Not long ago, the tendency was to explain the individual as completely the product of society. More recently, this tendency has been reversed: the individual is seen as creating society in his own image, and many social scientists have been led to explain both the individual and his society on the basis of specific infant disciplines—in particular, methods of feeding and toilet training. Harold Orlansky here analyzes the claims of this new method of explaining both individual personality patterns and national characters.
_____________
In recent years there has grown up a new theory of human behavior, based principally on certain aspects of psychoanalytic theory. The “oral-anal” view of personality and culture, as it may be called, holds that the infant’s early experiences in feeding and toilet training determine his adult character, and that this character, in turn, determines the nature of his culture, since most adults in any culture have received similar training during childhood. On the basis of this theory, an increasing number of pediatricians have been instructing mothers in methods of infant care, and some anthropologists have been conducting investigations of foreign cultures.
The tenor of these investigations may, perhaps, be suggested by an earlier study—Geoffrey Gorer’s analysis of Japanese culture. It attributes the “overwhelming brutality and sadism of the Japanese at war,” their type of ethics, their famous Tea Ceremony, and their landscape gardening to the early bowel training of Japanese infants. Gorer has also analyzed the “national character” of Andean Indians, Russians, Burmese, and Americans in terms of their mechanisms of infant rearing—how long and how often infants are breast-fed, when toilet training is instituted, and whether or not the infant is swaddled or cradled after birth. Some social scientists, following in his footsteps, seem engaged in the creation of a new science of history in which war, Nazism, Stalinism, free enterprise, and other manifestations of “national character” are explained by such infant disciplines.
Considering the theory’s popularity and importance, one must regret that it has so far been subjected to little intensive criticism. My own analysis is here restricted to the effect upon personality of nursing and sphincter disciplines in infancy—i.e., the first year or two of life, excluding other aspects of infant experience and the entire world of later childhood. It should also be stressed at the outset that the oral-anal theory in its extreme form (which attributes personality formation to a few mechanical features of infant care) is held only by a small group of writers most of whom are mentioned in this article. (Other students of personality and culture—Harry Stack Sullivan and Erich Fromm, for example—explain the formation of character in terms of the total social situation in which children are reared; with this approach the writer is in agreement.)
_____________
Let us first present the core of the theory in Somewhat greater detail, by quoting one of its proponents, Lawrence Frank:
. . .it appears that the infant who has had generous breast feeding and a benevolent weaning will face life with a benign, generous and optimistic attitude or disposition, while the infant who has been denied adequate nursing and mothering, and has been abruptly weaned will feel deprived, suspicious and fearful and bear resentment that may be crystallized into active hostility and aggression.
Again it appears that the child who has been trained in continence of urine and faeces slowly and without pressure or punishment will yield control over his elimination without anxiety or resentment, learning to release without conflict; but a precocious or harshly coercive training . . . will set up resistance . . . and focus the child’s behavior upon acquisitive or compensatory outlets for the denial of possession of his own eliminations. (Psychiatry, 1939, 2: page 22.)
It will be clear to anyone acquainted with psychoanalysis that this view stems directly from the “oral-anal-genital” characterology initially formulated by Freud and later elaborated by Karl Abraham, Edward Glover, and Ernest Jones. According to this analysis, the infant whose oral eroticism (pleasure in feeding, especially through satisfaction of the sucking function) is gratified by adequate nursing tends to develop an optimistic, confident adult personality; whereas inadequate nursing produces a sadistic or pessimistic adult who remains basically dependent upon others, attempting in all his important relationships, by force or supplication, to obtain again the love of which he was deprived during infancy. The infant whose anal eroticism (pleasure in free bowel movement) is frustrated by early and severe sphincter training will tend in his adult life to exhibit exaggerated and compulsive traits of cleanliness, orderliness, obstinacy, frugality, punctuality, etc. Where the infant is permitted to complete his “natural” cycle of growth, he develops beyond the stages of oral and anal eroticism into a normal or “genital” character.
