The Massacre of the Innocents

In the year 1974 one does not have to be a historian to sense that things are getting worse. Indeed, for the first time it is possible to see the face of Doomsday peering out just around the next corner, whether it takes the form of World Government by Universal Torture, or The Bomb, or Demographic Explosion, or just Running Out of Everything. But now comes Mr. deMause with some good news at last. With all the fervor and conviction of a nineteenth-century believer in the March of Progress, he tells us that in one aspect of human life, which he is confident is the key to all others, things have been getting better and better for the last two thousand years: this is the way we treat our children. He demonstrates this in Table 3—a graph of childhood felicity in which the curve turns upward after 1300, sharply upward after 1700, and is now rising almost vertically under the influence of books like A. S. Neill’s The Free Child and R. D. Laing’s The Politics of the Family.

DeMause’s basic thesis is set out in the first paragraph of his book: “The history of childhood is a nightmare from which we have only recently begun to awaken. The further back in history one goes, the lower the level of child care, and the more likely children are to be killed, abandoned, beaten, terrorized, and sexually abused.” Not so very long ago I rashly offered a hostage to fortune by writing that “Freudian psychology has not been much use to the historian, who is usually unable to penetrate the bedroom, the bathroom, or the nursery. If Freud is right, and this is where the action is, there is not much that the historian can do about it.” In his editorial foreword to the first number of the challenging new journal The History of Childhood Quarterly, of which he is the editor, deMause used this quotation as a whipping boy, asserting that on the contrary, “an entirely new field of history is about to be born.”

Although three quarters of a century have passed since Freud first drew attention to the crucial effect of childhood events in determining adult character and behavior patterns, it was not until the 1950s that there appeared the first general history of childhood in the West. Of the four significant studies of this phenomenon, all have been written by non-historians, persons who with respect to the profession are marginal men.

In 1955 J. H. van den Berg, a Dutch psychologist, published Metabletica, of Leer der veranderingen (The Changing Nature of Man), a bold, overarching, psychohistorical study of parental relations to children, based mainly on philosophical sources like Rousseau. In 1960 Philippe Ariès, a French director of publications at the Institute for Applied Research for Tropical and Subtropical Fruits, published his now famous Centuries of Childhood. In 1970 David Hunt, an American psychologist working with disturbed children, reinterpreted some of Ariès’s French seventeenth-century material in Parents and Children in History, using a modified Eriksonian model of ego-development. And now Lloyd deMause, an American academic dropout, a successful businessman and self-taught psychohistorian, has produced a collective volume, The History of Childhood, the key essay in which is his own lengthy general survey of “The Evolution of Childhood” ranging from the Greeks and Romans to the present day.

The first problem in studying the history of childhood is how to choose the appropriate psychological model. Nothing in the historical record disproves Freud’s theory about how at different stages of infantile development different erogenous zones become the foci of sexual stimulation, thus providing a logical explanation for the later relationship between oral, anal, and genital pleasure. Nor does the historical record do anything to belittle the importance of sublimation, or of the unconscious operating with a secret dynamic of its own. What it does do, however, is to cast very great doubt upon the assumption that the particular kinds of infantile traumas upon which Freud laid so much stress have been suffered by the whole of the human race at all times and in all places. It is now fairly clear that four of the main traumas Freud looked for and found among his patients, and therefore assumed to be universal, are dependent on particular experiences which did not happen to the vast majority of people in most of the recorded past, but which were peculiar to middle-class urban culture of late Victorian Europe. Provided that it was carried out slowly, as it apparently was in many cases, the oral trauma of weaning can hardly have been a serious one when it occurred as late as fifteen to eighteen months after birth. The anal trauma of toilet training is unlikely to have existed in a population which lived amid its own excrement, which hardly ever washed, and whose women and children wore no underpants.

The only detailed historical example of toilet training in the past that we have is, unfortunately, that of a less than typical person, a future king, the young Louis XIII in the early years of the seventeenth century. His training apparently did not begin until he was sixteen months or so, and was not internalized before about three years. He at any rate cannot have been traumatized by pressure to control his sphincters at an early age. We just do not know about how other children were toilet trained, but there is the strong negative evidence that contemporary child-rearing manuals do not even discuss the matter.

