructive instinct. Similarly, one might detect conceptual problems with the displacement or sublimation of self-destructive tendencies. According to psychoanalytic theory (cf. Toman, 1954), these mechanisms are forced into operation by the blockage of basal inclinations; they are not spontaneously activated. With the death instinct, however, the basal urges of self-destruction are never blocked. The avenue to self-inflicted death is always open. Thus, the proposal that displacement and sublimation are constantly involved in warding off self-annihilation by redirecting the immanent destructive forces toward the outside world is less than compelling intuitively (Zillmann, 1979).
Such objections may raise doubt, but they are not crucial. Freud's concept of the death instinct evades decisive criticism because it is sufficiently vague. It involves nothing concrete that can be operationalized. This vagueness applies equally to the transformed death instinct - the aggressive instinct accompanying libidinal urges rather than opposing them (cf . Hartmann, Kris & Loewenstein, 1949). In psychoanalytic theory, both the death instinct and the aggressive instinct have remained imaginary forces. As such, they are generally employed to 'shed light' on behavior that has already manifested itself. Since the forces are hypothetical and unmeasurable, behavior cannot be predicted on the basis of specific variations in the instincts in question. Post facto accounts of behavior may sound plausible, but they should not be confused with explanation that is established mainly by the accuracy of the prediction of outcomes. In psychoanalytic terms, any aggressive act can be readily 'accounted for' by mapping it onto the continuum of antagonistic libidinal and destructive energy (i.e. Freud's model) or that of jointly operating libidinal and destructive forces (i.e. the model of Hartmann et al.). Also, the distinction between primary inward-directed and primary outward-directed aggression is heuristically less relevant than it might appear. Since redirecting mechanisms are invoked in both cases, plausibility can be equally achieved by both reasoning procedures. Sadism, for example, 'results' in Freud's model from redirected self-destructive energy that dominates libidinal impulses, and in the model of Hartmann et al., from dominant destructive urges. In the case of masochism, the former model posits self-destructive urges that dominate the libido, and the latter model, the redirection of originally outward-directed, dominant destructive energy. Additionally, the notion of simultaneous cathexis - that is, the simultaneous concentration of libidinal and destructive energy on a particular object - which has been entertained by Hartmann et al., seems to be reduced to an exercise in semantics. In terms of the death-instinct paradigm, any combination of sexual and aggressive behavior tendencies can readily be 'explained' as a compound of life and death instinct in which either the life or the death component is dominant. Given these conceptual ambiguities, the two models are equally adequate or, more correctly, inadequate. With regard to the death instinct specifically, it becomes a matter of taste whether to endorse or to condems it. Epistemically, the assumption of such an instinct is simply pointless. It fails to further our understanding of aggression (Zillmann, 1979).

The element of Freudian theory that has proved most influential concerns the relief from libidinal or destructive pressures by appropriate consummatory action. It deals with the purgation of violent urges, an effect that may be transitory or may endure over longer periods. This phenomenon is best known as catharsis.
Although in Freud's dual-instinct theory, the forces of life and death are antagonistic, both instincts are conceived of as mechanisms that serve the conservation of energy. Both instincts actuate behavior that, at the very least, averts prolonged states of elevated energy mobilization. More characteristically, however, they actuate behavior that effects a reduction of such states. In Freud's view, the reduction of tension associated with a state of need (Reduzierung der Bedürfnisspannung) is a primary function of an instinct. The behavior is actuated to reestablish the state (generally associated with minimal tension) that prevailed before the instinctive forces were potentiated and to remove the stimuli that impinged upon the instinctive propensity. It is thus consistent with the model to say that libidinal energy is absorbed, neutralized, or reduced by the performance of direct sexual behavior or sex-related activities in a broader sense. More importantly here, it is equally consistent with the model to speak of the absorption or reduction of destructive energy by aggressive action.
The concept of catharsis is not simply identical to that of tension reduction, however. Freud discussed catharsis in connection with the expression of emotion. This relates, in fact, back to the original meaning of the concept, which grew out of Greek dramatic theory. In relation to drama, particularly tragedy, catharsis referred to a feeling state caused by witnessing tragic events. "It meant the stillness at the center of one's being which comes after pity and fear have been burned out. The soul is purified and calmed, freed from the violent passions" (Schaar, 1961). In the Freudian application of this notion, catharsis became the purging of hostile and aggressive inclinations brough about by the mere affective display - not the actual execution - of such inclinations (cf. Feshbach, 1970). Unlike in Greek dramatic theory - where the purgation of 'violent passions' is viewed as resulting from witnessing emotional expression - in Freudian reasoning, purgation is expected to come from the expression of aggression-related emotions, primarily hostile feelings and anger. By means of the presumed purgation, that is, by a reduction of destructive energy, the mere expression of hostile and aggressive feelings is seen to prevent truly harmful and injurious behaviors - or at least to lower their strength and the likelihood of their occurrence. This points out what must be considered the principal element of the cathartic process: Destructive behavior can be weakened or eliminated by some form of less destructive or nondestructive substitute action (Zillmann, 1979).
Many investigators (e.g. Berkowitz, 1962; Buss, 1961) have taken Freud's treatment of instinctive and cathartic processes to mean that Freud conceived of destructive energy as a finite, well-defined quantity, which fluctuates markedly with particular expenditures of energy. They attributed the so-called hydraulic energy model, in which forces are treated as analogous to liquids in a container, to psychoanalytic reasoning. In the hydraulic model, any increment in energy is associated with an increase in the amount of liquid held in the container, and any decrement is associated with a decrease. Liquid can be released through regular outlets. If it is not released, the reservoir may grow and build pressure to intolerable levels. By the same token, if liquid is drained through nonhabitual outlets, only the remaining reservoir can be discharged through the regular channels.
With regard to the cathartic process, this analogue suggests that if a nondestructive behavior absorbs destructive energy, less energy remains to motivate destructive behaviors. There is thus a purgation of hostile and aggressive forces. Obviously, the analogue also suggests that the motivation for hostile and aggressive activities is lowered by the performance of independent or related truly destructive behaviors (Zillmann, 1969).
Jakobi, Selg & Belschner (1971) have taken issue with this interpretation of Freudian theory. They criticized Berkowitz (1962), in particular, for arguing that the notion of a reservoir of aggressive energy that can be drained and abreacted through aggressive action is an integral, essential part of the psychoanalytic theory of aggression. Jakobi et al. insist that in Freudian reasoning, there is "no direct connection between catharsis and aggression which would lead to the expectation that certain aggressive acts (e.g. socially sanctioned ones) could effect a reduction in others (e.g. socially disapproved ones)".
How much and what kind of actual hostility can be curbed by hostile expressions remains unclear in Freudian writing. In some current psychoanalytically inspired therapeutic techniques (e.g. Bach & Goldberg, 1974; Lowen, 1967, 1970, 1971; Perls, 1969a,b), however, the cathartic powers of hostile expression are treated as unquestionable and seemingly unlimited - in spite of decisively negative research evidence on this point (cf. Berkowitz, 1973).
With regard to the catharsis notion, Freud seems to have been erroneously credited, for better or worse, with a paradigm of greater specificity than that which he proposed. His treatment of tension reduction has been 'forced' into a model in which: (a) the amount of available destructive energy is finite, (b) the discharge of energy by aggressive action drains the reservoir to a point where other destructive behaviors are deprived of their motivational force, and (c) the reservoir is not immediately replenished after energy discharge. According to this model, aggression against a particular target should indeed result in reduced subsequent aggressiveness against any other target. This is not what Freud suggested. If he had entertained this view, he should have had great hopes for the control of violence by the harmless abreaction of destructive impulses toward specially selected targets. There would, in fact, have been little cause for painting the bleak picture of inevitable violence. As Jakobi et al. (1971) suggested, the more specific model of the cathartic process just outlined is not due to Freud but derives instead from the work of Dollard, Doob, Miller, Mowrer & Sears (1939) on frustration and aggression. However, erroneous as the accreditation may be, the notion of catharsis as a mechanism by which aggression or pseudoaggression generally lowers subsequent aggressiveness has become integrally associated with psychoanalytic theory. The basic concept has been promoted with enormous success by Dollard et al., who were very much under the influence of Freudian thought. The instinct component was eventually replaced by drive forces (Zillmann, 1979).

Aggression as an Instinct and Drive

According to Brenner (1973), Freud's theory assumes "that in all of the instinctual manifestations which we can observe, whether normal or pathological, both the sexual and the aggressive drives participate". To use Freud's terminology, the two drives are regularly 'fused', "though not necessarily in equal amounts". Moreover, the aggressive drive can be traced through the transitions from oral to anal to phallic that Freud described for the manifestations of the sexual drive. Thus, Brenner states, "Aggressive impulses in the very young infant are apt to be discharged by oral activity such as biting. Somewhat later, soiling [and] retention of feces become important outlets for the aggressive drive, while to the slightly older child the penis and its activity are used, or at least conceived of (used in fantasy) as a weapon and a means of destruction respectively". That is, although "the child of five or six... does not actually use his penis as a weapon... the weapons he uses in his games and fantasies, such as spears, arrows, guns, and so on can be shown by analysis to represent his penis in his unconscious thought".
Even before Freud arrived at his conclusion that aggression is a basic drive lodged in man's unconscious, he put forth a number of ideas that became incorporated in theories of aggression. At first Freud postulated that oral frustration in the infant brings forth rage reactions. This idea, that frustration leads to aggression, became the basis for the frustration-aggression hypothesis later elaborated by Dollard et al. (1939).
Freud's theory of the libido also has components relevant to aggression. Specifically, Freud described cathexis as the investment of libidinal or psychic energy in a bodily part, a function, an idea, an object, or a person. Thus, in the oral stage there is cathexis of the oral region; when the teeth grow in, the 'oral erotic stage' is followed by the 'oral sadistic stage'. A 'fixation' at this stage - either through frustration or through parental mishandling or rejection - may, according to Freudian theory, lead to a strong sadistic component in the character structure; and this component may be manifested later as 'verbal biting criticism'. Similarly, in the 'anal sadistic stage' characteristics such as hostility and outbursts of anger may become channels for aggression. Thus, individual manifestations of aggression, such as sadistic sexual practices (genital stage) and other forms of sadistic behavior, would be explained by the vicissitudes of the libido (Kutash, 1978).
The Freudian assumption of the Oedipus complex also can serve as a possible explanation for certain types of aggression and violence, particularly when the so-called oedipal phase is not resolved adequately by identification with the parent of the same sex. Investigating the unconscious mental lives of neurotic patients, Freud uncovered fantasies of incest with the parent of the opposite sex, coupled with intense envy and murderous rage against the same-sex parent. He drew an analogy between these fantasies and the Greek legend of Oedipus, who unknowingly killed his father and married his mother. Later psychoanalytic studies confirm that this constellation, which Freud (1900) called the Oedipus complex, also is present in normal persons and is experienced by all people from the age of 3 to 6 years; and anthropological studies of different cultures confirm the existence of incestuous conflicts and parenticidal impulses in every culture studied (Roheim, 1970). In discussing the universality of the Oedipus complex, Brenner (1973) includes "inverse or negative oedipal wishes; that is, fantasies of incest with the parent of the same sex and murderous wishes toward the parent of the opposite sex". He asserts that this is also a general phenomenon. He goes on to stress that "The most important single fact to bear in mind about the oedipus complex is the strength and force of the feelings which are involved. It is a real love affair. For many people it is the most intense affair of their entire lives, but it is in any case as intense as any which the individual will ever experience". An unresolved Oedipus complex, coupled with the failure to develop sufficient ego strength and an adequate superego, has been offered as an explanation for various crimes of passion - particularly patricide and matricide and in instances where the victims are parent surrogates. The recent 'Son of Sam' case may well be considered an example of such a series of crimes. The fear of retaliation, which is part of the oedipal constellation, often results in the killer's unconsciously leaving clues by which he may be caught and punished; or he may actually send notes to the police, pleading to be stopped or apprehended (Kutash, 1978).
As mentioned, Freud's ultimate theory of aggression evolved gradually and was subjected to various revisions. Until 1905 he regarded sadism as an aggressive component of the sexual instinct, which becomes exaggerated and independent and, through displacement, assumes the leading position (Freud, 1905b). At the same time, he recognized that the impulses of cruelty arise from sources that are independent of sexuality but may become united with it at an early stage (1905b). This is the early inkling of his later idea that the two drives are independent but are regularly 'fused'. Four years later, Freud was still ambivalent about whether aggression represents an independent drive; he stated that he could not bring himself to assume the existence of a special aggressive instinct alongside of the familiar instincts of self-preservation and sex and on an equal footing with them (1909). By 1915 he was pursuing both lines of thought; he sometimes regarded aggressiveness as a component of the sexual instinct and sometimes as an independent drive separate from sexuality (1915a).
By 1920 and even more so in 1923, Freud revised his entire theory of instincts and resolved his ambivalence by a new hypothesis - that of the life and death instincts, Eros and Thanatos. He now hypothesized that the muscular system, serving as a 'special organ', can divert the destructive impulses stemming from the death instinct onto the external world, thus successfully neutralizing the death instinct. Through the muscular system, then, the death instinct expresses itself, though probably only in part, as an instinct of destruction directed against the external world and other organisms (1923). He later referred to this instinct as the instinct for mastery or the will to power (1924). Freud's assumption of the death instinct - which, he felt at the time, is inherent in all living things and can be handled only by turning the destructiveness against the outside world through aggression - led him to a painful conclusion: War and destructive aggression are inevitable; the alternative is illness, self-destruction, masochism, or passive and ineffectual behavior leading to self-defeat.
Freud's contribution consisted of defining aggression as a basic drive. "By far the most widely accepted notion of aggression among practical workers in the field of mental health, especially, and in political science and social organization has been the view that aggression is a fundamental human drive" (Singer, 1971). Freud was not the first to designate man's aggression as an instinct. McDougall (1908) did so but stressed the desirability of man's socializing his instinct for 'pugnacity'.
Freud struggled very hard to reconcile the conflict between his scientific objectivity and his therapeutic humanism, and he began to search for ways of controlling aggression and directing it into adaptive channels. It remained for later psychoanalytic thinkers, especially Hartmann. Kris & Lowenstein (1949), to refine and complete this part of the theory and to distinguish clearly between adaptive aggression and destructive or dangerous aggression and violence. Freud also at one point emphasized the ego instinct or the instinct of self-preservation, which was later taken up and developed as the source of adaptive aggression or self-assertion (Kutash, 1978).