Now psychoanalytic literature affords considerable documentation as to the existence of these three character types among adults—though there is a good deal of ambiguity in definition, especially with the “oral” and “genital” characters. But it offers almost no empirical evidence on the origin of these character-types in infancy; clearly, adult reconstructions of infantile experience derived from the analytic couch do not constitute empirical findings on the infantile situation. Melanie Klein and Anna Freud have added something to our direct knowledge of infant psychology, but they have generally been more interested in applying psychoanalytic theory than in testing it. We must turn to the literature of medicine, psychology, and anthropology for scientific data which, though deficient in many ways, can provide at least a partial test.
According to the theory, infants nursed at the breast for a long time and weaned late should be less frustrated than infants weaned early or those bottle-fed on a strict feeding schedule, and early bowel training should be more frustrating than late. Do the objective facts verify these hypotheses?
_____________
Two London physicians, B. C. F. and C. H. Rogerson, studied og9 children who had been infants in their clinic seven years previously; those children who had been artifically fed in infancy showed a higher frequency of feeding and sleeping problems, bed-wetting, fear, nervousness, etc., than those who had been breast-fed. (Journal of Mental Science, 1939, 85: pages 1163-82.) This investigation made no effort to equate the groups studied for factors other than method of feeding during infancy. The physical and psychological difficulties found among the artificially fed children can be accepted as accurate for these children, but the inference that artificial feeding was the cause of these difficulties is not warranted; indeed, the authors themselves expressly point this out. (To the social scientist, who almost daily encounters hypotheses of fundamental and universal import “verified” by the investigation of anywhere from ten to a hundred accidentally assembled “cases,” there is always something refreshing about a study that makes no final claims.)
Our suggestion that the difficulties of the bottle-fed children in the above study were probably due to factors other than the manner of feeding is supported by another study which comes somewhat closer to the scientific ideal of controlled experiment. At the Stanford University clinic, Drs. Harold Faber and Leonard Sutton compared forty-two babies who had been breast-fed less than six weeks and thereafter fed from bottles, with an equal number of babies who had been breast-fed for an average of forty weeks. In other respects than the type of milk, the diets of both groups were alike; twenty-six members of each group were paired for weight at birth, and all babies received excellent care. It was found that the breast-fed babies gained more weight than the bottle-fed babies during the first quarter-year, but that this relationship was reversed during the rest of the year. (American Journal of Diseases of Children, 1930, 40: pages 1163-76.) The significance of these findings lies in the fact that weight, physical maturation, and bodily health are generally accepted as indices of psychological development in early infancy.
In the light of present data, it does not appear possible to establish the desirability of breast or bottle feeding without consideration of the entire home environment and the psychological condition of the mother. Where nursing conditions are hectic and the mother’s health poor, as is frequently the situation in the lower-class home, bottle-feeding is certainly preferable, though the norm for lower-class families in England and America probably entails breast feeding in a greater proportion of cases than among middle-class families. Similarly, a middle-class mother who disliked breast-feeding would be ill-advised to institute it, or the child might pay the psychological penalty for the pediatrician’s obstinacy. I have seen no evidence that breast-feeding is inevitably more advantageous to the child’s emotional development than artificial feeding.
_____________
At least five studies have examined the possibility of a correlation between the duration of breast feeding in infancy and the psychological characteristics of the child or adult. All findings I have read agree that no direct correlation can be observed between the duration of breast feeding and any major aspect of personality; this is the only definitive conclusion that present evidence seems to permit. On the basis of two studies (Journal of the American Medical Association, 1929, 92: pages 615-19; and American Journal of Orthopsychiatry, 1931,: pages 284-91), one might presume that a medial duration of breast feeding is more likely than either brief or prolonged nursing to be associated with “normal” personality development in our society, but this hypothesis is flatly contradicted by another study which found the highest degree of emotional security in adults “who were breastfed little or not at all and . . . [in] those who were fed at the breast for over a year” (Journal of Abnormal and Social Psychology, 1946, 41: page 84). Leaving aside questions as to the validity of the personality tests and the methodological procedures employed, it is clear that extreme caution is necessary in making any generalization applying to large populations from the very inadequate samples upon which all of these studies were based.
