Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 27 June 2019

Children in the Roman Family and Beyond

Abstract and Keywords

This article demonstrates the status and conditions of Roman children in the family. The first section explains the demographic conditions of children in Roman society, including the high mortality rates and the various contraceptive methods used. The next section looks at the childhood of Roman children, from their birth to their coming of age. The roles of the mother and father are also discussed. The last section of the article discusses orphans and stepparents and stepsiblings. This article concludes that very few Roman children grew up in an intact core family; rather, they grew up in heterogeneous families, with stepsiblings and stepparents.

Keywords: children, demographic conditions, mortality rates, contraceptive methods, childhood, coming of age, parents, orphans, stepparents, stepsiblings

Introduction: Demographic Conditions

Procreation was considered by most Romans to constitute the most essential purpose of wedlock; and, according to ancient moral philosophers, that purpose alone justified sexuality within a marriage (Plu. Moralia 140B). Thus, Romans wed (at least ideally) in order to have children—and the Latin formulaic expression liberorum quaerendorum causa (for the purpose of begetting children) was to be found in most Roman marriage contracts. This desire for children seems generally to have arisen from several different needs: to secure the continuance of the family name; to provide heirs to the family property; or, especially among those of lesser wealth or status, to ensure a system of support in old age. To be sure, there is the occasional reference in ancient literature to family happiness, to the outright joy afforded parents by small children (e.g., Lucr. 3. 894–911). However, this is rarely mentioned as a reason for begetting a child. Rather, offspring were principally intended to guarantee the survival of the family: oikos (household) for the Greeks, domus (household) or familia (family) for the Romans.

Thus, remaining childless was considered a significant evil, and was feared in particular by the head of the household. On the other hand, we possess many sources that attest to what may have been reasonably widespread childlessness (Parkin 1992: 114). Pliny the Younger, to name but one example, remained childless, despite three (p. 624) marriages. His third wife, Calpurnia, did become pregnant, but had a miscarriage (Plin. Ep. 8. 10, 8. 11). Many divorces were, therefore, the result of a marriage's lack of issue.

An extremely high mortality rate for children caused many a Roman not to have immediate heirs. Probably about 25–30% of newborns died within the first year of life; by way of comparison, in modern industrialized countries, infant mortality amounts to just about 1%. Of the Roman children who survived their births, about 50% died before reaching the age of ten (for the statistics, see Frier 1982, 1999, 2000). Therefore, the number of surviving children remained low, despite an apparently high rate of pregnancies. Because of this high child mortality, families with many offspring were the exception in antiquity. And so, to name but one example, the Roman emperor Marcus Aurelius and his wife Faustina had at least twelve children; yet, only one son, the later emperor Commodus, survived his father. Given, then, the great likelihood of losing one's children in their formative years, the opinion is frequently voiced by scholars, that Roman parents could not risk investing too many feelings in their children. At first sight, this thesis seems to be supported by the fact that deceased infants were often not even buried properly; and it is true that, generally speaking, children are underrepresented on the gravestones of the imperial period (Shaw 1991). However, such an utter lack of emotional ties between parents and children is contradicted by many literary sources, which show that parents were much distressed by the death of their children, even of small children and infants (Golden 1988).

In any case, because of high and unpredictable mortality rates, Roman families were, unlike many of today's, significantly different from each other. For example, in present-day Western Europe, most married couples have one or two children, who are mostly born relatively soon after the parents' wedding, and who, as a rule, reach adulthood. As soon as the desired number of children has been reached, effective modes of contraception are practiced. This was different in Rome. To secure descendants, Roman parents were often forced to produce children until comparatively late in their lives. As a result, the age difference between parents and children tended to be much larger than in most contemporary European families (Krause 1994–95, 1: 33–34). This applied similarly to the age difference between siblings. It could lie in the vicinity of two or three years, but the census declarations preserved from Egypt show that siblings were frequently separated by ten years or more. Such age differences could be even more significant between step-siblings.

In short, then, the high and completely unpredictable rates of child mortality led to significantly diverse family settings in the Roman Empire. Marriages would often last just a few years. When children were born, no predictions at all could be made as to whether they would survive their parents. Families without surviving children must have been commonplace; and if there was only a surviving daughter, then property would likely be transferred from her family to that of her husband. On the other hand, if a large number of children in a given family unexpectedly survived, thus beating the odds, that family's property would have to be distributed among them. This could lead to a significant social decline for the next generation, (p. 625) and in the worst case scenario, to impoverishment. Therefore, in Rome, the number of surviving children could well decide the rise or fall of a given family.

Consequently, finding oneself confronted with a large number of surviving children could easily prove to be just as undesirable as could a dearth of children. Moreover, since Roman inheritance law did not follow the principle of primogeniture, it was desirable to keep the number of children small, and thus to avoid splitting the family property amongst too many heirs. Contemporaries were well aware of the risk of a large progeny, who might easily become a financial burden. And so, under Tiberius' reign, for example, Hortensius Hortalus, who came from a distinguished senatorial family, fell into severe financial difficulties; he no longer possessed the minimum wealth required of a senator. One of the reasons cited for his predicament were his many children—he had four (Suet. Tib. 47).

Now, assuming that a child did survive, then his education required great investments that, in the best case, found some return only late in life. The state developed merely a tentative social policy, which was intended to lessen the burden of educating children for members of the society's lower strata. Thus, the emperor Trajan offered monies to less-well-to-do parents and their children, which could be used for the purposes of education. Moreover, loans from the central government were also made available to Italian landowners; then, the interest payments on these loans were administered and used by the local communities to assist parents, at least partially, with the costs of educating their children. Such alimentary programs supported girls to the age of thirteen or fourteen, and boys for a few years longer (Duncan-Jones 1964; Woolf 1990; Wierschowski 1988; Rawson 2001; see also Horster in this volume).