Non-Freudian Psychodynamic Theories of Aggression

Some of Freud's earliest disciples eventually departed from classical psychoanalysis and founded their own psychodynamic systems. One of these early disciples was Alfred Adler (1927), who developed a theoretical system which he called 'individual psychology'. Of prime importance in Adlerian theory is the child's feeling of helplessness and his development of an inferiority complex, for which he strives to compensate by a drive toward superiority or power. Adler stressed attention-getting and power drives as strong motivating forces, particularly in neurotics and even more so in some varieties of severe pathology. To recognize the drive for power as a source of aggression, one need only recall the numerous examples of crimes committed under the influence of an inordinate need for attention and for power over others.
A theory of destructive aggression based on an inordinate drive for power has long been put forth as an explanation for war and rivalries in politics and in all kinds of organizations.
Adler postulated sibling rivalry as one of the basic behavioral motivations. Starting with Cain's murder of his brother Abel, the story of Jacob and Esau, and the observations of competition between brothers and contemporaries in all walks of life, one could support strongly the importance of sibling rivalry and its derivatives as an explanation for aggression. Adler himself was involved in an eventual rivalry with Freud, and the power motivation may have been a factor leading to a split between the two men. Another Adlerian hypothesis related to aggression is his theory of masculine protest. A comprehensive Adlerian theory of aggression would include as components the inferiority complex, the drive for power, sibling rivalry, ordinal position in the family, attention-getting mechanisms, and masculine protest in various combinations to explain each occurrence. It is interesting that Freud, in 'The History of the Psychoanalytic Movement' (1914) stated, "The view of life which is reflected in the Adlerian system is founded exclusively on the aggressive instinct; there is no room in it for love".
Carl Jung (1933) is another early disciple who broke with Freud and then went on to develop his own system, which included such concepts as psychological types, extroversion-introversion, archetypes, racial and collective unconscious, superior and inferior functions, and a mystical preoccupation with transcendentalism and even life after death. A theory of aggression constructed on the basis of Jungian hypotheses would regard violence and destructive aggression as an unleashing of primordial archetypical behavior inherent in the collective unconscious. An extrovert will 'act out' his aggression against others or against the environment, whereas the violently aggressive introvert will be self-destructive and perhaps suicidal. Somewhat similarly, a person in whom thinking is the superior function and feeling the inferior function will exert intellectual control over aggression; in contrast, the person in whom the feeling function is superior is more likely to express aggression overtly. Finally, some archaic experiences which seem to occur in a number of cultures - experiences such as 'the passion to spill blood' or 'blood lust' - may fit in with Jung's conception of the racial unconscious and archetypes (Kutash, 1978).
Otto Rank (1945) departed from the Freudian fold mainly by putting forth his theory that the trauma of birth is the original source of anxiety and the basic cause of neurosis. In Rank's view, therapy consists mainly in mobilizing the will of the patient so that he can be psychologically 'reborn'. The drive toward separation and differentiation from the mother through assertion of an independent will would serve as a Rankian explanation of aggression. A benign form of aggression or self-assertion results when the assertion of independent will is proportionate to the needs of the individual and is not destructive to the other person.
Destructive and violent aggression results when the assertion of will is exaggerated and disproportionate to the needs of the individual. This disproportion stems from an early struggle of wills between the child and the mother figure; this early struggle, if not fully resolved, becomes a lifelong struggle that sometimes results in the destruction of the opponent (Kutash, 1978).

Modified Analytic Theories of Aggression

Of some interest in relation to the theme of aggression is the holistic approach of Karen Horney (1937). Her modifications of classical psychoanalysis included, among other theoretical differences, a strong emphasis on the cultural influence, a here-and-now focus, and a rejection of Freud's theory of penis envy. Freud's penis envy theory and his famous statement that 'anatomy is destiny' implied that males are more dominant and females more submissive. Horney believed that this phenomenon might be, at least in part, culturally conditioned rather than the result of biological or glandular sexual differences. However, of major importance for the explanation of aggression is Horney's theory of character structure. She divided people into those who 'move away from people', those who 'move toward people', and those who 'move against people'. Aggression of a destructive nature would thus occur in individuals who move against people, whereas more benign self-assertion would be characteristic of persons who move toward people. Those who move away from people would react to threat by a 'flight' reaction rather than a 'fight' reaction. Horney also put forth a hypothesis of 'competition' as a substitute for Freud's libido theory, which she rejected (Kutash, 1978).
The theoretical framework of Harry Stack Sullivan (1953) places primary emphasis on the relationship between the person and other people in his environment as the determinant of personality development: "[Sullivan] holds that, given a biological substrate, the human is the product of the interaction with other human beings, that it is out of the personal and social forces acting upon one from the day of birth that the personality emerges" (C.Thompson, 1950). From this point of view, aggression (whether adaptive or destructive) and violence would result from disturbances (parataxic distortions) in an individual's interpersonal interactions and relationships. However, this theory does not clarify whether any biological predisposition or drives bring about disturbed interpersonal relationships, which in turn create interpersonal or displaced aggression. There is a chicken-egg problem involved: Do disturbed interpersonal relations cause destructive aggression, or does aggressiveness cause disturbance in relationships?
Related to the school of interpersonal relations is the British derivative of psychoanalysis, the object relations theory of personality and character development. Contributions to this theory have been made by Melanie Klein, W.R.D. Fairbairn, and Donald Winnicott (Guntrip, 1973) - all of whom emphasize the individual's early relationships with his mother or with a mothering person. From this point of view, aggression would be explained mainly on the basis of inordinate rage reactions developed in a frustrating relationship with a rejecting or inadequate mothering person (The ego psychologists discussed in a later section of this chapter also put a great deal of emphasis on the mother-child relationship.)

By far the most comprehensive theory of aggression to come from the cultural school of psychoanalysis is the seminal work of Erich Fromm (1973). He distinguishes between benign aggression and malignant aggression. Benign aggression is defensive and 'built in in the animal and human brain' and enables the individual to act 'against threats to vital interests'. Man's 'hyperaggression' or destructive aggression is "not due to a greater aggressive potential but to the fact that aggression-producing conditions are much more frequent for humans than for animals living in their natural habitat". Thus, "man's destructiveness and cruelty cannot be explained in terms of animal heredity or in terms of a destructive instinct, but must be understood on the basis of those factors by which man differs from his animal ancestors. The problem is to examine in what manner and to what degree the specific conditions of human existence are responsible for the quality and intensity of man's lust for killing and torturing". Fromm thus takes a position in direct contradiction to Freud's theory of a 'death instinct' to explain destructive activity in man.
Fromm goes on to detail a number of varieties of what he refers to as 'pseudo-aggression' - "aggressive acts that may cause harm, but are not intended to do so. Among these are 'accidental aggression', 'playful aggression' and 'self-assertive aggression'. He also discusses 'conformist aggression', "acts of aggression that are performed not because the aggressor is driven by the desire to destroy, but [because] he is told to do so and considers it his duty to obey orders", and instrumental aggression', which has the aim of obtaining that which is necessary or desirable. He regards war as the most important example of instrumental aggression. In his view, war is caused not by innate human destructiveness, but by "instrumental aggression of the military and political elites".
Fromm believes that cruelty and destructiveness are manifestations of malignant aggression peculiar to man. He postulates two forms of destructiveness: spontaneous destructiveness and destructiveness that is bound in the character structure. Examples of spontaneous destructiveness are vengeful destructiveness and ecstatic destructiveness. Vengeful destructiveness is a spontaneous reaction to intense and unjustified suffering inflicted upon a person or the members of the group with which he is identified. It differs from normal defensive aggression in two ways: (1) it occurs after the damage has been done and hence is not a defense against a threatening danger; (2) it is of much greater intensity and is often cruel, lustful, and insatiable. Ecstatic destructiveness occurs in ritualistic, primitive orgies or states of trance organized around rage and destructiveness. Examples of cruelty and destructiveness woven into the character structure are sadism and necrophilia. Fromm applies the term 'necrophilia' or 'love of death', to a character-rooted passion which he labels the 'necrophilous character'. In his view, however, the malignant forms of aggression - sadism and necrophilia - are not innate; "they can be substantially reduced when the socioeconomic conditions are replaced by conditions that are favorable to the full development of man's genuine needs and capacities; to the development of human self-activity and man's creative power as its own end".