_____________
The widespread opinion in modem pediatric and psychiatric circles is that “self-demand” feeding—in which the infant is nursed whenever it cries—is more likely to produce a confident and unneurotic personality than is scheduled feeding. As O. H. Mowrer and Clyde Kluckhohn have expressed this thesis, “The parent who responds to a clock rather than to the behavior of the child is, from the child’s point of view, not responsive at all. . . . An apathetic or an anxious or a hostile individual is likely to result [from scheduled feeding].” (Mc V. Hunt, ed., Personality and the Behavior Disorders, I, page 90.)
Clearly, such statements must be regarded as mere hypotheses, and there is evident in them a great deal of reading-in of adult emotion into the infant situation. Only a few years ago the “best scientific opinion” favored scheduled feeding to promote healthy personality development in the child. Certainly, even very young infants quickly learn to adjust their sleeping and eating habits to a schedule of regular feeding, and it would seem that this imposed regularity might contribute to a sense of security and confidence just as much as to a feeling of apathy or hostility.
I know of only two studies that have subjected the self-demand hypothesis to empirical test, and the published account of these studies is such a specimen of social science inadequacies that it is worth quoting at some length:
Miss Trainham . . . reported on a study begun at the Merrill-Palmer school. Two groups of children were chosen, one group on self-regulating feeding regimes, and the other not self-regulated. The parents of both groups were quite similar. . . . One group consisted of ten subjects and the other twelve. . . . The results suggested in a very striking way that the self-regulated children are in ail categories of the tests more advanced in their development. . . .
Dr. Escalona had conducted similar studies . . . and had arrived at opposite results. She, too, had only worked with small groups. One group consisted of babies on self-regulated feeding regimes, and the other group consisted of babies who were on the conventional rigid feeding schedules. The group of babies who were more frustrated [i.e., those fed on time schedules] . . . showed more rapid maturation [read, “were more advanced in their development”—H. O.] by test scores. Until she had heard Miss Trainham’s report, she had explained her findings by the theory that the frustrated babies are forced to recognize the environment earlier. (M. J. Senn, ed., Problems of Early Infancy, page 33.)
It is easy to manufacture a theory to “explain” a particular empirical finding. To explain all the relevant findings is not so easy, and I have no simple solution to offer here.
Means of measuring personality at present available are not very adequate; our knowledge of its genesis is even less adequate. But surely we may ask of the social scientist that minimum of historical and cultural sophistication which would enable him to give thought to the position that the human groups he studies occupy in the real social world to which they—and the social scientist—belong (as contrasted to the position the social scientist ascribes to them in the artificial world of his “control” laboratory). The scientist’s observations might then at least have the validity of descriptive statements about a designated portion of the real world; ultimately, after enough such statements had been accumulated, a fairly complete description of the real social world might be obtained. To generalize about psychological or social laws on the basis of limited observations of small, unrepresentative groups, can lead only to such predicaments as that of the unfortunate Miss Trainham and Dr. Escalona.
_____________
In summary, it must be conceded that social scientists—because of a general lack of historical sophistication, the difficulty of isolating single factors for study, and the difficulty of establishing the validity of the personality measurements employed—have failed to produce a definitive answer to the question of the relation between nursing disciplines and character development. Accepting the sparse experimental findings as tentative markers, however, it appears reasonable to reject the thesis that specific nursing disciplines have a specific and unvarying psychological effect upon the child. Instead, the hypothesis may be proposed that the emotional effect of a particular discipline will vary in accordance with the parental attitudes associated with its administration, the organic constitution of the child, and the entire social situation of which the discipline is a part. Erich Fromm has made a similar formulation: “For an infant who has confidence in the unconditional love of his mother, the sudden interruption of breastfeeding will not have any grave characterological consequences; the infant who experiences a lack of reliability in the mother’s love may acquire ‘oral’ traits even though the feeding process went on without any particular disturbances” (Escape from Freedom, page 293). In short, personality is to be viewed not as the product of instinctual infantile erotic drives mechanically channeled by parental disciplines, but rather as the product of a complex interaction of a unique organism undergoing maturation and a unique socio-cultural environment.