Although children in the past, as we shall see, had to endure far worse things, the passage through the oral and anal stages of childhood, in the purely technical sense, does not seem to have been particularly traumatic. As for the genital stage, the one example we have—again that of Louis XIII—suggests that no one was bothered by infantile and childhood sexual autoeroticism, sexual display, or sexual curiosity. Louis could get courtiers to kiss his penis, and was allowed to poke his little fist up the vaginas of several ladies-in-waiting. We also know that most families slept in one room, while even if they did not, houses were poorly constructed with thin board partitions through which it was easy to see or hear—as Fanny Hill soon found out. Consequently, from a very early age most children must have witnessed their parents and others—to say nothing of animals—engaged in sexual intercourse.

There is also the negative evidence that childhood and adolescent masturbation was not regarded as a mortal sin in the eighteenth-century Catholic confession manuals (although it was in the Middle Ages), and that although the paranoid drive to suppress all hints of autoeroticism began in 1710, it did not catch on before the early nineteenth century. Finally, we know that half of all children would have lost one parent before completing adolescence, and that in England a majority of them left home anyway between the ages of seven and fourteen, to act as servants in other people’s houses, to serve apprenticeships, or to go to boarding school. Under such circumstances the conflict of wills between parents and their adolescent children, which rips apart so many modern homes today, can have had little opportunity to develop. The identity crisis of puberty was normally passed away from home.

It is now possible to provide alternatives to these historically inappropriate traumas advanced by Freud as self-sufficient explanations of adult personality problems. As David Riesman has put it, “There has been a tendency in current social research, influenced as it is by psychoanalysis, to over-emphasize and overgeneralize the importance of very early childhood in character formation. Even within this early period, an almost technological attention has sometimes been focused on what might be called the tricks of the child-rearing trade: feeding and toilet training schedules….” This “assumes that once the child has reached, say, the weaning stage, its character structure is so formed that, barring intensive psychiatric intervention, not much that happens afterward will do more than bring out tendencies already set.”¹

No one doubts that child-rearing practices affect the adult personality, but acceptance of the theories of more recent ego-psychologists like Erikson and Hartmann opens up a new range of possibilities for the historian.² These theories involve hypotheses about the continued plasticity of the ego far into adulthood as it responds, through a series of crises, to the twin challenges of maturation and the influences of the family, the culture, and the environment. Not only do these theories have a strong ring of truth about them in the light of common experience, but they have the enormous advantage to the historian that they admit ego-development into periods of the life cycle when historical data becomes more readily available.

Secondly, since these developmental theories admit the influence of the social and cultural environment in affecting the nature, timing, and resolution of the recurring crises, they allow the historian to view the problem of ego-development in a broad historical frame. Clear evidence of distinctive features of national character, and fundamental shifts in character over time, for example from the other-directed to the inner-directed personality, can be explained in broader terms than those internal to the family itself.

This does not mean, however, that childhood experience in the past was without its effects on the adult personality. On the contrary, the experiences of the average child were so damaging that I believe that a large number of adults, at any rate of the gentry class in the period with which I am most familiar, namely, the sixteenth and seventeenth centuries, were emotionally stunted and found it extremely difficult to establish warm personal relationships with other people.

This was probably caused by four factors. The first was the frequency with which infants at that period were deprived of a single mothering and nurturing figure to whom they could relate during the first three or four years of life. Upper-class babies were mostly taken from their real mothers and put out to wet-nurse. These nurses were often cruel or neglectful, and they often ran out of milk, as a result of which the baby had to be passed from nipple to nipple, from one unloving mother-substitute to another. Even if it stayed with one nurse, to whom it became attached, the weaning process at about eighteen months inflicted the terrible psychological trauma of final separation from the substitute mother-figure and return to the alien world of the natural mother. This kind of experience is known to be psychologically and even physically damaging, leading to “deprivation dwarfism” and emotional atrophy in later life.