Ego Psychology and Aggression

Freud formulated his theory of aggression from the point of view of id psychology and made only a small beginning toward integrating it with ego and superego psychology. Like most analysts (for instance, Fenichel, 1945), Hartmann, Kris & Lowenstein (1949) reject Freud's theory of a 'death instinct' as the source of aggression and amplify the theory of aggression by depicting the role of the ego and the superego in controlling, displacing, and/or channeling the aggressive drive. They believe that aggression is an instinctual drive rather than an instinct in the biological sense. Mainly, they explain the difference in terms of the idea that the development of the ego is in part a learning process. Learning thus guarantees the gratification of the inner urges stemming from the instinctual drives. In contrast, the biological instincts can lead to gratification with virtually no learning or a minimum of learning: "Instinct in the ideal case guarantees the survival of the individual at least in lower animals, while in man the guarantee of survival rests with the ego".
Of some interest is the fact that many psychoanalytic theorists and practitioners are more inclined to accept the idea of a diversity of drives subsumed under libido theory. That idea is incorporated, for instance, in Murray's (1938) need-press system, which provided the rationale for such projective techniques as the Thematic Apperception Test (Murray, 1943). Bellak (see Spence, 1967) regards the psychoanalytic dual theory of drives, sexual and aggressive, as "another instance of unnecessary semantic confusion because each drive can be fragmented into many partial drives". For example, libido theory, in Bellak's view, contains such drives as the wish to see, the wish to exhibit, and the like. Hartmann, Kris & Lowenstein describe aggression in terms of impetus, source, aim, and object, just as Freud did for libido. With reference to impetus, they draw a strict parallel with libido, since it deals with the pressure of the drive. They then go on to indicate that activities of all kinds offer opportunities for the discharge of aggression. The absence of activity is in some instances an expression of suppressed aggression. Thus, 'active aggression' refers to the wish to harm, to master, or to destroy an object; 'passivity' refers to the wish to be mastered, harmed, or destroyed - a form of aggression turned inward. It is worth noting that the psychiatric nomenclature includes a diagnosis of passive-aggressive conflict (American Psychiatric Association, 1968).
Bellak and other ego psychologists do not agree that aggression is an independent drive. Bellak remarks, "It would seem that we are not dealing with a specific drive but instead with one aspect of an organismic reaction to environment (ordinarily subsumed under cathexis) which only later becomes more specific by maturation and learning, e.g. specifically aggressive". This view is closer to the position of Fromm and perhaps even to that of the learning theorists like Bandura. The learning theorists and the behaviorists, however, explain aggression as almost entirely a learned or conditioned reaction (Kutash, 1978).
Hartmann, Kris & Lowenstein (1949) enumerate four types of conflict which modify the aims of aggression. One is an instinctual conflict, created when the aggressive and libidinal drives are both vested in the same object. Another is a conflict with reality, which occurs when the object of aggression reacts to the attempts at completion of the aggressive acts in a way that may endanger the aggressor. A structural conflict involving the ego may result from anticipation by the ego of danger to the individual. In this instance the ego is in part already identified with the object, so that the ego may be in opposition to the completion of the aggressive acts. Lastly, there may be a conflict involving the superego, where the moral values are in conflict with the expression of the aggression.
The narcissistic personality, in whom the self is the primary object, has difficulty in acting out his aggression, since the aggressive act may involve danger to the self or ego. The work of Kohut (1971) clarifies this aspect further. Bellak (see Spence, 1967) regards the 'conflict with reality' as "a coexistence of an aggressive drive and learned inhibition related to self-harm". In commenting on the structural conflict involving the ego, Bellak states. "There is a coexistence of an aggressive drive and learned inhibition, the latter being predicated upon apperceptions of the self, superimposed on and integrated with apperceptions of the object". Bellak also refers to the structural conflict involving the superego as "a coexistence of aggressive drive and learned apperceptions concerning culturally acceptable modes of behavior".
The impact of aggression may be modified by four types of processes, according to Hartmann, Kris & Lowenstein: "(1) by displacement of aggression to other objects; (2) by restriction of the aims of the aggressive impulses; (3) by sublimation of aggressive energy; and (4) through the influences of libido mentioned above, one of these influences operating as fusion. These processes are frequently interdependent". Hartmann, Kris & Lowenstein regard displacement as the simplest of the four processes. The displacement of anger from the original target to a less threatening substitute target, or one that is nonlibidinized or inanimate, occurs frequently in human experience. Freud (1930) and later Bibring (1941) pointed out that social conflict is often exploited by political demagogues who manipulate the masses and thus provide each individual member of the society or group with an enemy who is accepted enthusiastically as a target of aggression. Social tension is thus used for the displacement of individual tension (Kutash,1978).
Through sublimation, aggressive energy is 'deinstinctualized' - just as, in Freudian terms, the libido is 'desexualized'. The sublimated libido, according to Freud, contributes to the formation and maintenance of permanent object relations and to the 'molding of psychic structures'; once the psychic structures are formed, the energy is at the disposal of the ego and the superego. Similarly, aggressive energy is 'neutralized' (Hartmann, 1952) and thus can be transferred from the id to the ego. According to Hartmann, Kris & Lowenstein, "the capacity to neutralize large quantities of aggression may constitute one of the criteria of ego strength".
In elaborating the functions of neutralization, Hartmann and associates indicate that not all internalized aggression leads to self-destruction, "just as not all internalized libidinal energy necessarily leads to self-infatuation". On the contrary, neutralized psychic energy supplies the ego and the superego with motor power and equips the ego for its function in action. The internalization of aggression is an essential condition for the formation of the superego. Once the superego is formed, modified aggression is used by the superego in relation to the ego. "Thus, what appears as displacement, restriction or sublimation, considered in relation to the id and to discharge of aggressive energy, is, if we take into account the total personality and its position in social reality, a most important prerequisite of mental integration and of mastery of the environment".
Once the ego is sufficiently cathected and established as a functioning organization, the ego and aggression become more closely linked, since the ego organization normally controls mobility, and the muscular apparatus serves as a medium for the discharge of aggression: "Musculature and motility, apparatuses for the discharge of aggression, contribute decisively to the differentiation between self and environment and, through action, to the differentiation of the environment itself. The environment in turn invites action and determines specific areas of action; it thus offers opportunities for the discharge of particular modes of aggression and their individual modifications".
Whether aggressive energy is utilized in a fight reaction to objective danger or results in a flight reaction depends, among other factors, upon the extent of the perceived danger and the strength of the ego. Aggressive energy not discharged in flight may be internalized. It may be used as cathexis of the superego, generating guilt feelings; or it may be neutralized in the ego without interfering with the individual's integrity. Internalized aggressive energy without neutralization can lead to self-destructive impulses (Kutash, 1978).

Ego Boundaries and Aggressive Behavior
Some of the crucial questions raised by those who have addressed the problem of aggression and its manifestations from the psychodynamic point of view are the following: (1) Under what conditions is aggression acted out directly (that is, expressed in overt behavior)? (2) When does aggression remain on a fantasy level and get expressed symbolically and indirectly? (3) Are there two kinds of aggression, which can be labeled benign and malignant, or does an undifferentiated aggressive drive or potentiality become either benign or malignant, depending on how it is channeled or controlled? (4) What determines the outcome of an aggressive drive: the strength of the drive; the built-in defenses; the reality situation, including the existence of objective danger; the quality of the ego organization with its conscious controls and planning capacity; or the value system and the qualities of the superego? (5) Does psychopathology play an overriding role in producing violence, or is violent behavior a learned or conditioned reaction? (Kutash, 1978).
A great deal of light can be shed on these questions by focusing on the ego and its intactness or deficiencies, as illustrated by the condition of the ego boundaries. The concept of ego boundaries was described by Federn (1952) in relation to the psychoses and developed further by Gutheil (1958), Zucker (1959), Ivey (1959), and Kutash (1963, 1976).
The ego when fully cathected develops two major boundaries, which enable it to carry out its integrative functions. One of these boundaries differentiates the ego from the id; the other differentiates the ego from outer reality. "In the psychologically well-functioning individual, these boundaries are optimally cathected and flexible, so that the ego functions are properly exercised, including suitable repression and selective admittance into consciousness of primordial and instinctual drives from within and adequate reality testing and cognitive, perceptual experiencing of the external world. The major task of the ego is the successful integration of these pleasure drives and needs from within, with the reality considerations and requirements of the external world" (Kutash, 1965). When the ego is thus 'optimally cathected', both boundaries serve as 'semipermeable membranes'; the urges, promptings, and impulses from within (including the impulses toward expression of aggression), as well as the stimuli from the external environment, are allowed to enter consciousness (the ego) only when they serve the goals and aims of the ego. Thus, in the well-functioning individual, the ego boundaries ensure aggression in the service of the ego. Aggression in the service of the ego has meaning over and above what Hartmann (1952) refers to as adaptive aggression, but fits in with his theory of neutralization, which makes the aggressive energy available to the ego. However, it involves more, in that it utilizes the strength of all the ego functions. It also is more than what Fromm (1973) refers to as benign aggression, since it also implies that aggression can be utilized by the well-integrated ego for self-development and self-actualization without harming others. It has some interfaces with Bellak's (see Spence, 1967) explanation of aggression modified by the ego's learned reactions but emphasizes more the exercise of its other perceptual functions (Kutash, 1978).
In terms of perceptual theory, the concept of perceptual flexibility (Ittelson & Kutash, 1961) can be related to optimal ego boundary cathexis. The field of criminal law, in its recognition of criminal intent, also takes into account the ego and its boundaries; that is, a defendant cannot be convicted of a crime unless the prosecutor can prove that the defendant intended consciously to commit it (Branham & Kutash, 1939).
Analytic psychotherapists and psychoanalysts have demonstrated clinically that people have unconscious homicidal, suicidal, and destructive aggressive fantasies and dreams, but the person with a well-functioning set of ego boundaries does not act out these fantasies. In the neurotic personality, one of these ego boundaries may be inauffiently cathected - too permeable; however, the other boundary is overcathected by defenses that prevent the acting out of 'malignant' aggression. When both boundaries are poorly cathected or damaged by excessive conflict and trauma, the individual may then use destructive aggression against others or himself and in some instances, under external eliciting circumstances, resort to bizarre and 'senseless' varieties of violence.
From this point of view, all the ego functions - reality testing, accurate appraisal of objective threat, learning, resolution of conflict, decision making, selective use of defenses such as sublimation and displacement, perception and awareness of internal and external stimuli, and others - are available for dealing with the aggressive drive (Kutash, 1978).
In the healthy functioning ego organization, objective danger or threat from the environment is reacted to by either fear or anger - normal emotions in response to such a thrust. If the situation is perceived as one that the individual cannot cope with directly, such as an armed holdup, then the emotion experienced is mostly fear. If there is a reasonable chance of overcoming the threat, then the individual feels predominantly angry. Fear leads to a flight or avoidance reaction, while anger leads to fight or a confrontation. In the neurotic or the person with insuffient ego strength, reality testing is poor; consequently, such a person may exaggerate, imagine, or misperceive the threat and thereby suffer from anxiety (neurotic fear) or hostility (neurotic anger), which may take a chronic form. In such situations, inappropriate aggression or withdrawal may occur (Kutash, 1978).
Kutash (1978) described the effects of an overcathected, too rigid ego-id boundary, whereby the drives from within are repressed in varying degrees and the libido is dammed up. Such a condition may have resulted from early specific trauma or a chronically traumatic or neurotic early childhood situation, in which the major defense of repression was developed and much of the painful material rendered unconscious. A good share of libidinal or psychic energy becomes tied up in maintaining the repressions, leaving an inadequate cathexis of the ego-outer world boundary, which remains too permeable. As a result, the individual develops characteristic symptoms of the hysterical personality structure - such as extreme suggestibility, histrionic displays of emotion, feelings of unreality and identity problems, and tendencies toward dissociation reactions. At the same time, the rigidity of the inner boundary results in sexual naiveté, 'belle indifférence', lack of drive, and conversion symptoms.
If we now turn to the implications of this type of ego boundary imbalance for the aggressive drive, we must emphasize that the sexual urges and the aggressive promptings are repressed, so that the sexually inhibited person is also passive and lacking in drive. The passivity and the repression of aggression contribute to the shallowness of the affect and to the fact that the 'emotional' displays are histrionic but not fueled with drive from the inner core of the personality. The ego's concern is to protect itself against the emergence into consciousness of forbidden sexual and aggressive urges and fantasies. When such a person becomes involved in aggressive activity or violence, it is only as a follower because of his extreme suggestibility. These individuals may be unduly influenced by violent behavior in television and motion picture presentations and newspaper accounts. They may 'take on' this behavior also because of their identity problems and their need to follow the leader (Kutash, 1978).
In the obsessive-compulsive personality structure, it is the ego-outer world boundary that is too rigid, not suffiently permeable, or overcathected (Kutash, 1976). The individual has erected a barrier between himself and the world around him, bolstered by such character defenses as intellectualization, rationalization, isolation of affect, and compulsions. The inner boundary, by contrast, is too permeable and may be fractured, so that sexual thoughts, homicidal and suicidal ideas, unacceptable fantasies of all types, impulses, and promptings from within continually enter consciousness in the form of obsessions. These are prevented from being acted out in the environment by the relatively inpermeable outer boundary. This neurotic arrangement sets up the typical symptomatology of the obsessive-compulsive reaction, in which "the anxiety is associated with the persistence of the unwanted ideas and of repetitive impulses to perform acts which may be considered morbid by the patient" (American Psychiatric Association, 1968). The reality-testing capacity of the obsessive-compulsive is preserved at the expense of rigid defenses around the ego-outer world boundary. "Preoccupied as he is with warding off the outside world, he is rigid, incapable of absorbing much that impinges upon him from the environment. His views are extremely conservative since, above all, he is anxious to maintain the existing order" (Gutheil, 1958).
In this type of ego organization, homicidal fantasies, suicidal preoccupations, sadistic ideas, and a variety of tabooed or forbidden antisocial thoughts herniate into the ego or consciousness, and the individual obsesses continuously without acting on them. He becomes indecisive and is beset with selfdoubts and uncertainties. The nonadaptive aggression, which could be disastrous if acted out in reality, remains on a fantasy level. Threats from the environment are handled by denial and negation. Such a person will resort to homicide, suicide, or irrationally violent behavior only when the threat becomes overwhelming and undermines the entire security system of the individual, penetrating the ego-outer world boundary. The obsessive-compulsive personality has then decompensated into paranoid schizophrenia (Kutash, 1978).
In the borderline personality organization (Kernberg, 1975) and in ambulatory schizophrenia, both the ego-id boundary (inner boundary) and the ego-outer world boundary (outer boundary) are unevenly cathected or poorly invested with psychological energy, so that there are 'breaks' or 'splits' in both ego boundaries. Under stress and under the influence of eliciting environmental circumstances, such individuals are highly likely to express malignant aggression in sometimes bizarre and irrational forms. In these individuals sexually violent acts such as rape may also occur, as well as perverse sexual expressions. Primary-process material from the primitive levels of the unconscious may be 'acted out' by such individuals in the outer world, resulting in impulsive acts of violence (Kutash, 1978).
The condition of the superego, or censor, also helps to determine whether aggression will be expressed in a malignant or violent manner. The superego represents the incorporation of parental and societal values, the capacity to differentiate what is considered right from what is thought to be wrong. It is partly conscious and partly unconscious. That part of the superego that is repressed into the unconscious operates automatically. When both the ego and the superego are working properly, the superego helps the ego prevent the emergence and acting out of destructive drives and actions that would violate the inherent value system or prove detrimental to the individual by causing extreme guilt and depression or putting him in conflict with the environment and established authority. Thus, defects or insufficiency of the superego may, in cases of psychopathy or criminality, result in malignant aggression or violence. (Cleckley, 1964, has called psychopathy 'the psychosis psychiatry refuses to face').
A theory of aggression must include some explanation of socially sanctioned aggression, such as war that is officially sponsored by the nations involved; in such an event, each nation considers its war machine and soldiers to be fighting a 'just war for human welfare'. What happens to the superego's function in ruling out killing? The great majority of young men who volunteer or are conscripted for service in war can kill under these circumstances because of the social sanction and approval that they get for fighting for their country. A few individualists may hold out as conscientious objectors or pacifists, but a person must have a particularly strong ego to be able to 'take an unpopular stand'. In an unpopular war such as the one in Vietnam, where the social sanction and approval were equivocal, the number of defectors and conscientious objectors was proportionately very high. Thus, in mass aggression the individual superego is superseded by the authority and sanction of the state, and most citizens as individuals do not have a strong enough autonomous ego to stand up against the state (Kutash, 1978).