_____________
The anal erotic, anal sadistic, or compulsive character is the most clearly-drawn picture in Freud’s album of characterology. So widespread is the anal character in Western society that Geza Róheim has argued that the entire culture is “based on sublimations or reaction formations of anal trends” (Essays Presented to C. G. Seligman, page 283). However, when one looks for empirical verification of the influence of sphincter training upon personality, the data are not very imposing.
Mabel Huschka, after studying 213 problem children at a New York hospital, concludes that children who were subjected to “coercive” bowel training manifested “undesirable” responses—such as constipation, fear, and rage—more frequently than children who were “adequately” trained; but no attempt is made to correlate these responses with any permanent effects upon personality, and no normal children were used for comparison with the problem children (Psychosomatic Medicine, 1942, 4: pages 301-8). In a similar study, the same author secured comparable data on the reaction of 215 problem children to the institution of bladder training (ibid., 1943, 5: pp. 254-65). G. V. Hamilton, comparing a group of adults who recalled constipation or other anal traits in childhood with a group who did not, found some evidence in favor of Freud’s theory of the anal origin of such adult personality traits as stinginess, sadism, masochism, and concern for clothes (A Research in Marriage, pages 467-72).
These inadequate studies are all that I have been able to discover in the psychological literature. Ardent believers in psychoanalysis have shown a tendency to regard these and even less substantial data as confirmatory of their hypotheses, but a reading of the original studies does not incline an impartial student to such optimism. Thus one writer, reviewing a series of studies by Margaret Fries, states that the direct influence of bowel training on personality “was proved beyond doubt.” Dr. Fries herself, however, had explicitly drawn a contrary inference from her work, stating that “any attempt to correlate age at which habit training was started with later personality traits, without consideration of the mother’s personality type, would involve a large factor of error.” (Psychoanalytic Study of the Child, 1946, 2: page 92.)
In opposition to the orthodox Freudian’s emphasis upon rigid bowel training as the cause of the anal character, it is possible to explain all elements of the anal character—obedience, punctuality, cleanliness, hoarding, etc.—and early bowel training as well—as traits inculcated in the child by middle-class parents in capitalist society: we recognize this pattern of character traits as being historically associated with the Protestant virtues that were economically so well rewarded during the rise of capitalism. The origin of the “anal” character can be understood as an adjustment of the individual, on the level of personality, to specific economic and social conditions. The contrary contention, advanced by Róheim, that these very economic and social conditions are the results of severe anal training, is historically and logically unsatisfactory: was money invented so that we might hoard it, or clocks so that we might be compulsively punctual? (One is reminded of the dialectic of Voltaire’s Dr. Pangloss: the nose was made to hold spectacles; therefore we have spectacles.)
Otto Fenichel, outstanding among psychoanalysts for his recognition of the importance of social conditions in the development of personality, argues persuasively against Réheim’s position on just this point. But even Fenichel holds that “anal eroticism produces the desire to collect something,” although he adds that “what is collected is determined by reality” (Psychoanalytic Quarterly, 1938, 7: page 89). To speak of a libidinal “desire to collect something” as existing apart from the demands of a specific cultural situation is, however, to invoke a factor for which no independent evidence can be advanced. The more economical theory would appear to be to explain “anal” traits solely in terms of the cultural situation which gives rise to them.
_____________
If the confirmatory evidence is so weak, what are the reasons for the popularity of the oral-anal theory?
The pediatrician cannot change the nature of the child’s home environment or the personality of its parents, but he can modify the type of discipline to which it is subjected. As Arnold Gesell and Frances Ilg have remarked, “In America the adoption and duration of breast feeding depends to a surprising degree on more or less fortuitous cultural influences. We know of a pediatrician with a wide metropolitan practice who . . . has induced three-fourths of his maternal clients to adopt breast feeding in preference to bottle feeding. In many other instances this ratio is reversed” (Infant and Child in the Culture of Today, page 86). One may surmise that acceptance of the oral-anal theory gives some pediatricians a gratifying, if illusory, sense of power.