But in spite of any and all such programs, it is clear that there was significant interest in limiting the number of children. However, just which methods of family planning were most often used, and how efficient they ultimately were, remains unclear. Ancient medical texts describe a number of contraceptive practices (Hopkins 1965–66; Parkin 1992: 126–27). The techniques, as they are portrayed in these texts, were based equally on methods involving magic and medicine. Now of course, some of these operations will surely have been effective, but it is questionable whether the difference between successful and useless methods was terribly well realized. And in fact, what would appear to be the simplest form of contraception, coitus interruptus, is not at all mentioned in our ancient texts.

Aside from contraception, the ancient medical texts also list a number of medications and surgical procedures that would lead to abortion; once again, we find a mix of effective and completely unsuitable approaches. Nonetheless, there were certainly successful abortions; and while doctors discussed the legitimacy of abortion well into the imperial period, this procedure did not pose a legal problem. Abortion was not considered criminal—unless it might serve to impede a father's right to a legitimate potential heir. Given that concern, however, it was illegal for a wife to pursue an abortion without her husband's consent; should she do so, then she was perceived to be interfering with her husband's absolute prerogative to determine his child's destiny. Abortion with a father's consent, on the other hand, was completely legal (Dig. 47. 11. 4; 48. 19. 39).

(p. 626) Abortion, though, posed very great risks to a mother's health; thus, the abandonment of children offered a less dangerous method of family planning. The decision to forsake a child was usually made right after birth and lay within the jurisdiction of the head of the family, the pater familias. Now, while we are in no position to quantify cases of abandonment, ancient authors routinely consider this a frequent practice (Boswell 1988; Harris 1994; Corbier 1999: 1261–73; Corbier 2001). Poverty, of course, was a continual motive for the abandonment of children. According to Plutarch, the poor often chose not to raise their children because they were concerned that they would not be able to provide them with an adequate education (Moralia 497E). But even the rich got rid of surplus children; in such circles, the exposure of the child was usually intended to avoid splitting the inheritance among too many offspring, a circumstance already encountered above.

Compared to abortion, the abandonment of children had the advantage that it could be used in a much more targeted fashion. And so, for example, while it is difficult to quantify this exactly, it would appear that girls were abandoned in higher numbers than boys (Apul. Met. 10. 23). For while sons began to contribute early to the economic well-being of a family through their work, daughters had to receive a considerable dowry at the time of their weddings. Thus, sons were generally perceived as being instrumental for the financing of old age; daughters tended to be, instead, a financial burden. Furthermore, it was through the offspring of sons that the family line would be continued.1

Often, children were abandoned at a crossing, or in front of a temple, in the hope or expectation that some passer-by would take care of the child. Thus, a few abandoned children had the chance to be raised as foundlings, and, consequently, the abandonment of children cannot be exactly equated with homicide. Abandonment of children was also most likely one of the more important resources for the slave supply. The well-known grammarian Melissus, for example, had been abandoned by his parents as a newborn, was brought up as a slave, received a superior education, and was given as a gift to Maecenas. Later, his mother demanded that he be freed; Melissus, however, preferred to remain a member of Maecenas' household—even if as a slave (Suet. Gram. 21. 1–2).

In short, then, Roman demographic realities posed serious hazards for parents and children alike. The vicissitudes of both child survival rates and the available methods of family planning always entailed great risks. Insofar as the latter is concerned, the options available (contraception, abortion, abandonment) were likewise treacherous for mothers and for those children who were forsaken. On the other hand, should parents produce too few children altogether, then the high mortality rates might easily bear another, more far-reaching danger: early deaths of the few born could lead to the family's extinction outright. And yet again, in the case of too many children, fragmentation of the familial property might threaten the economic well-being of subsequent generations. Thus, as we saw at the outset, the desire for children, and for enough surviving children to guarantee familial stability (p. 627) in various senses, was great. In short, children were highly valued, and producing children was a well-recognized social virtue. Nevertheless, the desire to control family size and make-up, and the lack of modern technologies for doing so, meant that precarious methods of family planning were widely tolerated. The forces of both nature and nurture, in the end, tended to result in a rather precarious existence for Roman children.

From Birth to Coming of Age

Usually, the birth of a child took place in the father's house, but in a much less private atmosphere than is usually the case nowadays. First, it was customary to have a midwife present. In the lower classes of society, neighbors often took over the midwife's function. Frequently, other women, mostly relatives of the couple, attended the birth. The midwife was the first person to examine the newborn, and therefore made a preliminary decision as to whether the child was fit to live. However, it was the father who had to decide whether the newborn was to be raised or to be abandoned.

On the dies lustricus, the ninth day after the birth of a boy, the eighth after the birth of a girl, the child of an elite family would be named. This delay can probably be explained by a very high rate of perinatal mortality, and the caution it must have generated. Until the dies lustricus, the child was, as it were, in a transitional phase. It was only after a few days had passed that it could realistically be determined whether the baby had an actual chance of survival. In any case, assuming that the child did survive, on the dies lustricus the newborn would also receive the so-called bulla, a leather locket containing amulets that was worn by aristocratic children until they reached adulthood, and that served an apotropaic function. Aside from the bulla, an aristocratic youth also wore a particular garment, namely, the toga praetexta, a toga with a purple border stripe, until he attained manhood.

In the upper strata of society, it was typically a wet nurse who fed the baby and was responsible for the child's earliest upbringing. In larger households, such wet nurses were primarily slaves; however, there were also poor free women who made a living this way (Joshel 1986; Bradley 1994; Corbier 1999: 1274–80). It is impossible to make generally valid statements about the duration of the nursing. In Roman Egypt, contracts with paid nurses were frequently set up for two years, and doctors generally recommended that babies should be nursed at least until they developed their first teeth, in other words, until at least their sixth or seventh month.