Psychopathology and Violent Behavior
Destructive aggression occurs with much greater frequency in individuals suffering from various types of severe psychopathology. Kutash (1978) covers briefly some of the more flagrant types of psychogenic psychopathology in relation to violence. Severe psychopathology is marked by poor or depleted ego strength, inadequately cathected ego boundaries, and/or ego decompensation, so that aggression no longer manifests itself under the control of the ego or conscious mind. In these individuals aggression no longer operates in the service of the ego; instead, the ego is at the mercy of the aggressive impulses.
Many acts of uncontrolled violence occur in individuals suffering from manic-depressive psychosis. In the manic phase, severely afflicted individuals may commit violent acts such as homicide, assault, or destruction of property; in the depressive state, there is a danger of suicide or self-mutilation, depending upon the severity of the depression. There is much evidence for a hereditary predisposition in this illness, but there is also a psychodynamic theory based on the work of Margaret Mahler (Mahler, 1968; Mahler, Pine & Bergman, 1975). She has traced the cycloid personality and particularly the manic-depressive psychosis to the period in early childhood when the young child is in conflict between feelings of omnipotence and helplessness. When the child feels omnipotent, he feels that he can do anything; he attempts to act upon this assumption but is restrained or controlled by the mother. In the 'helpless' mood or phase, the child is completely dependent and feels unable to do anything for himself; he demands mother's help and clings to mother. This can be a trying period for a mother to cope with; when it is handled inadequately by an immature, depressed, or negligent mother, the basis is laid for wide mood swings and ego weakness, which permits the emotions to take over.
The varieties of schizophrenia, particularly paranoid schizophrenia, also produce an inordinate number of crimes of violence and aggression, so that this form of mental illness - when adequately diagnosed - has served as a legal defense on the basis of the defendant's not knowing the nature of his act and being unable to distinguish right from wrong.
Other psychopathological conditions that produce an inordinate number of violent acts are states known as 'epileptic equivalence'; transitory psychotic episodes in borderline and narcissistic personalities; and numerous other disorders marked by loss of ego control over the inner promptings, urges, drives, and impulses. All these disorders have in common a weakness in the ego and a gross impairment of such ego functions as reality testing, judgment, channeling of drives, sublimation, inhibition, and neutralization and fusion of libidinal and aggressive drives (Kutash, 1978).

A Comprehensive Theory of Aggression and Violence
The central concept in an integrated psychodynamic theory of aggression is that of ego strength. A well-functioning ego screens out the stimuli, promptings, and urges from the unconscious as well as the stimuli from the environment and the outer world. To accomplish this purpose, the ego has flexible and evenly cathected boundaries. It has achieved functional autonomy and a keen perception of both bodily processes and outer reality. Healthy or benign aggression, better called self-assertion, is aggression in the service of the ego. It is adaptive; that is, it helps the individual achieve a healthy heterosexual adjustment, success in the performance of work and creative activities, and an adequate social life, as well as involvement with the environment and people. It is 'fused' adequately with libido, which tends to give it a positive life-supportive direction. Social learning as an ego function enhances the proper channeling of aggression in acceptable, adaptive, rewarding activities. Benign aggression is thus aggression, self-actualization, or self-assertion under the control of the ego and in harmony with a socially adaptive value system or superego (Kutash, 1978).
By contrast, malignant or pathological aggression occurs in individuals with a weak ego, identity problems, and poor ability to control impulses from within or to appraise the reality situation adequately. The weak ego also has a distorted or poorly developed superego. The varieties of partially impaired or immature and overdependent egos that have a high potential for poor impulse control and overdependent egos that have a high potential for poor impulse control and unrealistic expression of aggression and violence - the so-called borderline and narcissistic personalities - are described in the book edited by Hartocollis (1977) and can be understood by a thorough study of ego psychology (Blanck & Blanck, 1974; see also The Culture of Narcissism by Lasch [1979]). In severely damaged or inadequate egos, as in the psychoses, the potential for malignant aggression and violence is even greater. This theory also encompasses the Adlerian idea that feelings of inadequacy, sibling rivalry, and compensatory masculine strivings may lead to violence, since these are individuals with weak egos. It would also include the acting out of primitive drives, as in Jungian theory. The theories advanced by the 'cultural' schools of Erich Fromm and Karen Horney are also applicable, since individuals at odds with the culture but unable to alter or challenge it effectively often suffer from weak egos, poor reality testing, and impotent rages.
Thus, an integrated psychodynamic theory based on ego malfunctioning, ego weakness, immaturity, or damage can encompass other theories and does not rest on whether one regards aggression as an inborn instinct, a drive, a tendency, or a survival mechanism. It is based primarily on the idea that ego control - which includes good cognitive functioning, adequate contact with reality, positive self-regard, a healthy superego, flexible and evenly cathected boundaries, and a host of other ego functions - modifies the aggressive drive or tendency so that it serves adaptive functions and prevents destructive violent behavior (Kutash, 1978).

Buss (1961) on Freud(ians)

In his earliest writings Freud was preoccupied with libido (sexual energy) and psychosexual development, as he attempted to establish the notion that sexuality (defined broadly) underlay all neurotic conflicts. The emphasis on libido had the effect of relegating aggression to a minor role, aggressive impulses becoming manifest only in relation to the stages of psychosexual development. In the late oral stage the child acquires teeth, and his tendency to bite objects (including the mother's breast) is an expression of oral-sadistic impulses. The peak of aggression is reached during the anal stage, when sadistic urges to hurt and dominate others are notable for their frequency and intensity. Finally, in the Oedipal stage, rivalry with the same-sex parent for the love of the opposite-sex parent leads to death wishes toward the hated rival. In this scheme sadism appears first in the sequence, and only later is masochism seen (as the inversion of outwardly directed aggressive impulses) (Buss, 1961).
In the next phase of Freud's theorizing he explored more fully the question of 'ego instincts', and nonlibidinal urges played an increasingly important role in his formulations. He was no longer as concerned with the source of instincts as with their aims, e.g. the prototype of an ego instinct would be hatred (which has an aim) rather than hunger (which has a source but no particular aim). The general aim of ego instincts was self-preservation, and the major constituent of such instincts was aggression. Aggressive trends were thus transferred from aspects of libido to part of the ego instincts, and Freud believed that aggressive urges could occur in the absence of sexual conflict:
"The ego hates, abhors and pursues with intent to destroy all objects which are for it a source of painful feelings, without taking into account whether they mean to it frustration of sexual satisfaction or gratification of the needs of self-preservation. Indeed, it may be asserted that the true prototypes of the hate relation are derived not from sexual life, but from the struggle of the ego for self-preservation and self-maintenance' (1925, Vol. 4).
This revision of earlier views emphasized the reactive nature of aggressive urges. Their source was not biological, as were sexual urges, but in the self-preservative tendency of the ego to strike back at whatever threatens it or denies it satisfaction. This reactive view was adopted two decades later by the Yale group, who expressed it in more behavioral terms as the frustration-aggression hypothesis (Dollard et al., 1939).
The destructiveness of World War I evidently had a profound effect on Freud's theorizing, and subsequently he gave aggressiveness an even more important role in his theorizing. Freud's final theory of aggression is more balanced in terms of the polarities that characterize psychoanalytic theories; life-death, expression-suppression, tension-increase versus tension-decrease. The novelty in the revision was the postulation of a death instinct, which represents the organism's wish to return to the state of nothingness whence it emerged. This notion is consistent with the 'conservative' nature of all instincts, which orient the organism toward conservation of energy, i.e. a return to a tensionless state. The death instinct is opposed by the life instinct, which consists of both libido and self-preservative tendencies. Both life and death instincts have the aim of reducing tension, the life instinct seeking release mainly from sexual tension and the death instinct seeking release from the tension of simply living. Since all life is a tension state that ends with death, the death instinct predominates ultimately. The organism's wish to return to the passive, tensionless state of death is opposed by the life instinct, whose aim is to maintain life and allow the release of only sexual tension, not all tension.
If the organism did not prevent death instinct from fulfilling its aim, death would soon ensue; and the deaths that ostensibly occur when infants are not fondled are cited by Freudians who adopt this view. Most individuals do not allow this primal self-destructiveness to manifest itself in behavior; self-destructiveness is opposed by the life instinct, which succeeds, at least in part, in turning destructive urges from the self to external objects. The life history of every individual may be construed as a struggle between the life and death instincts, ending only when the life instinct is no longer capable of opposing the death instinct. The stronger the death instinct in a person, the more necessary is it for him to direct aggression outward against objects and people. Whatever aggression is not vented against external objects will be turned back on the self. In this scheme there is a primal masochism (self-destructiveness) that must be directed outward and turned into sadism; to the extent that it is not expressed in sadism, this aggressiveness is turned back on the self in the form of secondary masochism. This is clearly an inversion of the earlier formulation, in which sadism antedated masochism.
In this final formulation, the parallel between sexual instinct and destructive instinct is exact. Sexual instinct derives from life instinct, which also includes self-preservation. In the developmental scheme of sexual instinct, primary narcissism (self-love) is followed by object libido, i.e. love is diverted from the self to external objects. This may be followed in turn by secondary narcissism, a return to self-love, when sexual impulses cannot be completely discharged onto external objects.
Destructive instinct derives from death instinct. In the developmental scheme of destructive instinct, primary masochism is followed by sadism; aggressiveness is diverted from the self to external objects. This may be followed by a return to (secondary) masochism when destructive impulses cannot fully be discharged against external objects. Thus narcissism and masochism follow parallel paths during the developmental sequence.
Freud realized that there is no direct evidence for the existence of a death instinct but he reasoned that the absence of its manifestations was due to its being a 'silent instinct'. Its actions are ostensibly concealed by those of the more obvious and flamboyant life instinct, which blocks and fuses with the death instinct in order to prevent it from fulfilling its aim. The most direct fusion of life and death instincts leads to masochism, the pleasurable experiencing of pain.
Freud was not concerned with behavioral proof of the death instinct because he believed that the existence of any instinct was a matter best reserved for chemists to decide. He predicted that eventually instinct would be placed on a firm foundation of physiology, a belief that is consistent with his biological approach when discussing the fundamentals of behavior (Buss, 1961).