Again, the liberal psychiatrist, keenly aware of the prevalence of psychological difficulties in our society, may seize upon the oral-anal theory as offering a simple mechanism for the prevention of neurosis and a “safe” program of social reform—safe in the sense that it does not involve difficult and controversial political action. (And what efficacious program of reform is ever personally safe?) Illustrative of this tendency is the Cornelian Comer, a national organization of psychiatrists which energetically propagates the view that “a psychologically healthy race of Americans could be produced by . . . consistently loving care to the pre-conscious child,” and that “through the scientific application of correct mothering . . . a new form of mature collectivization could be established.” (See Psychiatric Quarterly, 1946, 20: pages 603-9.)
In anthropology, similarly, the oral-anal theory offers an attractively simple solution to complex cultural problems. In the work of writers like Réheim, Gorer, Weston La Barre, and Erik Erikson, the psychology of infant discipline performs the role of monistic simplification once filled by racial or geographical explanations of culture. Thus we find a long list of cultural phenomena—trade, gardening, the plow, domestication of cattle, money, etc.—explained by one or another of these authors as originating in infantile erotic experience.
The psychological interpreters of culture would profit by reading Leslie White’s article on “Culturological vs. Psychological Interpretations of Human Behavior” (American Sociological Review, Dec. 1947). White points out that psychological “explanations” of cultural phenomena—such as that war results from man’s “warlike” propensities, mother-in-law avoidance from a “horror of incest,” or a taboo on eating pork from an innate “loathing” of pigs-are virtually meaningless, since they do not explain why the phenomena occur in some times and places but not in others. If the psychological factor is “innate” or “universal,” it can obviously not account for variable events; an adequate explanation can only be given in cultural or historical terms; indeed, analysis of the cultural situation often shows that presumed psychological “causes” are rather the effects of this very situation—thus it is closer to the truth to say that war produces “warlike” qualities, and that Jews loathe pork because their culture places a taboo upon it, rather than the reverse.
_____________
The main strength of the oral-anal theory, However, as has been indicated, derives from its psychoanalytic roots, and before passing ultimate judgment on the theory one must evaluate these roots. It is obviously beyond my province to attempt any criticism of psychoanalytic theory as a whole, but it should be useful to summarize, all too briefly, certain aspects of the psychoanalytic view of infancy upon which the oral-anal theory of personality rests, and to criticize this view in the light of the empirical evidence.
- The infant is generally regarded as a miniature adult who experiences adult emotions of anger, fear, anxiety, and love in response to maternal treatment. “When a child comes into this world, he comes in lonely and he is afraid,” one psychoanalyst writes. The psychoanalyst interprets infant experience in terms of what it would mean to him if he were an infant. But this view is intensely colored by the cultural bias of the adult and fails to acknowledge the fact that no infant shares this bias—i.e., infants are culturally “neutral.” Thus, when Freud states that little children “cannot help conceiving the sexual act as a kind of maltreating or overpowering, that is, it impresses them in a sadistic sense,” he may be giving us insight into the Viennese, but not the infantile mind; and when Otto Rank writes of the “trauma of birth” he is anthropomorphizing and not observing the animal-fetal state. Direct observation reveals only that the newborn infant is an immature organism, mainly capable of generalized responses, in which a stable conditioned reflex cannot readily be established; indeed, many child psychologists have abandoned use of the word “emotion” to describe the physical states of this organism during its first three months.
- What is frustrating to the adult is conceived to be frustrating to the infant, and aggression (direct or disguised) is considered the automatic and invariable consequence of frustration. But both halves of this formula are very questionable. The evidence indicates that what is frustrating to a person is determined by his previous experience, and the infant has not had the adult’s experience. Thus, restraint of motion is frustrating to an adult, but it is not necessarily frustrating to an infant who has been strapped into a cradle from birth and has never experienced freedom of motion. Nor is aggression a necessary consequence of infantile frustration, as David M. Levy has observed: “There are any number of frustrations that do not evoke aggressive response in the sense of discharging hostility against a social object or its surrogates. There are, for example, a number of experiments in which animals are frustrated and in which such aggression does not occur. The sucking frustrations in infancy cause finger sucking or sucking other objects . . . rather than increased aggression. There is no proof that the so-called weaning traumas of infancy cause more aggression or even more phantasies of hostility . . . than in other children. The same may be said of all those frustrations that have to do with bowel and bladder control.” (Newcomb and Hartley, eds., Readings in Social Psychology, page 264.)