Even after the weaning process, the wet nurse would usually remain with the child. She possessed great influence on her ward's personal development. These relationships with wet nurses often lasted a lifetime, and sometimes reached a level of stability that did not exist in the parent-child relationship. Pliny the Younger, for example, gave his wet nurse the gift of a country estate, valued at 100,000 sesterces, (p. 628) to support her in her old age (Plin. Ep. 6. 3). The philosopher Favorinus opposed the practice of using wet nurses, complaining that, among other things, the mutual love between mother and child was bound to suffer as the children developed closer loving relationships with their wet nurses (Gell. 12. 1. 22–23). We can only speculate as to the effects of employing wet nurses (and later teachers) upon the relationships of children with their mothers. At least in the upper classes, wives were only loosely tied into their husband's family through the task of raising their children, and of this occupation they were mostly relieved by slaves. This might partially explain the comparatively high number of divorces in the Roman Empire. Hardly ever does the divorced wife's separation from her children, who normally remained in their father's household, seem to have posed a problem.

As soon as the child was older, a slave, who would function as a teacher, was assigned to him. This person would accompany the child everywhere, not only to school, but, for instance, also to the circus or the theater. This slave-teacher was of very great importance for the child's socialization (Bradley 1985a). Consequently, wet nurse and teacher shared, at least in the upper classes, the tasks of a child's most basic early education. Especially the father, it would seem, had very little contact with his small children. Thus, Seneca comforts a father regarding his son's early death with the argument that the wet nurse had anyhow known the child much better than did the father (Sen. Ep. 99. 14). Very young children in Roman aristocratic families, then, had caregivers beyond father and mother, and these individuals were of potentially greater importance than were the parents during those children's formative years.

Still, there could well be input, early on, from the adult members of the family. In Roman houses, at least insofar as the available archaeological evidence would indicate, there were no rooms specifically intended for use by children. Youngsters simply lived and played among the adults. Indeed, many probably shared sleeping quarters with the slaves, who were assigned to take care of them. Thus, when Roman authors do show us children at play, the scenario is not in a children's room of any sort, but in the central living area and parlor of the house, namely, the atrium. To this extent, then, a Roman aristocratic child was pretty well constantly under the vigilance of his or her mother and father.

With respect to care-giving by the parents, it was generally expected that the mother would look after the small children. A son's education, starting at about the age of seven, was preferably the father's task, for it was an old republican principle that boys and young men were supposed to learn primarily from their father's example. However, as soon as the boys reached the stage of early adolescence, they were entrusted to one or more experts, mostly relatives or friends of the family, whose example they were likewise expected to imitate (Tac. Dial. 34). With the Roman adoption of the Greek school system, beginning in the second century BC, the educational system changed completely. In upper-class families, the children were now usually educated at home by a slave or freedman, primarily of Greek origin, who was well versed in literature. However, only few parents could afford such private tutors. The alternative was to have the children attend a school. While Quintilian, in (p. 629) his handbook on rhetoric, stresses the importance of education within the family, he, as opposed to many of his contemporaries, prefers schools over private education at home (Inst. 1. 2. 4–5).

In poorer circles, childhood ended quickly. Boys were expected, from a young age, to contribute to the family's support. In agriculture, for example, children were utilized to herd small animals (Var. R. 2. 10. 1; Col. 8. 2. 7; Ov. Fast. 4. 511). From Roman Egypt, we have sources documenting children as substitute workers during the olive harvest; they received up to four obols per day, while the daily payment for adults amounted to six obols (P. Fay. 102, ca. AD 105). Boys learned a trade very early—at the age of twelve or thirteen—and were soon in a position to supplement the family finances. So, for example, the writer Lucian was originally supposed to learn such a trade; it was planned that he would become a stone-mason. This education was not as costly as one that would lead to the intellectual professions, and the son would be able to help support the family much more quickly (Lucian Somn. 1–4). Quite correctly, research has insisted on the immense importance of children for their parents' support in old age, and this especially among the lower classes. Their early professional training, then, should be seen in this context (Bradley 1985b: 327–30; Saller 1988: 405–6; Wiedemann 1989: 153–55).

Coming of age for elite boys was marked by the exchange of the toga praetexta, again, a boy's special toga, for the toga virilis (the manly toga). There was no absolutely fixed point in time for this ceremony; however, boys were between fourteen and sixteen years old when they received their adult clothing. For the girls, marriage was the point in time that separated childhood from adulthood, and Roman girls could legally marry from the age of twelve.

The Father's Role

Children simply belonged to their father—they took his name and they received his social status. The Romans talked of the power of the father (patria potestas), but not that of the parents. There were only loose legal ties with the mother's family—indeed, with the mother herself. And in many areas, such as inheritance law, or with regard to the arranging of a guardianship, agnatic relationships were all-important.

In short, patria potestas was arguably the most distinctive characteristic of the Roman family (and was similarly important in the law; see Kehoe in this volume). It encompassed the physical structure of the house with everything that belonged to it, the free members of the family, the slaves, and the assets. This power was almost unlimited, and the pater familias was technically entitled to kill his slaves as well as his children without suffering any punishment—he possessed the so-called right of life and death (ius vitae necisque) over these members of the household. This right to kill applied not only to the newborns, but even to an adult son, who continued to (p. 630) be subject to this patria potestas. Legally, there was almost no difference between a slave's and a son's position with regard to the head of the family.

During his lifetime, the father (again, technically) controlled the economic activities of the children, who were under his potestas. He could give them an allowance (peculium), but was not obligated to do so. Legally, the children did not own anything so long as they were subject to this fatherly power. This patria potestas must have been especially oppressive because there was no definite coming of age that ended it. It simply lasted until a father's death. Only then did children become legally independent (sui iuris).

The demographic conditions of the Roman world (especially the high mortality rates), however, created a situation in which adult children must rarely have remained under their fathers' power. When young women married, often shortly before they turned twenty, in all likelihood half of them had already lost their fathers. Men would marry (ideally) at about the age of thirty; at this point in their lives, perhaps only a quarter of them would still be under their fathers' patria potestas. In short, given the mortality rates that must have been operative in that world, only a small fraction of adult Romans will have had very long to tolerate a father, who was still alive, and who thus still controlled the family fortune.