Reactions of Freudians

Freud's theory of the death instinct split psychoanalysts into three camps. One group accepted it fully; one group rejected the death instinct with its metaphysical connotations but accepted aggression as an instinct of equal importance with libido; and the third group rejected the entire revision, retaining Freud's earlier version of aggression as reactive and noninstinctual (Buss, 1961).

Complete Acceptance
Even among those who fully accept death instinct there is room for a difference of opinion, and this group may be divided into the extremists and the moderates on the basis whether they are willing to consider any modifications. The views of Nunberg (1955) may be taken as an example of the extremists. He follows Freud in regarding the death instinct as silent and not striving for objects but is more active than Freud in seeking evidence for it. He regards an infant's gnawing at its fingers or toes as evidence of primal masochism, the earliest manifestation of the death instinct. He distinguishes between this primal masochism and 'actual' masochism, in which the death instinct fuses with narcissistic libido to acquire an erotic tinge. Nunberg also proceeds further than Freud in attempting to specify the 'organs' of aggression; he notes that, just as the genital is the instrument of sexuality, the striped muscles of the extremities and trunk constitute the instrument of aggression.
Nunberg's view of psychosexual development deviates from that of other Freudians mainly in the way he sees the oral period. He assumes that from the earliest period the destructive instincts attempts to protect the ego from all external stimuli. When tension in the erotogenic zones of the mouth disturbs the infantts rest, this tension is projected to the mother's breast, which must be destroyed. The early oral period is marked not only by primal masochism (gnawing on his own fingers and toes) but also by cannibalistic tendencies to incorporate the mother, or at least her breast. In the anal-sadistic period the child acts out wishes to take possession of and destroy objects, using as a means of aggression the muscles of the extremities and trunk. The phallic period is also marked by destructiveness, but now inflicting pain is associated with pleasurable genital sensations, as hatred of objects appears for the first time (Buss, 1961).
Waelder (1956) also accepts death instincts but has more moderate views. He distinguishes between reactive and essential destructiveness, the former arising from three sources. First, there are threats to ambition or self-preservation and frustration of libidinal impulses; second, there are aggressive by-products of the ego's attempts to control its own body and to master the external world; and third, aggression may be a part of libidinal drive, as in incorporation and penetration fantasies. Aggression from these three sources is reactive, and there is no need to postulate a death instinct.
Essential destructiveness, on the other hand, requires postulation of a death instinct because the aggression is too deeply rooted in the individual to be ascribed to any other source. Consider psychotic acts of murder and suicide, and acts of revenge that occur 10 or 20 years after their instigation; Waelder argues that such acts cannot be accounted for by a transient, reactive aggressiveness but must be attributed to an inborn drive to destroy (Buss, 1961).

Partial Acceptance
Most psychoanalysts reject the metaphysical notion of death instinct but accept aggression as an instinct. Loewenstein (1940), one of the leading proponents of this view, distinguishes three aspects of the theory of death instinct:
1. A primary self-destructive instinct
2. The turning outward and projection of self-destructive instinct, which leads to aggression
3. Destructiveness as an independent instinct, opposed to the united sexual and life instincts.
He rejects the first two assumptions but accepts the third, and in collaboration with others (Hartmann. Kris & Loewenstein, 1949) has formulated a theory of aggression as an independent instinct, as follows.
Libido and aggression may be compared as to source, nature of discharge, and aims. Certain organ zones are the source of both stimulation and gratification of libido, but aggression is not zone-specific as to source or gratification, and the zones involved in discharge involve widespread musculature. Aggressive discharge is less structured than libidinal discharge, but there are elements of forepleasure and satisfaction in aggressive behavior as well as in sexual behavior. While the aims of sexuality are diversified, the aims of aggression are rigid and narrow; but the plasticity of aggression may be seen in the variety of means that can lead to satisfaction.
Hartmann et al. reject the formulation that internalized aggression leads to self-destruction because internalized aggression, once it has been neutralized by the ego, constitutes part of the motive power of both the ego and superego. Internalized aggression is necessary for the normal development of the superego, and only in rare instances of psychopathology is the superego self-destructive. The impact of aggression may be modified by four processes: (1) the destructive urge may be displaced from the original 'object' to 'objects' that do not retaliate or are not prized by society; (2) the aims of aggression may be restricted by a scaling down of the intensity of the act, e.g. intent to injure becomes intent to humiliate; (3) aggression may be sublimated and directed into socially acceptable channels, such as fighting disease or crime; and (4) there may be a fusion of aggression with libido, e.g. mastery of man over woman in the sexual act. Although these various means of modifying aggressive urges are at the disposal of the ego, it cannot prevent conflicts over aggression. Hartmann, Kris & Loewenstein outline four kinds of conflict. The first may be labeled instinctual conflict: aggressive drive versus the libido when both are directed toward the same object. The dual impulse to love and destroy the same person must lead to severe conflict and turmoil. Most affectional relationships are marked by ambivalence, but individual members of society fail to understand the dual nature of their feelings toward another. It is considered reprehensible to harbor ill will toward loved ones. Often aggressive urges are repressed, to the accompaniment of considerable anxiety and guilt.
The second kind of conflict is between instinct and reality. The aggressive act would be met with retribution, endangering the aggressor. No one can escape punishment for his aggressive acts, at least not during childhood, and the awareness of punishment for aggression sets up an approach-avoidance conflict.
The third type of conflict is similar to the first type, but it involves only the ego. The ego may identify with the 'object' of destructive urges; the child identifies with the parent against whose domination he is rebelling. Since the ego has an investment in the victim, any attack against the victim is similar to an attack against the self.
The fourth type of conflict involves the superego. When the lessons of civilization ('Thou shalt not...') have been well learned, the inhibitions against aggression are strong. After parental disapproval of aggression has been introjected, there is self-disapproval whenever aggression occurs. The discharge of aggressive drive is held back by the superego, which severely punishes the ego for transgressions. If the conflict becomes severe, thought may be equated with action, and the superego punishes the ego merelyfor the impulses to aggress.
The outcomes of these conflicts depend on the strength and maturity of the ego and superego. An immature ego has weak identifications and is insufficiently aware of reality; consequently there will be little inhibition of aggressive urges. An undeveloped superego has not introjected the values of society, and it fails to inhibit aggressive acting out. On the other hand, the ego may overestimate the danger of possible punishment and place excessive inhibitions on aggressive acts. Similarly, the superego may be too demanding, refusing to allow even the mildest forms of aggression. Finally, in the mature individual there is a balance between instinctual demands for aggression and ego inhibitions against aggression, permitting the expression of modulated aggression along lines that do not endanger the ego and are acceptable to society. When there is a clear danger signal, the normal individual can aggress; but if danger situations become sexualized, the result is likely to be masochism, i.e. danger and pain become a source of sexual stimulation.
In the face of danger, aggressive energy has only two courses, discharge or internalization. Internalized aggression may follow three paths: (1) it may be neutralized by the ego in ways outlined above, with the individual's integrity remaining intact; (2) it may be used by the superego to attack the ego via guilt feelings; and (3) it may become a source of self-destructive impulses. This last path reveals a crucial difference between the doctrine of aggressive instinct and the doctrine of death instinct. The death instinct doctrine states that there is a primal masochism that must somehow be turned outward against others; the aggressive instinct doctrine states that there is a primal aggressiveness against others that may be turned inward under certain circumstances (Buss, 1961).
Building on the formulation of Hartmann et al., Beres (1952) described the varieties of maladaptive behavior that result from faulty fusion of aggression and libido. Oral drive merging with aggression may lead to insatiable demands for food and its psychic equivalent, nurturance, e.g. continued demands for attention and affection (aggressive helplessness). Combined aggressive and anal drive may lead to clutching, possessive love, in which the person must dominate and ensnare the partner; or it may lead to sadism, which is seen early in childhood in tormenting animals and later in the infliction of pain. In combination with phallic drive, aggressive drive may give rise to exhibitionism and competitiveness (Buss, 1961).

Most European psychoanalysts accept the doctrine of death instinct; most American psychoanalysts reject death instinct but accept the notion of aggressive instinct. Some psychoanalysts reject both notions; they adhere to Freud's previous view of aggression as a reaction to frustration. The writings of Saul (1956) serve to illustrate the last position.
Saul notes that there is no evidence that hostility is inherited and immutable, and he emphasizes the importance of events during the early formative years in the genesis of hostility:
"Hostility is a disease of development and has its chief sources within the personality. The distortions which cause it may be in the id (excessive demands, dependence, envy and the like), in the superego (either through hostile imagos which stimulate hostility or through deficiencies and disorders of standards and ideals), or finally in the ego (the highest faculties), insofar as an individual's whole way of thinking and outlook are warped by the persisting emotional effects of unwholesome childhood influences" (Saul, 1956).

Non-Freudian Psychoanalytic Views

Within the psychoanalytic movement there were a number of deviations from Freudian doctrine, the earliest and most important being those of Adler and Jung. During the last few decades there have been numerous modifications of psychoanalytic theory, but of the latter-day deviationists only Horney has written extensively on aggression. Psychoanalytic views are, of course, not limited to psychoanalysts, and there have been systematic contributions from psychologists. The major statement on aggression by a psychoanalytically oriented psychologist is that of Munroe.

Many of Adler's previously untranslated writings were translated by Ansbacher & Ansbacher (1956), and the following account leans heavily on their volume. Adler's early views on aggression date back to 1908, but his approach is similar to that of many present-day psychoanalysts. He saw aggression as a drive (instinct) toward fighting for satisfaction of all needs, a drive not restricted to a single organ or organ system. Aggressive drive dominates all aspects of motor behavior: in pure form aggression is seen in fighting and cruelty; in modified form in sports and war; and when directed against the self, in masochism and exaggerated submissiveness.
Aggressive drive may be directed into a number of channels or modes of expression. If it is expressed in fantasy by an artist or poet, the creation will show cruelty or destructiveness. If it is expressed in terms of an individual's occupation, it may show up in such diverse roles as criminal, revolutionary, or police officer. Or it may undergo cultural transformation and emerge as converse behavior (reaction formation), as in charitableness, sympathy, and altruism.
In normals, aggression is seen predominantly in modified and modulated forms, but in neurotics and psychotics it is more intense and naked. It can be seen directly in temper tantrums, hysteria, epilepsy, and paranoia - all directed against others. Directed against the self, it emerges as hypochondria, hysterical pain, accident neurosis, ideas of reference and persecution, or self-mutilation and suicide. In this early formulation, anxiety was seen as a phase of aggression inverted against the self, occurring only when aggression is suppressed. The inverted aggression (anxiety) may be expressed in the motor system as tremors and in other body systems as blushing, palpitations, sweating, and vomiting (Buss, 1961).
Freud knew about these 1908 views of Adler, and in a paper written in 1909 he stated his reaction:
"Alfred Adler, in a suggestive paper, has recently developed the view that anxiety arises from the suppression of what he calls the 'aggressive instinct', and by a very sweeping synthetic process he ascribes to that instinct the chief part in human events, 'in real life and in the neuroses'... I cannot bring myself to assume the existence of a special aggressive instinct alongside of the familiar instincts of self-preservation and of sex, and on an equal footing with them. It appears to me that Adler has mistakenly hypostatized into a special instinct what is in reality a universal and indispensable attribute of all instincts and impulses - their 'impulsive' and dynamic character, what might be described as their capacity for initiating motion. Nothing would remain of the other instincts but their relation to an aim, for their relation to the means of reaching that aim would have been taken over from them by the 'aggressive instinct'. In spite of all the uncertainty and obscurity of our theory of instincts I should prefer for the present to adhere to the usual view, which leaves each instinct its own power of becoming aggressive" (1925, Vol. 3).
Many years later Freud was forced to recant his strict opposition to aggression as an instinct, but he never agreed with Adler. When Freud elevated aggression to an equal position with libido, it was in the form of death instinct that could be turned outward toward external objects. He refused to the end to acquiesce to Adler's views on aggression, despite their partial similarity to his own. History has many strange quirks. While moat Freudians have rejected Freud's doctrine of death instinct, they have accepted the notion of aggression as an instinct; but it is clear from both Adler's 1908 paper and Freud's reply, that the idea of an instinct of aggression and the ways in which it is manifested are Adlerian. Adler tended to play down the role of aggression in his later formulations. Aggression was no longer considered a drive (instinct), but a partly conscious, partly irrational reactive tendency toward overcoming obstacles and everyday life tasks. It was seen as subordinate to a general striving for superiority and power, aggression being a pathological form of the more general tendency to 'overcome'.
Thus the paths of Adler and Freud crossed, going in opposite directions. Adler's formulation of aggreesion as a basic (biological) instinct yielded to a later view of aggression as reactive and noninstinctual and of relatively minor importance. Freud started out by assigning aggression to a minor role; then promoted it to a major ego instinct; and finally gave it equal statue with libido as an instinct (death instinct). It changed from a reactive tendency to a basic, biological tendency of the individual (Buss, 1961).