The second factor was the very high death rate. The constant threat and reality of the sudden loss of a parent, nurse, brother, sister, or friend soon taught the child to be wary of sinking too much emotional capital in any human being. Thirdly, the practice of tight swaddling in the first months or even year of life is believed to isolate the infant from its surroundings and to give it a sense of both frustrated rage and yet helpless acceptance of the cruelty and duplicity of the world. Thus there could be, and often was, a combination of sensory deprivation; motor deprivation, and emotional deprivation—to say nothing of oral deprivation—in the first critical months of life, of which the consequences upon the adult personality are now known to be very serious and long lasting in reducing the capacity for warm social relationships.

Finally, there was the deliberate breaking of the young child’s will, first by the harshest physical beating and later by overwhelming psychological pressures, which were thought to be the key to successful child-rearing in the sixteenth and seventeenth centuries. These four factors all contributed to a “psychic numbing” which created an adult world of emotional cripples, whose primary responses to others were at best a calculating indifference and at worst a mixture of suspicion and hostility, tyranny and submission, alienation and rage.

Given the validity of “psychohistory” as a legitimate enterprise, which is the most profitable field in which this research can be pursued? In my opinion, it will not be in the application of this or that psychological theory to the analysis of some particular person in history—Luther or Leonardo da Vinci or Woodrow Wilson or Hitler or Gandhi. What can more fruitfully be done is to study changes in family patterns and structures of specific classes or status groups in particular places. These changes will include relations of the nuclear core to the kin and the community, and economic and social power and affective relations between both husbands and wives and parents and children.

In that sense van den Berg, Ariès, and deMause are pursuing a far more promising line of historical inquiry than those who have tried to use psychology to interpret the behavior of individual figures in the past. I just do not think that such things as the extermination of six million Jews can be explained by the alleged fact that Hitler’s mother was killed by treatment given her by a Jewish doctor in an attempt to cure her cancer of the breast; or that Luther’s defiance of the Roman church can be explained by the brutal way he was treated by his father or by his chronic constipation.

These things may perhaps be necessary causes, but they certainly are not sufficient, and the result of such work to date has been disappointing, partly because of the flimsiness of the evidence of childhood experience, partly because of the speculative nature of the causal links with adult behavior, partly because of the neglect of the influence of the great processes of historical change in religion, economics, politics, society, and so on. As Malinowski pointed out in 1927, “Man disposes of a body of material possessions, lives within a type of social organization, communicates by language, and is moved by systems of spiritual values.”³ Any explanation of his history which ignores these cultural facts is not likely to be very convincing.

The first general model of childhood development in the West was that of Philippe Ariès. It is a pessimistic one of degeneration from an era of freedom and sociability to an era of oppression and isolation. According to him the Middle Ages and the sixteenth century were a period of happy social polymorphism, in which there were no divisions of ranks or ages, no separation of the child from the adult, no privacy, no external pressures from the state or the needs of an industrial economy, no internalized work ethic. Children and adults mixed together easily and naturally, wearing the same clothes, playing the same games, and working together on the same jobs. They also shared from the beginning a common knowledge of both sex and death.

In the seventeenth century, as a result of the spread of new kinds of Christianity into both Protestant and Catholic regions, a new attitude toward children developed, an event he describes as “the discovery of childhood.” This was not the work of Renaissance humanists but of seventeenth-century clergy. There was a rising concern for the child, which took two forms. First, there was a tightening of family bonds, along with the isolation of the family from external influences and a growing concern by the parents for their children; and secondly, there was a growing fear of the inherent corruptibility of the child by sin, leading to severity toward him in the home and to his isolation in schools regimented by age groups and disciplined to suppress all signs of moral backsliding. Medieval sociability was replaced not by Enlightenment individualism but by the isolated child-centered family and by the school, in both of which the prime concern was the taming of the will.