There were yet further mechanisms that weakened patria potestas. For example, an institution called emancipatio was a fictive sale, via which a pater familias released his son or daughter from his power. The father ‘sold’ his child to a third party. This person then freed the child; but instead of gaining freedom and full legal rights, the child fell back into the father's power. This process was performed three times, and after the third ‘sale’ the child would be released entirely from the father's power. Until the early imperial period, though, emancipatio was most likely not a regular practice. This might have changed in late antiquity. This may be so, given that one of Constantine's laws mentions the coming of age, and says that this usually released children from paternal power (CTh 8. 18. 2, AD 319). In other words, what previously had been an irregular practice seems by Constantine's day to have become quite normal.

Furthermore, patria potestas could be moderated by the grant of a peculium, in other words, a specific set of assets (sometimes plain cash), which legally remained part of the father's property but could nonetheless be used relatively freely by a son; daughters sometimes received such grants, but this was rare. While the father could theoretically rescind a peculium he had granted, such a situation, or the various problems that might conceivably have resulted, play hardly any role whatsoever in the Roman legal sources. It was obviously common to allow a dependent the use of such a peculium once granted, and to withdraw it only for very important reasons. In most cases, the peculium was more than mere pocket money. It was generally large enough for sons to be able to enter contracts, even large enough to serve as security for obligations vis à vis the state. And so, with the peculium, a son received the opportunity to form and run his own household, and to live according to his social status. This institution thus served a double purpose—it led to a certain (p. 631) independence for the individual granted such a peculium, yet, at the same time, it continued the financial dependency on the father (Thomas 1982).

The alternative to granting funds that a son could himself administer was the payment of an annuity, which allowed him to live in accord with his social station. Like a peculium, such an annuity continued the economic dependency on a father. Some young men were not satisfied with the monies supplied by their fathers, and therefore sought loans from willing lenders. The resultant indebtedness of youths still under their fathers' potestas posed a continuous problem. Apart from fathers, who often tended to be frugal, or outright stingy, with regard to their children, there were others who favored a more liberal educational style. The existence of such liberal fathers must have made the situation of those who were dependent on a frugal father far into their adult lives even more difficult.

All of this, of course, raises the question as to just when a Roman reached adulthood, and just how the Romans understood the attainment of adult status. More important, really, than attaining a particular age was the family situation, in other words, whether a person was still dependent on a father or not; and the described Roman family structures led to considerable inequalities. The son of a long-lived father would remain under patria potestas potentially far into his life, and all the while did not possess full legal capacity; and yet, many far younger men became, due to their fathers' earlier deaths, much sooner the heads of their households. The potential for conflict here is obvious.

We may now turn to the matter of provision for the heirs upon a father's death. Even from early on, the custom of establishing a will was widely common in Rome.2 It should be noted, however, that the standard rule of inheritance, namely, that children would be instituted as heirs, could be modified through individual arrangements. This gave wealthy fathers the possibility to exert pressure on their children by threatening them with disinheritance. While the children would have the right to contest an unfounded disinheriting via a prescribed legal procedure (the querela inofficiosi testamenti), it was still possible for a father to leave his children with the legal minimum, namely, a fourth part. Clearly, then, there were important reasons for a son to obey his father. Nevertheless, certain limitations had to be observed. Due to considerable social pressures, a father could not afford to disinherit his son without a clear cause. It was generally assumed that each legitimate child, both male and female, should receive an equal, or at least a substantial, portion of the inheritance. Reasons for excluding a child completely from an inheritance had to be very convincing. And while disinheriting was pretty certainly fairly rare, a hint toward a future inheritance may well have served as a means to achieve obedience and respect from children.

Another institution, that of a domestic court that was presided over by the pater familias, and in which he could render judgments over his family members, continued to exist into the imperial period. In this context, the power of the father over his son, the husband over his wife, and the proprietor over his slave was lent a (p. 632) quasi-legal sanction. Crimes committed within a family were mostly not brought before public courts, but instead were handled in this more private realm. And while the context was more private-seeming, the decisions rendered in such a family court could be quite effective. The power of a household head over his children included, as we have already noted, his ability to put them to death (the so-called ius vitae necisque). No public court was needed to exercise this prerogative; it was, however, customary for a father to consult with a group of relatives, serving as a board of advisors (in Latin, a consilium), before he rendered such a judgment.

That having been said, it must nonetheless be realized that there are very few sources supporting the actual utilization of this ius vitae necisque by Roman fathers (Thomas 1984; Harris 1986). Indeed, during the reign of Augustus, people in Rome almost lynched a father who had whipped his son to death—public opinion tended to be hostile toward any excessive use of paternal force (Sen. Cl. 1. 15. 1). Obviously, a simple right to kill was, at least in the early imperial period, no longer undisputed, and future enactments limited its scope further. For example, the emperor Hadrian (AD 117–138) deported a father who killed his son during a hunting trip because the son had engaged in an affair with his wife, the son's stepmother. Hadrian supported his decision by stating that the father had killed his son in the manner of a robber, rather than according to paternal law. The point was that a father's power was supposed to be based on dutiful respect (in Latin, pietasDig. 48. 9. 5, Marcianus), and the father in question had, apparently, acted simply out of anger. By the fourth century, at the latest, a father's right to kill his children had been abolished.

But despite any and all limitations, the ius vitae necisque was, so long as it survived, not entirely without meaning. It existed as a threat, and must have had an according psychological effect on children. In sum, then, a father's power became legally more limited over the imperial period. Still, this power did continue to exist, and the basic patriarchal family structures were never seriously questioned (Arjava 1998).