- Infancy (the first year or two of life) is considered the period in which personality is irrevocably structured; the initial mold into which the infant libido is cast cannot be recast by later experience. This conclusion derives from Freud’s biological and instinctual orientation, and one may say it represents a mechanical rather than a dialectical view of personality development. But the personality of the adult cannot legitimately be regarded as a mere projection of the infant’s personality. Although the experience of later years often reinforces the character structure laid down in infancy, this is not invariably the case. Thus, Navaho infants receive ideally indulgent and unfrustrating care according to the oral-anal theory, with unlimited breast feeding and no interference with free bowel movements; nevertheless, they grow into anxiety-ridden adults because of the nature of their post-infant experience. As Lois Murphy has written, “a satisfying infancy does not necessarily compensate for economic deprivation in the next ten years, and there is increasing evidence that later gratification may go far toward offsetting the effects of early frustration” (Hunt, ed., Personality and the Behavior Disorders, Vol. II: page 657).
_____________
Are we, then, to assert that infant disciplines bear no relation whatsoever to the development of personality? Obviously not. I can do no better here than quote an admirable statement by Gerald Pearson, summarizing his conclusions from a study of personality development in seventy-two children: “There is no question that the physical situations of a child’s life bear a definite relation to the formulation of his personality, but the number of associated etiological factors renders it impossible to consider any single situation as causative. This study does indicate, however, the marked importance of parental attitudes. . . . The child forms most of his ideas about the world of humanity from his impressions of his parents, and their attitude to him . . . must affect the formation of his personality to a far greater degree than the length of [labor at] his birth or the duration of his breast feeding. The events of a child’s life enhance and crystallize the parental attitude toward him . . . and the child reacts to the parental attitude through its association with the events of his life rather than to the events themselves.” (American Journal of Orthopsychiatry, 1931, 1: pages 289-90.)
But how are “attitudes” communicated to the infant if not by physical behavior?
Obviously, emotion must be communicated to the child by bodily actions—but this does not mean that some simple prescription for parental behavior can automatically create a beneficent atmosphere for child rearing. In the last analysis, the oral-anal theory is an attempt to make things simple, reducing the complex problem of infant care to a few clear “duties” easily defined by pediatricians and easily carried out by parents. But simplification distorts reality. The infant’s awareness of those who care for him is surely far more complex than the oral-anal theory would have it.
Some students write of the infant’s sensing the mother’s attitude to it by a process of “empathy.” This concept is dangerously similar to Margarethe Ribble’s semi-mystical concept of “mothering,” but it may have objective validity when reduced to certain motor indicators by which the infant learns to distinguish emotions which the mother feels as anxiety, anger, or love. After the first six months of life, the infant becomes increasingly sensitive to forms of parental emotional expression—as Margaret Fries has expressed it, he is bathed in “a twenty-four hour emotional atmosphere.” And it is by this atmosphere—the entire mother-family-society environment of infancy and childhood—that the personality is presumably shaped, rather than by the mechanical disciplines of infancy. Thus, the consistency with which a discipline is enforced by the mother may be a more important factor than the nature of the particular discipline itself; for example, the infant may be able to adjust contentedly to a wide range of feeding regimes, but a stable emotional adjustment would be difficult if his regime were repeatedly altered.
From this point of view, the efforts of an organization like the Cornelian Corner to reform American society by having mothers breast-feed their infants are misguided. And similar efforts to produce decisive alterations in German and Japanese society by changing certain techniques of infant discipline would be equally ineffective. A much wider range of childhood experience—not to mention the surrounding adult environment—would have to be changed before any significant change in “national character” or culture could be effected.
No theory of personality formation can be simple and yet fit all the facts. Lest we slip into another oversimplified theory that would explain the genesis of personality solely in terms of the personality of parents or the social environment of the home, it would be well to bear in mind the following genealogy which Thomas Fuller has derived from a well-known journal:
Roboan begat Abia; that is a bad father begat a bad son; Abia begat Asa; that is a bad father begat a good son; Asa begat Josephat; that is a good father begat a good son; Josephat begat Joam; that is a good father begat a bad son.
_____________