Jung did not spell out his views on aggression in detail, and they may be stated briefly. He grouped life and death instincts into a single drive, libido. This single, all-encompassing drive becomes differentiated in its manifestations into polar opposites, e.g., life-death, love-hate. Since libido is dynamic, when one aspect does not operate, its opposite must function: when it cannot create, it must destroy. When an undifferentiated function sinks to a lower level in the unconscious, it necessarily becomes destructive.
Note the contrast to Freud's doctrine of death instinct. Freud opposed death instinct with life instinct, and the various personality processes reflected the fusion and conflicts between these polar opposites. Jung started out with a single instinct, out of which emerged the polar opposites of life and death instincts (Buss, 1961).

Karen Horney
Horney (1939, 1945) was explicit in rejecting both death and aggressive instincts. She attributed aggressiveness and hostility to the individual's response to 'basic anxiety'. Basic anxiety supplies the core motivation for all tendencies; briefly it is "... the feeling a child has of being isolated and helpless in a potentially hostile world" (1945). The response to this feeling may crystallize into one of three patterns: moving toward, moving against, and moving away from people. In the normal person all three approaches are possible, and the individual is capable of shifting his approach as conditions demand it. The neurotic, however, is stuck in one or the other of these three molds and is incapable of shifting. The individual who moves against people is an aggressive personality. He takes for granted the hostile nature of the world and reacts by fighting; people cannot be trusted, and in the jungle of life one must strike first. Beyond fighting for his own protection, he is also strongly motivated toward revenge, which is closely related to sadism. As might be expected, Horney rejects sexuality as the basis for sadism. She traces the roots of sadism to hopelesness and futility. The miserable person, unhappy with his own lot and jealous of others' happiness, seeks to impart suffering to others. His own self-loathing must be projected onto others, and by hating them, he has less need to hate himself: "The degrading of others not only allays his intolerable self-contempt but by the same token gives himself a feeling of superiority" (1945). His torture of others not only projects his self-hatred but also lifts him from self-negating futility by allowing him to step on those he humiliates. The sadistic individual thus lives aggressively and destructively, but he achieves meaning for his life only vicariously, i.e., through the suffering he causes.
Horney makes hostility an all-pervasive tendency rooted in rejection. The child wishes only to be cared for and loved, hopes only that the world will be kind. Since the world is not a comfortable, warm place and because some rejection and hostility are inevitable, there must be at least some hatred in everyone. What is not made clear is the basis for some individuals tending to move against others, while others tend to move away. In the face of rejection and hostility, moving against people is no more probable than moving away from them, and Horney's failure to state the conditions for the development of one or the other tendency is a weakness in her theorizing (Buss, 1961).

Ruth Munroe
Munroe's (1955) views may be described as eclectic psychoanalytic. While she is predominantly Freudian, her position reflects other psychoanalytic views, and, in part, those of modern psychology. She notes that several different kinds of behavior are lumped under the heading of aggression and attempts to distinguish between them. Some aspects of aggression, she argues, stem from the ordinary activities of the individual as he makes his way in the world. While they might be interpreted as aggression, they are more likely "... essentially neutral behavior, or the by-product of the effort to establish an effective idea of the self" (1955).
Another aspect of aggression is the emotional response to danger or frustration, namely rage. Unlike sex or hunger, rage is a response to an external stimulus, rather than a tension arising from within the organism. Yet the antecedents of rage are so universal in early childhood, and the rage response is so much more than an appropriate reaction to stimuli that it has the same drive qualities as sex and hunger. Thus Munroe comes close to an instinctual view of aggression, without becoming involved in the theoretical complexities of instinct doctrine (Buss, 1961).

Fisher & Hinds
Psychoanalytic theory makes extensive use of the hydraulic analogy, with implications of deep level tensions exerting pressure on surface structures. The analogy is spelled out in some detail by Fisher & Hinds (1951), who compare normal with pathological control over aggressive impulses. They distinguish between deep level hostility, which is entirely unconscious, and peripheral hostility, which is under more conscious control. In the normal personality, surface and deep level controls over aggression function as follows. Surface controls operate primarily to discharge anger outward and secondarily to turn anger inward; the number of outward channels should exceed the number of inward channels. Outward channels function cooperatively, so that one channel is directing hostility outward, the others are also operating. On the other hand, the relationship between outward-directing and inward-directing channels is reciprocal: the more hostility directed outward, the less directed inward.
Normal deep level control involves a distinction between hostile urges associated with guilt and hostile urges associated with assertiveness (which do not arouse anxiety). The adjusted person does not confuse these two kinds of aggressive urges. Deep level assertive urges are allowed expression, but anxiety-laden aggression inhibited. The inhibition of disturbing hostile urges is not complete because there is at least minimal drainage in order to prevent the build-up of too much tension due to unexpected deep level urges. When such urges do accumulate, there is an increase of peripheral hostility directed outward and a decrease of peripheral self-blame, both of which allow sufficient drainage of deep level impulses (like water spilling over a dam).
Maladjustment tends to disrupt the functioning of this complex system of controls. Mild maladjustment affects only peripheral controls, leaving deep level controls unaffected; deep level controls must be left free to operate in an organized fashion, and, fortunately, they show the greatest resistance to disturbance. As the personality disturbance becomes more severe, deep level controls become more disorganized. In the extreme of maladjustment, psychosis, the major effect is on deep level controls, surface controls remaining virtually unaffected. The psychotic cannot distinguish between realistic aggressive urges and those laden with guilt or anxiety. All aggressive urges are inhibited, and too many are directed inward by peripheral controls. The subsequent tension serves to increase confusion and disorganization, creating a spiral of ever-increasing turmoil. Surface controls cannot channel enough of the deep level hostile impulses, and in persons with suicidal tendencies the resulting tension may spark an attempt at self-destruction (Buss, 1961).

Recent Developments
Recent developments in psychoanalytic theory have been summarized in Yakeley & Meloy (2012), on which the following account is based.

Klein (1946) understood aggression to be instinctual in origin and primarily destructive; Winnicott (1971) rejected the concept of the death instinct and differentiated between normal aggression as a constructive and essential ingredient of normal development necessary for separation and individuation, and pathological aggression as a reaction to environmental trauma and loss [cf. Fromm]. Hartmann and the ego psychologists' (Hartman, Kris & Lowenstein, 1949) position on aggression concurred with Klein that it was an instinctual drive, but differed in their rejection of the death instinct. The tendency of some of these theorists’ followers to simplify these arguments, however, has not done justice to the complexity of Freud's views of aggression which he himself never fully resolved, and has led to false dichotomies and the fallacy that there is only one type of aggression, and correspondingly one type of violence (Yakeley & Meloy, 2012: 231).

Early contributors
One of the earliest psychoanalysts who wrote extensively about aggression, but also documented his clinical experience with many cases of real and often severe violence, was Menninger (1938, 1942, 1968). Menninger agreed with Klein that there were very early unconscious murderous fantasies, but he also stressed the ego’s homeostatic properties, which, under conditions of stress, will attempt to discharge the aggressive drive against the environment in the service of self-preservation. He was also one of the first psychoanalysts to propose that all violent acts contain meaning, and that murder may be committed to preserve the sanity of the individual in an attempt to stave off psychic disintegration (Menninger, 1963; Perelberg, 1999a,b) (Yakeley & Meloy, 2012: 231).
Glover is another important contributor to the early psychoanalytic literature on violence. In his seminal work The Roots of Crime (1960), an investigation into psychopathic and delinquent behavior inspired by his clinical work at the Portman Clinic, he distinguished between unmodified or primary aggression, which he believed was fused with libidinal energy and included sadism; and reactive aggression, which was intimately linked to hate and anxiety, and is activated by frustration and other psychic dangers. He traced the development of antisocial behavior from the infantile period, and was keen to emphasize that both constitutional and environmental factors, including ‘traumatic stimulation’ and the ‘broken home’, were at play. His work suggested that there are different forms of aggression; and he also distinguished between more temporary ‘functional’ aggressive reactions to anxiety, and more permanent structural modifications of the ego and superego by mechanisms such as unconscious identification with the aggressor (A. Freud, 1936) that could result in psychopathy (Meloy, 1988a,b) (Yakeley & Meloy, 2012: 231).

Post-Kleinian developments
Menninger's idea that violence can be a defense against psychotic breakdown and unbearable states of mind is one which has been developed by later analysts working with violent patients, notably those working in the Kleinian tradition. These analysts highlight the primitive defenses, particularly projection and projective identification, that are mobilized in the violent act. These defenses work together with a constellation of object relations, fantasies, and impulses to form pathological defensive organizations (Hyatt-Williams, 1998; O'Shaughnessy, 1981; Steiner, 1982) that function to ward off psychic pain and unbearable states of mind. This causes a splitting of the personality so that areas of the mind containing affects which may stimulate actual violence and primitive object constellations remain encapsulated and underdeveloped (Yakeley & Meloy, 2012: 231).
Hyatt-Williams (1998), in recounting and systematizing his extensive work with murderers in UK prisons in the 1960s and 70s, described how these individuals’ minds are dominated by persecutory anxieties they are unable to tolerate, and which they attempt to expel via projective identification. However, this defensive maneuver is ineffective, in that once projected, the victim or recipient of the projections may act in an aggressive way towards the projecting person, provoking him to violent attack. He proposed that an essential feature of murderous acts is the collapse of symbolic thinking, releasing previously encapsulated unmetabolized death experiences, which he calls ‘the death constellation’ (Hyatt-Williams, 1998), which overwhelm the mind and have to be completely projected in homicide, or introjected in suicide (Yakeley & Meloy, 2012: 231).
Sohn, who has worked in high secure forensic psychiatric hospitals in the UK since the 1960s, also emphasizes failures in symbolic functioning in violent offenders. Sohn (1995) described working with psychotic patients who have carried out apparently unprovoked violent attacks on strangers, in which projection and projective identification fail altogether. He proposed that in these patients projective identification cannot be used to rid the self of frightening states of mind; instead, these states of mind are experienced as concrete mental objects that have to be kept in the mind to avoid the worse dread — a completely empty mind. Sohn suggested that these patients’ aggressive instincts, and lack of a maternal object into which they could project their feelings, interfered with their capacity to symbolize, including the symbolization of ‘loss,’ so that they were left with the terror of a mind devoid of content. This lack of symbol formation and inability to project means that the only way of ridding the mind of these intolerable feelings is by violent physical action (Yakeley & Meloy, 2012: 231).
These analysts, influenced by the theories of Klein and Bion, highlighted the defects in ego-defensive functioning, as well as the presence of primitive unconscious fantasies of death, destruction, and annihilation, which they believed dominated the minds of the violent individuals with whom they worked. They assumed, as Klein (1946), that innate envy and destructiveness were manifestations of the death instinct and predominated in early life, giving rise to primitive anxieties and defenses, unconscious fantasies and an archaic superego. Bion’s (1959, 1962) concept of ‘containment’ was also important for these writers in emphasizing the relative lack of emotional receptivity that violent individuals experienced in infancy from their maternal objects; their destructive impulses and fantasies could not be adequately contained, modulated or represented, and therefore remained in a primitive and toxic form (Yakeley & Meloy, 2012: 231-32).