Now, the legal documents that provide much of this picture are not necessarily well suited to describe the daily realities of family life. They determined only what which was legally possible. Thus, a father could indeed disinherit his children, and he could kill them without being punished, but that does not mean that most or many fathers acted that way. The relations between fathers and children were, just as one would expect, multifarious. There were strict as well as liberal fathers, fathers who spoiled their children, and stingy fathers. Indeed, it would appear that Romans increasingly expected fathers to show not only qualities such as strictness, but also benevolence, interest, or tender love. Furthermore, our literary sources repeatedly hint at the educational role of the father. Relations between parents and children were ultimately, and ideally, to be formed on the basis of pietas, and pietas had to be understood reciprocally. It was owed to the father by the son, but also to the children by the father (Eyben 1991; Saller 1991; Saller 1994).

The important thing to remember, in the end, is that for all the variety of relationships that will have existed between Roman fathers and their children, in the background always loomed the institutions of patria potestas and the ius vitae (p. 633) necisque. We cannot quantify the effects of these on the relationships in question. However, effects there must surely have been.

The Mother's Role

A woman who had entered a sine manu-marriage (a particular legal category of marriage, which was more liberal toward the wife) was by law bound to her husband's family only loosely. A mother never shared in the patria potestas, even when her husband was not up to the task of overseeing the family. Indeed, even after her husband's death, she did not gain any such power over her children. As a rule, she guided their education, should the father not be able to, but a guardian had to be appointed to administer the children's financial concerns. The mother was not entitled to become the legal guardian.

So long as a marriage continued, the mother bore very little financial responsibility for the children's support; principally, it was the husband's task to take care of this. However, the proceeds from the dowry could be used for the support and education of common children. As early as in republican times, a husband was entitled to keep a share of the dowry on behalf of the children in the case of a divorce initiated by his wife or her pater familias. Therefore, a mother had a certain financial responsibility for her children, and this potentially continued even after the end of her marriage. Proceedings after the death of a wife were similar to those after a divorce. In that situation, the dowry (or more exactly, the dowry specifically given by the bride's father, in Latin, the dos profecticia) went back to the wife's father, minus one-fifth for each surviving child; in other words, the support of the children had to be co-financed from the deceased mother's property. In the case of the husband's death, the entire dowry was at the wife's disposal (or that of the person with legal powers over her). The children's support, however, was to be financed solely with the deceased husband's assets; or in any case, this was the legal perspective on the matter.

In the imperial period we can observe a tendency to give greater legal weight to the cognatic than the agnatic relationship. Such a strengthening of the blood relationship implied an acknowledgment by the jurisprudents of closer relationships between mothers and their children. This was true in particular for inheritance law, the rights of custody upon a divorce, and regulations regarding guardianship.

Thus, beginning in the second century AD, children and their mothers possessed the mutual right to inherit on intestacy (by virtue of the senatus consultum Tertullianum and the senatus consultum Orfitianum). Now, while it was a general rule that children of divorced parents would grow up with their fathers, legal developments of the imperial period allowed the mother greater influence with regard to the children's education. In certain critical circumstances, for example, a divorced mother might even be allowed to gain guardianship (Wacke 1980). According to (p. 634) classical Roman law, the administration of fatherless children's property was their guardian's affair alone, while a mother was simply not entitled to that role. Nonetheless, even in the early imperial period, a mother had a certain influence over the administration of the property, which could lead to conflicts with the guardian. In late antiquity, a mother could become the guardian of her children officially, especially in cases when no agnatic relative was available (CTh 3. 17. 4 = CJ 5. 35. 2, AD 390; Krause 1994–95: 3, 113–29). Thus, while the law increasingly took close relationships between mothers and children into consideration, the Romans were always far from treating mother and father equally in this regard.

However, once again, it must be admitted that the law simply does not reflect the entire reality of the situation. While there was no maternal equivalent for the patria potestas, on which the mother could have based a (legal) right to respect from her children, she might very well be the object of love and respect for her children—a mother was at least thought to be owed pietas, in the same fashion as was a father.

One of the most important tasks of a wife was the education of the children (cf. Osgood in this volume). She was primarily responsible for the young children. For them, the mother was the most important contact in the nuclear family; however, this depended on the family's social status. In the upper strata of society, parents were assisted in the education of their children by various household personnel, slaves, but also by professional educators and teachers from outside the household (see Horster in this volume). Among the lower classes, it is a likely guess that mother-child relationships were, for better or for worse, on the whole, closer. While the rich seem generally to have handed newborns over to a wet nurse, the mothers in poorer families will have raised their children by themselves. In any case, apart from the care for young children, mothers were also in charge of the education of their daughters, who remained in their mothers' care until marriage.

But even with regard to the sons' education, mothers possessed a certain influence. In some cases, both parents made basic decisions with regard to their children's upbringing. For example, the choice of a marriage partner for one's offspring lay within the father's domain, but it was generally common to consult the mother as well, and over the imperial period her influence in this regard seems to have grown.

Ancient literature frequently mentions the different educational styles of fathers and mothers. Fatherly strictness is commonly contrasted with a mother's greater indulgence. Seneca points out that fathers usually insisted that the children pursue their studies, while mothers nursed and pampered their offspring (Prov. 2. 5). It was usually thought to be a father's task to reprimand and discipline a son. It may be that we are simply dealing here with literary topoi; however, there may also be a kernel of reality in such stories.

Now, the fact that a wife in a Roman marriage frequently possessed her own property served actually to strengthen her position toward her children. Larger expenses on behalf of the children were often afforded by using the mother's assets. This provided the mother with, for example, some influence over her children's (p. 635) education. And just as a father could threaten his children with disinheritance, so a mother could exert some pressure by hinting at her will. Given the property distribution within Roman families, it was therefore important for children to court their mothers as well as their fathers.

The role of the mother in the Roman family clarifies, then, how purely legal strictures and social reality could lie some distance apart. According to Roman law, while a mother belonged to her husband's family only loosely, and although connections between mothers and children were strengthened legally only in the imperial period, there was no question that in Roman society generally, a mother's responsibility for and connection to her children was very large, both within a marriage and after its dissolution.