The rise of the repressive boarding school is the significant feature of this development, involving as it did a progressive extension of the period of childhood into adolescence and even beyond: “The central event is quite obvious: the extension of school education.” This transformation of attitudes toward childhood preceded demographic change, and indeed became itself the cause of demographic change when in the late eighteenth and nineteenth centuries it inspired a deliberate policy of contraception.

Ariès’s book has had a dazzling success and has been the primum mobile of the study of family history in the last decade. As a pioneer work, erudite, imaginative, and inventive, it deserves all the praise and attention it has received. It is the kind of path-breaking book no traditional historian could have written, and without it our culture would be the poorer. But for all its seminal brilliance, there remain unanswered certain basic questions: Is its methodology sound? Is its data reliable? Is its causal hypothesis valid? Are the alleged facts and alleged consequences true? In short, is the model correct, and if so, for what areas and for what classes?

In the first place, Ariès omits to point out the undeniable fact that between the Middle Ages and the nineteenth century the institution of the family lost many of its older functions to a series of impersonal institutions, such as poorhouses for the indigent, alms houses for the old, hospitals for the sick, schools for the children, banks for credit, and insurance companies for protection against catastrophe. Its legal, political, and economic functions declined before the ever-encroaching march of the institutions of the modern state. This functional erosion enhanced the prominence of the last area of family concern, the nurturance and socialization of the infant and young child.

Furthermore, the power of the state undermined the influence of the kin, and thus increased the isolation and privacy of the nuclear family. This process can hardly be called the rise of the family, but rather its reorientation to serve a narrower, more specialized function. The rise of the school is best seen not as part of the same process as the growth of the child-oriented family, but as its very antithesis, the transfer to an impersonal institution of a socializing function previously performed by the family. Moreover, although the repressive school was based on the theory of original sin, it was only in its first stage, in the seventeenth century, that the more child-centered family was also repressive, and there is clear evidence that by the eighteenth century among the English upper classes it was loving, affectionate, and nurturant.

Thus Ariès’s model is broken-backed, for his two agents of change, the child-centered family and the repressive school, were pulling in different directions and were caused by different ideas and influences. It thus lacks explanatory cohesion, as both Hunt and deMause point out. Moreover, its use of evidence, particularly iconographic evidence from art, to prove that the “discovery of childhood” actually happened, is not very convincing. For example, we now know for certain that although the Florentine bourgeoisie of the fifteenth and sixteenth centuries decorated their houses with painted and sculptured putti, they emptied them of flesh and blood babies, who were all sent out to wet nurses in the countryside. Putti, which Ariès uses as evidence of the discovery of childhood, are therefore really not evidence at all.

In addition, the thesis has a unilinear view of historical evolution which is contrary to the known facts. Children were more harshly treated in the sixteenth and seventeenth centuries and again in the nineteenth century at the two peaks of religious zeal for moral reform than they were in the eighteenth or twentieth centuries, and perhaps also than in the fifteenth century. Ariès’s chronology is very vague. One can never be quite sure whether one is dealing with the sixteenth or the seventeenth or the eighteenth century, and the book dodges about from century to century in a most confusing and indeed ahistorical way.

It is as vague in its geography as it is in its chronology, ranging casually from Italy to France to England for its evidence. For example, the presence of effigies of long-dead babies on tombs was relatively rare in France, but extremely common in England in the late sixteenth and early seventeenth centuries—a distinction whose significance Ariès completely ignores. Flogging died out in French schools in the eighteenth century but persisted in English schools into the twentieth. Wet-nursing died out in eighteenth-century England but persisted well into the nineteenth century in France on a very large scale. Geography clearly matters.

There is also too little attention paid to the particular class which is being dealt with. Ariès deduces the attitude to infantile sexuality of the society as a whole from that of the entourage of the future Louis XIII. The development of the boarding school teaching the classics, which affected only a tiny minority of the population, becomes a key event in early modern history. And finally, the powerful historical forces that affected the family so profoundly, changes in religion, political power, industrialization, urbanization, and poverty, are virtually ignored. Ariès’s book is in fact a history of French schools, and of upper-class and middle-class parents and children, that lacks the necessary historical context of time, place, class, and culture. A fascinating pioneering book, it is now recognized to be badly flawed in both its methodology and its conclusions.