The maternal object and focus on pre-oedipal pathology
Other psychoanalysts working with violent patients have focused more on the role of the environment, in particular the maternal object, in the genesis of aggression and destructive behavior. Winnicott (1971), in his work with disturbed children, posited a central role for the mother in the etiology of violent and antisocial behavior. He saw aggression as a creative force necessary for healthy development by enabling individuation and separation, processes which were dependent on the ‘facilitating maternal environment’ and adequate maternal ‘holding’ functions. He believed that pathological aggression arose as a reaction to early deprivation and trauma (Winnicott, 1956, 1986), but viewed antisocial behavior as a hopeful phenomenon indicative of some early positive experiences, interpreting the young person’s ‘antisocial tendencies’ of anger, resentment and violence as an attempt to regain the lost object. Winnicott believed that object relatedness superseded aggression as an instinctual drive (Yakeley & Meloy, 2012: 232).
Another important contributor to the field is Glasser, who also worked for many years at the Portman Clinic in London, and whose ideas remain influential today. We will highlight two of Glasser’s contributions that we consider important both theoretically and clinically: his proposal that there is more than one type of violence, and his concept of the core complex.
Glasser’s (1998) understanding of violence concurred with Menninger’s; he believed violence could be evoked by any threat to psychic homeostasis. Returning to Freud’s economic model of the mind, Glasser asserted that a fundamental task of the ego is the protection and integration of all psychological systems, including narcissistic equilibrium, and to maintain a steady dynamic balance at optimal levels. He proposed that there were two distinct modes of violence, ‘self-preservative violence’ and ‘sado-masochistic violence,’ which differ according to the role of the object. Self-preservative violence was considered a primitive response triggered by any threat to the physical or psychological self. Such threats might be external, and could include attacks on a person’s self-esteem, frustration, humiliation or an insult to an ideal to which the person is attached. The person could also feel threatened by internal sources such as feeling attacked by a sadistic superego or fearing a loss of identity by feelings of disintegration and internal confusion. The violent response is fundamental, immediate, and aimed at eliminating the source of danger, which may be an external object or an attack on the person’s own body — the latter experienced as an external persecutor in self-harm or suicide. Such self-harming behavior can be considered as an extreme regression to a primitive level of functioning where parts of the body are related to as external objects to be destroyed (Yakeley & Meloy, 2012: 232).
Glasser proposed that his second form of violence, sadomasochistic, resulted from the sexualization of self-preservative violence. The modes also differed in their relationship to the object. In self-preservative violence, the object at the time of violence is perceived as an immediate danger, but holds no other personal significance and its emotional responses as a whole object are of no interest — it just needs to be eliminated. By contrast, in sadomasochistic violence, the responses of the object are crucial: the object must be seen to suffer, but to do so it must be preserved, rather than eliminated as in the case of self-preservative violence. Sadomasochistic violence also involves pleasure, which is not a component of self-preservative violence, where anxiety is always present (Yakeley & Meloy, 2012: 232).

The core complex
Glasser (1996), like Winnicott (1971) a generation earlier, considered the relationship to the maternal object as being fundamental in the genesis of aggression and violence. He described a particular constellation of interrelated feelings, ideas and attitudes that he called the ‘core complex’. A major component of the core complex is a deep-seated and pervasive longing for an intense and intimate closeness to another person, a wish to merge with them, a ‘state of oneness’ or a ‘blissful union’. Such longings occur in all of us, but where there has been an early pathologically narcissistic relationship to the mother they persist in a primitive form unmodified by later stages of development. Merging no longer has the quality of a temporary state but is feared as a permanent loss of self, a disappearance of the person's existence as a separate, independent individual into the object. To defend against this annihilatory anxiety, the person retreats from the object to a safe distance. But this flight brings with it feelings of emotional isolation, abandonment, and low self-esteem. These painful feelings prompt, in turn, the desire for contact and union with the object once more, so that the core complex assumes the qualities of a vicious circle (Yakeley & Meloy, 2012: 232).
Aggression is a central component of the core complex. The annihilatory fear of a loss of separate existence provokes an intense aggressive reaction on the part of the ego. In order to preserve the self, the object has to be destroyed. But this, of course, would mean the loss of all the goodness of the mother, and the security, love and warmth that she offers. The infant has only two options — to retreat into a narcissistic state or to resort to self-preservative aggression against the obliterating mother. This, Glasser proposes, is the origin of self-preservative violence. Glasser also formulates that in the perversions there is an attempt to resolve the vicious circle of the core complex by the sexualization of aggression, and conversion of aggression into sadism. The mother is now preserved and no longer threatened by total destruction, as the intention to destroy is converted into a wish to hurt and control. This can lead to sado-masochistic rather than self-preservative violence (Yakeley & Meloy, 2012: 232)..
We can see how Glasser, with his concept of the core complex, locates the origins of violence in the pathological early relationship between mother and infant, where violence is used by the child as a measure to create space and separation between himself and his experience of an overwhelming maternal object. Although weaving an elegant and complex theory, Glasser commits a reductive fallacy by focusing almost exclusively on the maternal object. Other psychoanalytic authors did the same by fixating on the role of the maternal object in creating a pathological internal object world in which pre-oedipal object relations predominate and predispose to destructive aggression and violent enactment. Shengold (1989, 1991) coined the term ‘soul murder’ to describe the process whereby the mother overstimulates her child and views him as an extension of herself rather than a separate being, replacing the child’s own development of self by his mother's idealizations; this theoretically leads to a sense of emptiness and disillusion when he inevitably discovers that the world is not such an ideal place. Shengold believed that such individuals who were subjected to the violence of psychotic or psychopathic parents developed primitive ‘anal–narcissistic defenses’ and were likely to become perpetrators themselves due to the power of the repetition compulsion and identification with the aggressor (Yakeley & Meloy, 2012: 232-33).

The role of the superego and third object
This leads to the role of the father or paternal object in the genesis of aggression and violence, and associated deficits in the formation of the superego in violent and psychopathic individuals. In his paper ‘Some character-types met with in psycho-analytic work,’ Freud (1916) described ‘criminals from a sense of guilt’, individuals who were drawn to committing antisocial deeds because they were forbidden, to relieve a pre-existing unconscious sense of guilt stemming from fantasies of oedipal transgression. This stimulated a series of classical psychoanalytic clinical reports about people who displayed antisocial and delinquent behavior. Alexander wrote several papers investigating both the biogenic and psychogenic origins of the psychopath (Alexander, 1923, 1930, 1935), while during the same period Aichhorn (1935) postulated that psychopathy was a failure of early identifications, narcissistic aberration, and oedipal pathology in his seminal text Wayward Youth. At the end of World War II, Horney (1945), Fenichel (1945), Reich (1945), Greenacre (1945), and Friedlander (1945) all published works in the same year which examined the unconscious life of the psychopath, highlighting the failures of the superego, deficits in early identifications, and early disturbed parent–child relations. Later analysts emphasized not only the developmental deficits evident in the mind of the psychopath, but also considered psychopathy as a defensive structure. Bowlby (1944) coined the term ‘affectionless psychopaths’ for children whose apparent indifference to others concealed their terror of ‘the risk of their hearts being broken again.’ This formulation was very similar to Winnicott's ideas (1956) regarding antisocial or psychopathic behavior as behavioral markers of an unconscious attempt to regain a good object (Yakeley & Meloy, 2012: 233).
Meloy (1988a,b, 1992), expanding on earlier work by Jacobson (1964) and Kernberg (1984), described the pathological identifications and failures of internalization that affect the development of the superego in psychopathic individuals. He also elaborated upon the presence of a particular model of the ‘grandiose self’ — one which is predatory by nature — an ego structure that is the primary identification of the psychopath, and which determines much of his interpersonal and affective functioning. It is maintained by continual behavioral denigration of others, often through the use of planned and purposeful violence (Meloy, 1988a, 2001).
Other analysts have emphasized the essential role of the father in the development of violence (e.g., Bateman, 1999; Campbell, 1999; Fonagy & Target, 1995; Limentani, 1991; Perelberg, 1999a; Stoller, 1979), as well as the role of the third object in the creation of internal space (Yakeley & Meloy, 2012: 233).

Contemporary approaches

Different types of violence
Although the concept of the death instinct remains contentious, it is difficult to maintain that aggression has no instinctual origin at all. Attempts to formulate aggression in purely psychological terms ignore crucial evidence from other disciplines, including the ethological, genetic and biological sciences. A few psychoanalytic thinkers have, therefore, attempted to integrate the instinctual and reactive models of aggression into a unified theory or postulated different forms of aggression. Rage has been isolated as a specific form of aggression that can express itself as explosive violence (e.g. Hyatt-Williams, 1998; Kernberg, 1992; Kohut, 1972; Kutash, 1978; Menninger, 1963; Parens, 1993; Meloy, 1988a,b, 1992). Both Kernberg (1992) and Kohut (1972) distinguish between different types of aggression. Following other authors (Parens, 1993; Schafer, 1997; Shengold, 1991), who have highlighted the role of the painful affects involved in narcissistic injury in precipitating rage-type aggression, Gilligan (1996) has proposed that all forms of violence are precipitated by feelings of shame and humiliation. Working as a psychotherapist with violent high-security inmates in prisons in the United States, he was struck by how often, when asked why the offender had been violent, they replied that they had been ‘disrespected’. He believes that their early experiences of being rejected, abused or made to feel as if they did not exist, predisposed them as adults to be sensitized to feeling ostracized, bullied or ignored, leading to unbearable feelings of shame and humiliation which need to be defended against by violent means. Such anecdotal data based upon self-report in a population of individuals known to be chronically deceptive, however, is not a firm empirical basis for such a broad theoretical formulation (Yakeley & Meloy, 2012: 233-34).
Cartwright (2002), in his investigation of rage-type murder, described people who appear to be successful, stable, perhaps sometimes rather controlling, but non-aggressive individuals who, seemingly out of the blue, commit explosive violence or even murder. Drawing on Winnicott's (1960) concepts of the ‘true’ and ‘false’ selves, Cartwright proposed that such individuals have a defensive organization that he calls the ‘narcissistic exoskeleton’, consisting of a rigid split in the psyche between a constellation of idealized object relations which form a defensive field around a disowned internal world of bad objects associated with aggression. Rage-type violence can suddenly erupt when there is collapse in this defensive system, precipitated by conflict or threats experienced by the person in relation to separation or antagonism toward the victim, which may have begun long before the murderous event. This causes a breakdown in their narcissistic idealized defensive system that functions to deny and control the existence of the hidden aggressive bad object system, which then intrudes into and overwhelms the mind and personality, cannot be subject to thought given its poor representational capacity, and is therefore evacuated physically in violent action. Cartwright’s work also appears to draw heavily on theories concerning intermittent explosive disorder first advanced by Menninger (1963) and Monroe (1989) (Yakeley & Meloy, 2012: 234).