Probably about 40% of all Roman children lost their fathers before they reached the age of 15. These youths would then be put under guardianship (tutela), usually of a close (male) relative. If the father had died intestate, the closest agnatic male relative would usually become guardian (this was called tutela legitima); the guardian was, as a rule, a paternal uncle, or an older brother of the now-fatherless child. This tutela legitima fell to a person who could legitimately inherit the ward's assets, with the exception that a woman was entitled to such an inheritance but was not permitted to serve as guardian. Boys were freed from guardianship when they turned fourteen; however, at that point, they did not gain full legal capacity. They continued to be subordinate to a person who would look after their interests (in Latin, a curator) and whose agreement was necessary with regard to all business transactions; however, as opposed to a guardian, a curator was not legally in charge of his ward's property.

A mother's possibilities to influence financial decisions were legally fairly limited. Single women were hardly considered capable of controlling their own property. Even less was there any willingness to trust women with the administration of a third party's property, even in the case of the woman's own children. Therefore, a mother was not at all in any position to make autonomous decisions with regard to all the important issues affecting her children after her husband's death.

As a rule, and as has been indicated, the mother took a lead in her children's education, while a guardian would be in charge of their property. Especially after the death or remarriage of a mother, orphans were frequently taken in by other relatives (e.g., grandparents, or uncles and aunts). As just one of many possible examples, Octavian, subsequently the emperor Augustus, grew up on his grandfather's estate; and his education was guided by his maternal grandmother, Julia, upon the remarriage of his mother. After his grandmother's death, Octavian moved to his mother and stepfather's house; he was then twelve. The young Caligula, son (p. 636) of Germanicus, seems to have lived with his great-grandparents over periods of time while his parents were still alive. After his father's death, he at first lived with his mother, and after her exile, with his great-grandmother, Livia. When Livia died, he was sent to his paternal grandmother, Antonia, and finally moved to Tiberius' household on Capri at the age of nineteen (Krause 1994–95: 3, 49–77).

Thus, grandmothers and aunts often had an important function with regard to the education of children. Children were, on occasion, and even while a father was still alive, but especially after a mother's death, transferred to the grandparents, often enough to the grandmother, or perhaps to an aunt. This was an experience shared not only by those who had lost both parents, for children who had lost a father were often sent to their grandparents, or uncle and aunt, even while their mother was still alive. Especially if the mother remarried, such a solution was preferred to the children's inclusion in a stepfamily. Thus were kinships beyond the core family often groomed.

Frequently enough, maternal relatives assisted with the education of orphaned children. A widowed mother was usually perceived only with difficulty to be able to take on the education of her children alone, and women in this position often sought out the support of their own parents or a sister (a maternal aunt of the orphans). Thus, while the general preference was to give orphans into the care of paternal relatives, nevertheless, the engagement of maternal relatives was a possibility. Even after the dissolution of a marriage, close contacts and relations between the children, who technically belonged to their deceased father's familia, and their maternal relatives could remain in existence.

Stepparents and Stepsiblings

The Roman family is characterized just generally by the following distinctive features, among others: the inconstancy of marriage and its easy disintegration; the frequent early death of one partner; and the availability of a surviving partner for a new marriage. In short, the Roman family formed a dynamic unit, which could be continuously dissolved and newly reconstituted (Bradley 1987, 1991a). And that, of course, will have had its effects on the upbringing and socializing of children.

Men who had been married once before, in other words, were widowed or divorced, tended to prefer younger women for their second marriages, since a young woman would be better suited to producing children. When a man married again, it was thus possible that the marriage partners belonged to very different generations, and that the second wife was possibly little older, or potentially even younger, than the children from the man's first marriage.

Upper-class families in particular were not based on marriage connections of terribly great permanency; children could not count on their parents' marriages lasting until their deaths at an old age. A fairly large number of men and women (p. 637) married at least twice during adulthood. Proportionally, many more children than in (say) modern Western society may well have been confronted with stepparents and stepsiblings. It is difficult to guess the emotional consequences of such marriage patterns. In ancient literature, we continually encounter the motif of the bad stepmother (Gray-Fow 1988; Noy 1991; Watson 1995). This hints at frequent ‘real-life’ tensions between children and stepparents. There is evidence for quite a few ‘good’ stepmothers; however, they do not dominate the overall picture drawn of stepmothers in Roman society. In Roman literature, the topos of the bad stepfather is not quite as prominent as the one of the bad stepmother, primarily because most children lived in their father's household upon a divorce, and thus were confronted with a stepmother rather than a stepfather. If we believe ancient sources, not only relationships between stepparents and stepchildren were strained, but relationships between stepsiblings were usually distant. Financial matters especially are often portrayed as having caused disputes; so, for instance, it was considered a certain risk that a mother might neglect the children from her first marriage when distributing her wealth.


Given all of the above, it seems safe to put things as follows. The group in which a Roman child was raised and socialized was very often more broadly cast than the core family consisting of father, mother, and children. Indeed, it would appear that a significantly small percentage of children in antiquity experienced (at least for all of their childhood) something like an intact core family. Rather, the Roman family, at least in the upper echelons of the society, seems to have been much more fluid. This was the result, on the one hand, of the natural mortality rates, and on the other hand of the socially determined frequency of divorce, and the consequences thereof, for children's living conditions. In short, very many children grew up in heterogeneous families, with stepparents and stepsiblings.

Apart from their parents, and others of their relatives, there were various care-givers with whom Roman children must frequently have established close emotional relationships. For the aristocratic child there were female wet nurses and male slaves, who frequently supervised the children's education. Though we obviously cannot gauge exactly the effects of such an upbringing, it seems hard to imagine that these different persons did not exert influence over the being of their young wards.