David Hunt’s book is a psychological gloss on that by Ariès. It begins with a brilliant critique of Eriksonian egopsychology, pointing out that the latter’s optimism is ill-founded, for generativity is a fragile cultural artifact, not an instinctive human response. Consequently, in reality children have often been neglected and abused. Hunt also criticizes Ariès’s historical model for its nostalgic, even reactionary, Gemeinschaft-Gesellschaft Durkheimian view of change, and for its exaggerated stress on the school.

He then settles down to a detailed analysis of Dr. Héroard’s account of the upbringing of the infant Louis XIII. He stresses the child’s very close relation to his father, the all-powerful and virile Henry IV, and his very distant relations with his mother; and the way the child’s will was deliberately broken from the age of two by frequent whippings in order to instill the basic principle of obedience. He points to Louis’s later life as an unhappy, semi-impotent husband and, by a great leap of the imagination, attributes this to the experiences of his upbringing: knowing the physical facts of sex but not their psychological meaning, confused by contradictory signals about what was permitted and what was not, cowed by frequent whippings, and more or less isolated from his mother. Hunt further stresses the traumatic nature of the break at the age of seven, when he put on adult clothes, and was transferred from the control of women to that of men. His conclusion is that not sex, not anality, but “infantile autonomy was the major child-rearing problem in seventeenth century society,” and that this was linked to anxieties about status in a hierarchical law-and-order society.

Hunt is correct to stress that the breaking of the will was the key element in child-rearing in the sixteenth and seventeenth centuries, but it has to be pointed out that his evidence is in many respects less than satisfactory. In the first place (as I have mentioned), he is dealing with the son and heir of a king, and it is probably not legitimate to extrapolate the education of such an exalted person to that of, say, the middle-class child at the same period. Secondly, he has worked exclusively from the printed record of Dr. Héroard’s diary, which was published in the mid-nineteenth century, and which contains only a relatively small part of the whole text. It appears that in the manuscript the doctor’s main preoccupation was neither sexuality nor discipline, but the child’s health, and that much of it is therefore a detailed record of the daily input-output flow of food and excreta respectively. Until the total diary has been published, any conclusions based upon the published Victorian extract must remain suspect. Finally, the link between the adult personality traits and the infantile experience remains no more than an interesting speculation.

DeMause’s model of historical change is the exact opposite of that of Ariès’s, for it is optimistic. It is based on the following five propositions: (1) Parent-child relations are an independent variable in history. (2) There are only three possible adult reactions to children: projective, reversal, and empathic; the first two resulting in hatred and cruelty, and the third in love and affection. (3) Changes in parent-child relations are not affected by religious, social, political, economic, or any other factors, but operate by “psychogenesis,” a process by which the parents’ capacity to regress to the psychic age of the child has improved slowly over the centuries. Thus from generation to generation parents do a little better each time, until we reach the present when the perfect child is just around the corner. It is apparently as inevitable and self-contained a process as Darwinian evolution, “a powerful private force for change in historical personality,” operating “wholly independent of public events, economic, social, whatever.” (4) There has been an upward linear progression in the history of childhood for the last two thousand years, from the Infanticide Mode of classical antiquity to the Abandonment Mode of the early Middle Ages to the Ambivalent Mode of the late Middle Ages and early modern period to the Intrusive Mode of the eighteenth century—the great watershed—to the Socialization Mode of 1800-1950 to the Helping Mode of 1950 onward. (5) Child-rearing practices provide the key to the transmission of all other cultural traits visible in the adult.

Prodded by critics, deMause asserts that his model is not unilinear but multilinear, and involves “methodological individualism”—whatever that means—not “psychological reductionism.” These disclaimers do nothing to solve the problem of how to regard so bold, so challenging, so dogmatic, so enthusiastic, so perverse, and yet so heavily documented a model. For deMause child-rearing has replaced Marx’s control of the means of production and the class war as the key element around which the whole of history has to be conceived: our task as historians is to construct “a scientific history of human nature.” Are we dealing here with what Clifford Geertz has described as “the natural tendency to excess of seminal minds” or with a hopelessly unscholarly aberration, hanging loosely in the void between history and psychology, and lacking the methodological rigor of either discipline?