Bimodal distribution of violence
The recognition that there are different types of aggression leads to the plausible assumption, backed by empirical evidence, that there are correspondingly different types of violence. What has largely been missing, however, among both psychoanalytic theorists and practitioners is an appreciation of the heterogeneity of violent acts, empirically studied and analytically elaborated upon in the forensic literature (see, for example, Meloy, 1988a,b, 1997, 2006; Revitch & Schlesinger, 1981). Most salient is the absence of an understanding that certain violent acts are neither anxiety nor affect-based, are usually related to specific character pathologies, have relatively distinctive neurobiological underpinnings, and deserve psychoanalytic understanding (Yakeley & Meloy, 2012: 234).
During the same period that Glasser was formulating his ideas concerning self-preservative and sadistic aggression with colleagues at the Portman Clinic, Meloy (1988a,b, 1992) applied object relations theory and attachment theory to a bimodal model of violence that originated in the animal physiology literature a half century earlier (see reviews by McEllistrem, 2004; Weinshenker & Siegel, 2002). Born in the observation that cats, mice, and other mammals appeared to engage in two behaviorally distinct modes of violence, affective defense and predatory attack, this research carried forward through the second half of the twentieth century and empirically measured the relatively distinctive neuroanatomical pathways and neurochemicals which biologically defined the two modes. Behavioral manifestations of affective violence included the presence of an immediate threat, elevated heart rate, elevated blood pressure, other indices of autonomic arousal, the presence of intense fear or anger, and the cessation of such violence within seconds or minutes once the threat was no longer perceived (Meloy, 2000, 2006). Predatory violence was behaviorally manifest in the absence of emotion and autonomic arousal, and a purpose and goal direction that was not dependent on physiological arousal (Meloy, 2006). Both modes of violence were measurable in humans through observation and self-report (Meloy, 2006), and also apparent in physiological (Stanford et al., 2003), neuropsychological (Houston, Stanford, Villemarette-Pittman, Conklin & Helfritz, 2003), psychopharmacological (Barratt, Stanford, Felthous & Kent, 1997; Eichelman, 1988), and neuroimaging (Raine et al., 1998) studies. There is a growing body of evidence that the evolved human capacity for both predatory and affective violence lives on in the species (Yakeley & Meloy, 2012: 234).
Psychopathic individuals, moreover, have been found to actually engage in more frequent predatory and affective violence than others, including non-psychopathic criminals (Cornell et al., 1996). Predatory and affective violence have also been studied in children and adolescents; in the former group, these constructs are labeled proactive and reactive violence, respectively (Crick & Dodge, 1996). Children who engage in proactive violence are typically less anxious, more callous, and more severely conducted disordered when diagnosed utilizing DSM criteria. The evolutionary basis of affective violence is survival against an imminent threat. The evolutionary basis of predatory violence is hunting (food gathering),which also promotes survival and genetic reproduction (Yakeley & Meloy, 2012: 234).
Attachment theory associated with predatory violence among personality disordered individuals suggests a generally insecure attachment, and a more specific dismissive attachment (Bartholomew, 1994). Meloy (1988a) hypothesized a primary identification as a predator within the grandiose self-structure of the psychopathic character, an antisocial variation of the positively affectively valenced fusion of the ideal self and the idealized object originally described by Kernberg in his work on the grandiose self-structure of the narcissistic personality disorder (Kernberg, 1975). Other research supports this formulation. Certain character disorders, such as psychopathic individuals, engage in prey–predator relationships as their predominant mode of relating to others in the absence of any reciprocal or affectional ties. They define their relations to others on the basis of power gradients — dominance and submission — rather than reciprocity. This is often clinically apparent in their behavior over time in relation to others, but also suggested by a historical observation of such individuals, their predatory stare, that has been observed but still not scientifically understood, for decades (e.g., Meloy, 1988a,b). There is also new evidence that psychopathic individuals have a ‘predatory acuity’ absent in normal — their ability to discern more accurately fear and distress on the faces of others (Book, Quinsey & Langford, 2007; Wheeler, Book & Costello, 2009) — which may be a product of their limbic disconnection (Meloy, 1988a,b) and paralimbic abnormalities shown in functional magnetic resonance neuroimaging research (Kiehl, 2006). Clarity of observation is more easily done when emotional states and other anxieties are absent from the mind (Yakeley & Meloy, 2012: 234).
The point of convergence between Glasser (1998) and Meloy (1988a,b) is the inherent dominance and submission within sadism. The sadist most desires submission and control, and the predator also operates within the paradigm of submission and control. However, we view the sadomasochistic violence defined by Glasser as a sexualized type of predation, wherein there are many other reasons for predation among humans that do not involve sexual arousal: territorial dominance, revenge, money, and power being four that are most common. Most predation is not sadomasochistic, but when it is, we agree that the object, and the object’s reactions, become a more intimate and personal focus of the predatory individual. Affective violence (Meloy, 1988a,b) and Glasser’s self-preservative violence (1998) are virtually synonymous (Yakeley & Meloy, 2012: 235).

The violent true believer
A variation of such predatory violence is the extreme aggression of the violent true believer, a subject who knows he is entitled to take his own life and the lives of others to advance a particular religious or political belief system, and intends to do so (Meloy, 2004, 2011). This is not the valueless violence of the psychopathic character, but a violence driven by a predominant and rigid superego that often carries the subject’s conscious beliefs, and unconscious identifications, as an agent of a spiritual or religious authority. Such sanctioning may stimulate conscious terrestrial fantasies that are overtly violent and grandiose, e.g., the establishment of a world theocracy, and derivative sexual fantasies and behaviors, such as the disavowal of sexual gratification in immediate reality while entertaining fantasies of unlimited virginal supplies in the afterlife. The homicidal superego demands are often manifest in several accelerating and evolving mental states. Meloy (2004) commented on this in his analysis of Timothy McVeigh and Muhamed Atta, violent true believers who separately carried out large civilian massacres in the United States in 1995 and 2001, respectively: “Within this identification arose a totalitarian state of mind in which omnipotence was idealized, intolerance of difference was magnified, hatred was exemplified, paranoia was rampant, and the entitlement to kill those who do not believe was embraced” (Meloy, 2004, p. 144) (Yakeley & Meloy, 2012: 235).

Attachment and mentalization
Others, such as De Zulueta (1994), Gilligan (1996) and Fonagy (Fonagy, 2003, 2004; Fonagy & Target, 1995; Fonagy, Steele, Steele, Moran & Higgit, 1991; Fonagy, Moran & Target, 1993), emphasize the role of attachment in the etiology of violence. Fonagy believes that aggression is biologically rooted, but that pathological aggression and violent behavior arise in response to perceived threats to the psychological self. He has developed a model of how early trauma and disrupted attachments involving physical and emotional abuse may lead to aggression and violence, with the concept of ‘mentalization’ (Fonagy, Gergely, Jurist & Target, 2002) (Yakeley & Meloy, 2012: 235).
Fonagy and Bateman propose that certain violent people with complex personality pathology, particularly those who would fulfill diagnostic criteria for antisocial personality disorder, have experienced significant trauma and disruptions to their attachment system in childhood, which has interfered with their neurobiological development and the development of psychological defenses, compromising their mentalization capacity and lowering their threshold for emotional reactivity. Empirical evidence to support the hypothesis that antisocial personality disorder is a developmental disorder rooted in attachment comes from studies showing abnormal attachment patterns in forensic patients and prisoners diagnosed with ASPD (Frodi, Dernevik, Sepa, Philipson & Bragesio, 2001; Levinson & Fonagy, 2004; Van IJzendoorn et al., 1997). These individuals are dependent as adults on relationships with others into whom they can project alien aspects of themselves in order to stabilize their minds and feel a sense of self-coherence. Their interpersonal relationships tend to be rigid, hierarchical and controlling, and notions of recognition and respect assume a special importance. When such relationships are challenged by threats to the person’s self-worth or ‘respect’, as described by Gilligan (1996) above, or when, for example, their partner refuses to be the recipient of malign projections, the return of the alien self threatens the person’s fragile stability of mind, leading to unbearable feelings of shame and humiliation that cannot be managed by representational means within the mind. They are now experienced very concretely as feelings that the person needs to expel in violent action in an attempt to regain control and a sense of integrity. In summary, Fonagy and Bateman propose that reactive violence is caused by failures in mentalization, and that being able to mentalize protects against violent behavior (Yakeley & Meloy, 2012: 235).
Fonagy and his colleagues, however, do not address predatory violence, but focus on violence that is trauma-based and a product of socialization abuse or neglect by the primary objects. This results in a partial theory, which may explain why some individuals are affectively or reactively violent, but it does not address the capacity of other individuals to engage in violence that is planned, purposeful, and emotionless — predatory violence — that is not a behavioral product of emotional reactivity or an unconscious wish to maintain the illusion of the self. The field of attachment research has tended to underscore early parenting deficits as the cause of attachment pathology; however, an equally plausible hypothesis is that certain individuals do not have the neurobiological capacity for attachment despite adequate parenting efforts. Attachment pathology may, in turn, be one measurable consequence of the predominance of neurobiological influences that predispose habitual violence. Empirical work, largely outside of psychoanalytic efforts, supports these assumptions (e.g., Raine, Venables & Williams, 1990; Raine, 1993) (Yakeley & Meloy, 2012: 235-36).

Unconscious phantasy
Unconscious phantasy has also been posited as a contributor to the origins of violence. Menninger (1963) and Biven (1977) linked violence to castrating or mutilating phantasies against the parents or others. Kernberg (1984), Shengold (1989), and Sohn (1995) believed that phantasies of being engulfed or attacked by the maternal object underlie the type of self-preservative violence with its core complex dynamics described by Glasser (1998). The pathologically intrusive early symbiotic relationship with the mother and experience of her as overwhelming and obliterating have influenced others such as Bateman (1999), Campbell (1999), and Fonagy & Target (1995), who see violence as stemming from phantasied attacks on the mother’s body. Such work has been empirically supported in research on sexual homicide (Meloy, 2000), and has been theoretically linked to the psychoanalytic work on perversions and symbiotic anxiety by Stoller (1972, 1974) and Greenson (1968). Hyatt-Williams (1998) emphasized the phantasies of annihilation and evacuation that underlie violent acts aimed at defending against indigestible psychic experiences of trauma or obliteration. He believed that murder only occurs concretely after it has been committed many times in phantasy, often unconscious, that has never entered the person’s conscious awareness. Empirically proving a negative — in this case, the presence as a motivation of a consciously absent thought — however, is virtually impossible, and the homicide research would suggest that a wide variety of fantasies occur, and do not occur, among those who intentionally kill (Meloy, 1992; Revitch & Schlesinger, 1981) (Yakeley & Meloy, 2012: 236).
Perelberg has been foremost in drawing our attention to the function of unconscious phantasy in the genesis on violence. One of the few psychoanalysts to have treated violent patients with intensive psychoanalysis, Perelberg (1995, 1999a,b) proposed that violent acts always have underlying specific phantasies or unconscious narratives that motivate them. She noted that in Freud’s writings there are many associations between his ideas on violence and the mythology of the origins of humankind, evident in unconscious phantasies of the primal scene or the Oedipus complex. Perelberg extends Freud’s thinking to linking violence to a core phantasy of a violent primal scene that involves an engulfing and also violent pre-oedipal mother. The violent act is therefore an attack on the mother's body, the mother being experienced in phantasy as not only being in possession of the child’s body, but also the child’s intellectual and affective experiences. The function of violence here is survival in a pre-oedipal world dominated by an obliterating maternal object with no conception of a paternal object present (Yakeley & Meloy, 2012: 236).
Cartwright (2002) raises important questions, however, regarding the place and function of phantasy. He warns, and we agree, that it would be simplistic to assume that one particular group of phantasies could explain all violent acts. He makes an important Kleinian distinction between ‘phantasy,’ referring to deep unconscious structures of the mind from which thought and behavior emanate, and ‘fantasy’ referring to more conscious surface mentation — fantasizing or day-dreaming — that has a sublimatory function. It appears that Perelberg and others, once again, are offering a partial theory by only referring to affective or self-preservative defensive violence that is motivated by unconscious phantasies of the primal scene and maternal object. However, sadistic and/or psychopathic violence is very often preceded by conscious violent fantasies, sometimes sexual in content, that the perpetrator may rehearse many times before committing the violent act (Yakeley & Meloy, 2012: 236).

One of the clearest threads that has emerged in understanding the heterogeneity of violence through psychoanalytic thinking — and supported by empirical data — is the existence of primarily two modes of violence, affective (reactive, impulsive, emotional, hot blooded, self-preservative) and predatory (instrumental, cold blooded, premeditated), with relatively distinctive behavioral manifestations and neurobiological underpinnings. Nevertheless, predatory violence, although long recognized as a mode of mammalian aggression, has received little attention from psychoanalysts despite its ubiquity and ancestral adaptation in our species (Yakeley & Meloy, 2012: 237).