DeMause’s essay undoubtedly makes enthralling, if horrifying, reading. One learns about the way the writers of antiquity treated infanticide as a normal and sensible way to dispose of unwanted children; of how they amused themselves by using little children for fellatio or anal intercourse; of how the bones of child sacrifices are to be found in, the foundations of buildings ranging from 7000 BC to 1843 AD; of how seventeenth-century nurses played catch-ball with the tightly swaddled infants and sometimes dropped them, with lethal consequences; of how infants were dipped in ice-cold baths, in order to harden them, or perhaps merely to baptize them, but in practice sometimes killing them; about how the doorsteps and dunghills of eighteenth-century European towns were littered with bodies of infants, dead, dying, or just abandoned; how some wet nurses systematically starved their charges to death to save money or simply because they had accepted too many babies for their milk supply; of how children were ferociously beaten, shut up in the dark, deprived of food, terrified by bogeymen, taken to see hangings and corpses, sold into prostitution, blinded and otherwise mutilated to attract alms, castrated to supply testicles for magic, had their teeth ripped out to supply dentures, and in the nineteenth century suffered clitoridectomy, the attachment of toothed penile rings, and even nightly imprisonment in strait jackets to prevent masturbation—and so on, and so on, and so on.

What is one to make of this catalogue of atrocities? One obvious problem is the extent to which deMause is generalizing from the particular in constructing his linear model of child care. He clearly has a special taste for the macabre, and often grossly exaggerates, but in general it looks as if some of his basic conclusions are probably well founded. Antiquity undoubtedly regarded infanticide as casually as most of us regard abortion today, and certainly Christianity changed attitudes on this subject. There can also be no doubt that children were often neglected and exploited in the past, and there is growing evidence that the critical change to a more affectionate parent-child relationship took place in the eighteenth century.

The first question is whether this adds up to a linear theory of progress. The eighteenth-century change occurred mainly in England and America, and was largely confined to the professional and gentry classes. But middle-and upper-class parent-child relations worsened again significantly in the nineteenth century, before improving in the twentieth. The middle-class Victorian father was a terrifying, and often cruel, authority figure. As for the children of the poor, their condition probably deteriorated during the demographic, urban, and industrial explosion of the late eighteenth and early nineteenth centuries. But during the nineteenth century, contraception, humanitarian legislation, slowly improving economic conditions, welfare services, and schools probably improved the lot of the children of the poor, just at the time when that of the rich was worsening. The theory of linear progress is thus clearly a false one, and the story of change will have to be traced country by country and class by class, since each has its own individual life history.

Secondly, the “psychogenic” causal theory of change in parental attitudes is mere mystical nonsense. The reaction of parents to children is far from being limited to the projective, reversal, and empathic, as proposed by deMause. Everything suggests that in the past most parents have treated their children as inevitable by-products of sexual pleasure, sometimes bitterly unwelcome, sometimes barely tolerated, sometimes useful to be exploited economically, and sometimes cherished and loved. Most frequently, however, the response seems to have been one of relative indifference. The cruel truth—crueler, perhaps, than anything deMause has suggested—may be that most parents in history have not been much involved with their children, and have not cared much about them. Hence the staggering infant mortality rates—between a quarter and a third were dead before the age of one—were caused not by deliberate parental hostility, as he suggests, but rather by ignorance, poverty, and indifference. Most children in history have not been loved or hated, or both, by their parents; they have been neglected or ignored by them.

For all its brilliance, deMause’s model is thus defective in certain critical respects. There is no unilinear upward progression of childhood felicity, different stages apply to different classes in different countries, while there are huge time lags between different countries at the same period; the psychogenic theory of parental evolution is an unproven and implausible hypothesis; parent-child relations have altered in response to cultural determinants such as religious beliefs, economic pressures, customary practices, state power, etc.; parents have not normally had intense relations with their children but rather have regarded them with some indifference; and finally, it is an oversimplification to argue that “the child is father to the man” and that brutalized children automatically result in brutalized adults, who then take out their frustrations in war, violence, and murder. Unfortunately, neither modern ego-psychology nor modern genetics nor modern anthropology suggests so simplistic a chain of causation.

The other essays in deMause’s book do something t flesh out, or sometimes to contradict, his model, but many of them are a little disappointing. One reason for this is that the source materials for such a study are very unsatisfactory, and that they get thinner and thinner the farther back one goes in time. They consist either of advice on child-rearing by doctors, theologians, and moralists, which are often totally contradictory and may or may not bear any relation to behavior in practice; or else a number of individual well-documented examples, which may or may not be representative of a class or nation or period. The gathering of such materials often results in no firm conclusions at all, but merely in a series of contradictory and ambiguous impressions.

Taken chronologically, the first successful essays are those by Professor Marvick on Louis XIII, based at last on the manuscript record, and by Professor Illick on seventeenth-century England and America, in which the reader is brought face to face with the hard demographic facts of high infant and child mortality. Under these conditions, no parent could retain his or her sanity if he or she became too emotionally involved with such ephemeral creatures as young children. Aloofness, or the acceptance of God’s will, or sending one’s children away from home were three natural solutions to this problem of how to deal with their deaths.

Professor Illick also makes the very interesting point that American practice preceded Lockean theory of childrearing, a suggestion further developed by Professor Walzer in his essay on eighteenth-century American childhood, where again the evidence suggests closer relationships between parents and children in America than was the case in England, owing partly to the shortage of servants to look after the children, and partly to the lower death rate. There is a dramatic contrast between this fairly humane picture of family life in eighteenth-century America and the brutality and indifference which Professor Dunn shows was still the norm in nineteenth-century Russia. Could it be that the child-centered family started first in America, moved then to England, and slowly traveled east, not reaching Russia until the twentieth century? If so, it is a purely cultural artifact which has been slowly diffused from class to class and from country to country, but without any clear line of progress.

In view of the weaknesses as well as the brilliant insights of the models presented by Ariès and deMause, what are we to put in their place? One of the many problems of studying childhood in isolation is that it lends itself to passionate polemics—all the authors have an obvious axe to grind. Another more serious objection is that it is impossible to study children in isolation from those who killed them off or fed them, nurtured them or neglected them, beat them or fondled them, namely, their parents. The history of childhood is in fact the history of how parents treated children. (A similar objection can be made against the impending flood of books about the history of women, an even more emotion-laden subject, which also cannot properly be studied in isolation from those who dominated them economically and sexually, exploited them, beat them, denied them education and opportunities for professional advancement, and who also pampered them, pelted them, supported them in comfortable leisure, put them on pedestals, sexually fulfilled them, and sometimes even loved them—namely, men.) The proper unit of historical studies in this area is thus neither children nor women nor men, but the family, the institution within which all these personal interactions take place.

My own view of the evolution of the family in the West is that there have been two fundamental shifts, both occurring in the early modern period between 1500 and 1800. The first was the shift in loyalties and functions from a kin-oriented system to the nuclear core, and the second was the shift from predominantly economic to predominantly emotional bonding. The first stage of the nuclear child-centered family in the sixteenth and seventeenth centuries was repressive and authoritarian, the second in the eighteenth century was permissive and affectionate, although there was a striking relapse into authoritarianism again in the nineteenth century. These changes were caused by external factors, not by internal self-generated “psychogenesis,” and were slowly diffused from class to class and from country to country.

The first influence was that of the fifteenth- and early sixteenth-century humanist educators, who postulated the malleable innocence of the child and the need for his careful training; the second, which soon overlaid the first, was the sixteenth- and seventeenth-century Puritanical fear of sin and the Devil, which drove newly caring parents to practice deliberate harshness toward their children in order to crush their wills—not for mere self-interest but for the supposed good of the children. The repressive boarding school evolved at the same period, partly because, as Locke later pointed out, the classical, grammatical, and linguistic training given to the elite in these schools was so tedious that only brutality could induce children to put up with it, and partly because parents were anxious that their children should continue to be severely controlled and isolated from the manifold temptations of the world.

In the eighteenth century there was a striking shift away from the authoritarian family and the repressive school and college, mainly as a result of that mysterious process, the growth of individualism. The family became increasingly child-centered, but also far more egalitarian and tolerant in its internal distribution of power between husband and wife, parents and children. The causes of this vast change in human attitudes are far too complex to be described in this essay, involving a check, in the Anglo-Saxon world, to the power of the state, a decline in religious fanaticism, and the revival of a more optimistic view of human nature. The first influential book setting out the new educational ideal was written by that crusty English bachelor John Locke.

It should be stressed that these changes took place first in America and England, then later spread to France, and later still further east, and also that they were at first exclusively confined to the properties classes, gentry, professional men, and higher bourgeoisie—families which were not so grand as to be able to maintain a small army of nursery staff to take care of the children for them, but rich enough to indulge in the luxury of sentiment. There are plenty of examples of truly loving parents in the eighteenth century, and the first peak of permissiveness in gentry class child-rearing was reached toward the end of that century, only to decline thereafter. These new attitudes spread very slowly, by a process of stratified diffusion, upward into the high aristocracy and downward into the lower middle class and then to the poor, encouraged and supported by the humanitarian impulses first set in motion by the eighteenth-century Enlightenment, and transformed into legislative action during the course of the nineteenth century.

It should also be stressed that this stage theory of family evolution does not imply any value judgment about the march of progress, or any assumption that the nuclear, individualistic, emotionally bonded family type which has emerged must necessarily always be better, in some social and moral sense, than the family types which preceded it. It is arguable that possessive individualism is an ideal which has no demonstrable basis in psychological or social reality, and has brought with it some evil consequences as well as many benefits.

As Philip Slater has pointed out in his recent book The Earth Walk, “The notion that people begin as separate individuals, who then march out and connect themselves with others, is one of the most dazzling bits of self-mystification in the history of the species.” The result is not only the Bill of Rights but also the undermining of community organizations, the consequent enhancement of the power of the centralized state, and a narcissistic obsession by the individual with self-fulfillment which is so often inevitably self-defeating. Since affective ties are limited to the nuclear family, both husband and wife tend to develop exaggerated expectations of sexual and emotional satisfaction, which produce the rocketing figures of divorce. This concentration also often leads to over-intense parent-child relationships, which result in children who are obsessive over-achievers and who experience great difficulty in cutting the umbilical cord at the period of adolescence and emergence into the world: they find themselves still tied to their parents by strings of love and/or hate.

Despite its many virtues, the rise in the West of the individualistic, nuclear, child-oriented family which is the sole outlet of both sexual and affective bonding is by no means always an unmixed blessing. Nor is it at all clear that it will necessarily survive the twentieth century, in view of the impossible demands which are so often made upon it, and the many signs of a reaction against it by men, women, and children. It is a final ironic thought that just as deMause is heralding the advent of the perfect parent-child relationship based on the permissive theories of A.S. Neill, many American young are losing interest in children, and are choosing not to have any at all. When they do have them, they are also, it seems, either turning away from permissiveness in the home or are dumping them in day-care centers at the earliest opportunity. The cycle of history is revolving once more: “progressive” parents, “progressive” schools, and “progressive” colleges are all on the decline, just at the moment when deMause sees the promise of the arrival of the “free” child-rearing millennium.

Thanks to the interest generated by the brilliant pioneering historical work of Erikson, van den Berg, Ariès, Hunt, and deMause, and by the energy and entrepreneurial enthusiasm of deMause himself, a huge amount of historical research is now in progress on the history of the family. There is little doubt that in ten years time the picture will be much clearer, and that all of us who have ventured to advance theories about this subject will have been proved wrong. The results could change our understanding, not only of the past but also of the present and the future.