The ‘typical’ Roman family was, therefore, quite different from (say) the typically perceived modern Western core family. The latter has an expected format. Two partners, almost equal in age, form a new household; there are relatively few children; these children are reasonably close in age; adult children leave the household, generally in their late teens or early twenties; and parents are, at that point, left by themselves. The period of post-paternal companionship in a marriage, then, may (p. 638) possibly last for decades. Such a model is certainly not at all suited to describe the Roman family.

The result of this all is that in several ways one might consider children (or in any case, many children) in the Roman world to have been marginalized by their society. Clearly, they were not outsiders in anything like the sense that magicians, or bandits, or prostitutes were. Yet, one can easily enough point to similarities between the positions of slaves and children, which might encourage a much less than optimistic picture of the life of a Roman child. Thus, it is well worth considering just exactly what the place of these young people in Roman society was.

Within the family, children were, one might argue, instrumentalized, to perpetuate the gens, or to support their elders. Furthermore, the fact that there often was a whole range of persons who saw to the care and upbringing of children could perhaps also be thought to have isolated these young people in various ways. That a child was subject to exposure, indeed to every whim of a father wielding the patria potestas, has also caused some despair about the plight of Roman children. Indeed, there was a quite influential ancient intellectual trend to perceive children as being absolutely irrational, hence, in many ways excluded from the ‘inside’ of Roman society. As one scholar has recently put it:

A moral and philosophical tradition ranked children, like women and irrational animals, among outsiders excluded from the male social hierarchy. Children still had to develop the qualities that would turn them (or a tiny minority of them) into full-fledged, independent Roman aristocrats. They still lacked the ability to act in a morally correct way. They were unable to control anger and other emotions, they did not speak in a rhetorical style, they were weak and infirm, and, above, all, irrational.3

What is more, a forceful current in social history generally has perceived childhood in pre-modern European society as having been altogether dismal, as having been largely characterized by an utter lack of any recognition for this as a particular and discernable (not to say fragile) stage of the human life cycle (e.g., Ariès 1962).

Most recently, however, there has been a tendency to see things otherwise. A significantly more affective model of the Roman family, and thus of the place of children within that family, has largely gained sway (see now esp. Rawson 2003). To name but one fundament of the more recent scholarly trend, the use and effectiveness of the quintessentially Roman institution of patria potestas has been called seriously into question (Saller 1994). In short, it is nowadays much more difficult than it once was to think of Roman children as outsiders, or as (p. 639) marginalized by their society. Nonetheless, it is important to realize that there was a time when, and there were scholars among whom, Roman children could indeed be viewed as little more than slave-like creatures, as truly marginalized figures in their community. Nowadays, though, we would do much better to see these young people as having been “welcome and valued and visible in Roman society” (Rawson 2003: 1).

Suggested Reading

Recently Beryl Rawson has produced a very good and convenient overview of children in the Roman world (Rawson 2003). This is probably the best place to begin reading about this subject. And as mentioned just above, she argues against the idea that Roman children were marginalized by their society. Keith Bradley's various writings (see just below in the bibliography) on Roman children are likewise most valuable. Saller 1994 is the most important work in recent years on the subject of patria potestas, and Frier 2000 provides a brief, but excellent introduction to demography in the Roman world.


Ariès, P. 1962. Centuries of Childhood. A Social History of Family Life. New York: Vintage Books. (Orig. L'Enfant et la vie familiale sous l'Ancien Régime. Paris: Plon 1960.)Find this resource:

    Arjava, A. 1988. “Paternal Power in Late Antiquity.” Journal of Roman Studies 88: 147–65.Find this resource:

      Balla, P. 2003. The Child-Parent Relationship in the New Testament and Its Environment. Peabody, Mass.: Hendrickson.Find this resource:

        Boatwright, M. T. 2005. “Children and Parents on the Tombstones of Pannonia.” In George 2005: 287–318.Find this resource:

          Boswell, J. 1988. The Kindness of Strangers: The Abandonment of Children in Western Europe from Late Antiquity to the Renaissance. New York: Pantheon Books.Find this resource:

            Bradley, K. R. 1985a. “Child Care at Rome: The Role of Men.” Historical Reflections / Réflexions Historiques 12: 485–523 = Bradley 1991a: 37–75.Find this resource:

              —— . 1985b. “Child Labour in the Roman World.” Historical Reflections / Réflexions Historiques 12: 311–30 = Bradley 1991a: 103–24.Find this resource:

                —— . 1987. “Dislocation in the Roman Family.” Historical Reflections / Réflexions historiques 14: 33–62 = Bradley 1991a: 125–55.Find this resource:

                  —— . 1991a. “Remarriage and the Structure of the Upper-Class Roman Family.” In B. Rawson (ed.), Marriage, Divorce, and Children in Ancient Rome. Oxford: Oxford University Press. 79–98 = Bradley 1991b: 156–76.Find this resource:

                    —— . 1991b. Discovering the Roman Family. Oxford: Oxford University Press.Find this resource:

                      —— . 1994. “The Nurse and the Child at Rome. Duty, Affect and Socialisation.” Thamyris 1: 137–56.Find this resource:

                        —— . 2005. “The Roman Child in Sickness and in Health.” In George 2005: 67–92.Find this resource:

                          (p. 640) Champlin, E. J. 1991. Final Judgements. Duty and Emotion in Roman Wills 200 BC–AD 250. Berkeley: University of California Press.Find this resource:

                            Corbier, M. 1999. “La petite enfance à Rome: lois, normes, pratiques individuelles et collectives.” Annales 54: 1257–90.Find this resource:

                              —— . 2001. “Child Exposure and Abandonment.” In S. Dixon (ed.), Childhood, Class and Kin in the Roman World. London: Routledge. 52–73.Find this resource:

                                Dasen, V. (ed.). 2004. Naissance et petite enfance dans l'Antiquité. Actes du colloque de Fribourg, 28 novembre–1er décembre 2001. Fribourg: Academie Press.Find this resource:

                                  Dixon, S. 1988. The Roman Mother. Norman: University of Oklahoma Press.Find this resource:

                                    —— . 1999. “The Circulation of Children in Roman Society.” In M. Corbier (ed.), Adoption et fosterage. Actes du colloque international, Paris, 4 et 5 juin 1993. Paris: De Boccard. 217–30.Find this resource:

                                      —— . (ed.). 2001. Childhood, Class and Kin in the Roman World. London: Routledge.Find this resource:

                                        Duncan-Jones, R. P. 1964. “The Purpose and Organisation of the Alimenta.” Papers of the British School in Rome 32: 123–46.Find this resource:

                                          Evans, J. K. 1991. War, Women and Children in Ancient Rome. London: Routledge.Find this resource:

                                            Eyben, E. 1986. “Sozialgeschichte des Kindes im römischen Altertum.” In J. Martin and A. Nitschke (eds.), Zur Sozialgeschichte der Kindheit. Freiburg: Verlag Karl Alber. 317–63.Find this resource:

                                              —— . 1991. “Fathers and Sons.” In Rawson 1991: 114–43.Find this resource:

                                                French, V. 1991. “Children in Antiquity.” In J. M. Hawes and N. R. Hiner (eds.), Children in Historical and Comparative Perspective. An International Handbook and Research Guide. New York: Greenwood Press. 13–29.Find this resource:

                                                  Frier, B. W. 1982. “Roman Life Expectancy: Ulpian's Evidence.” Harvard Studies in Classical Philology 86: 213–51.Find this resource:

                                                    —— . 1999. “Roman Demography.” In D. S. Potter and D. J. Mattingly (eds.), Life, Death, and Entertainment in the Roman Empire. Ann Arbor: University of Michigan Press. 85–109.Find this resource:

                                                      —— . 2000. “Demography.” In A. K. Bowman, P. Garnsey, and D. Rathbone (eds.), The Cambridge Ancient History 11: The High Empire, A.D. 70–192. 2nd ed. Cambridge: Cambridge University Press. 787–816.Find this resource:

                                                        Garnsey, P. 1991. “Child Rearing in Ancient Italy.” In Kertzer and Saller 1991: 48–65.Find this resource:

                                                          George, M. (ed.). 2005. The Roman Family in the Empire. Rome, Italy, and Beyond. Oxford: Oxford University Press.Find this resource:

                                                            Golden, M. 1988. “Did the Ancients Care When Their Children Died?” Greece & Rome 35: 152–63.Find this resource:

                                                              Gray-Fow, M. J. G. 1988. “The Wicked Stepmother in Roman Literature and History: An Evaluation.” Latomus 47: 741–57.Find this resource:

                                                                Harlow, M., and R. Laurence. 2002. Growing Up and Growing Old in Ancient Rome. A Life Course of Approach. London: Routledge.Find this resource:

                                                                  Harris, W. V. 1986. “The Roman Father's Power of Life and Death.” In R. S. Bagnall and W. V. Harris (eds.), Studies in Roman Law in Memory of A. Arthur Schiller. Leiden: E. J. Brill. 81–95.Find this resource:

                                                                    —— . 1994. “Child-Exposure in the Roman Empire.” Journal of Roman Studies 84: 1–22.Find this resource:

                                                                      Hopkins, K. 1965–66. “Contraception in the Roman Empire.” Comparative Studies in Society and History 8: 124–51.Find this resource:

                                                                        Huskinson, J. 1996. Roman Children's Sarcophagi. Their Decoration and Its Social Significance, Oxford: Oxford University Press.Find this resource:

                                                                          (p. 641) Joshel, S. R. 1986. “Nurturing the Master's Child: Slavery and the Roman Child-Nurse.” Signs 12: 3–22.Find this resource:

                                                                            Kertzer, D. I., and R. P. Saller (eds.). 1991. The Family in Italy from Antiquity to the Present. New Haven: Yale University Press.Find this resource:

                                                                              Kleijwegt, M., and R. Amedick. 2004. “Kind.” Reallexikon für Antike und Christentum 20: 865–947.Find this resource:

                                                                                Kneissl, P., and V. Losemann (eds.). 1988. Alte Geschichte und Wissenschaftsgeschichte. Festschrift für Karl Christ zum 65. Geburtstag. Darmstadt: Wissenschaftliche Buchgesellschaft.Find this resource:

                                                                                  Krause, J.-U. 1994–95. Witwen und Waisen im Römischen Reich. 4 vols. Stuttgart: Franz Steiner Verlag.Find this resource:

                                                                                    Laes, C. 2007. “Inscriptions from Rome and the History of Childhood.” In M. Harlow and R. Laurence (eds.), Age and Ageing in the Roman Empire. Portsmouth, R.I.: Journal of Roman Archaeology. 25–37.Find this resource:

                                                                                      McWilliam, J. 2001. “Children among the Dead: The Influence of Urban Life on the Commemoration of Children on Tombstone Inscriptions.” In Dixon 2001: 74–98.Find this resource:

                                                                                        Néraudau, J.-P. 1984. Etre enfant à Rome. Paris: Payot.Find this resource:

                                                                                          Noy, D. 1991. “Wicked Stepmothers in Roman Society and Imagination.” Journal of Family History 16: 345–61.Find this resource:

                                                                                            Parkin, T. G. 1992. Demography and Roman Society. Baltimore: Johns Hopkins University Press.Find this resource:

                                                                                              Rawson, B. (ed.). 1991. Marriage, Divorce, and Children in Ancient Rome. Oxford: Clarendon Press.Find this resource:

                                                                                                —— . 1997. “The Iconography of Roman Childhood.” In B. Rawson and P. Weaver (eds.), The Roman Family in Italy. Status, Sentiment, Space. Oxford: Clarendon Press. 205–32.Find this resource:

                                                                                                  —— . 2001. “Children as Cultural Symbols. Imperial Ideology in the Second Century.” In Dixon 2001: 21–42.Find this resource:

                                                                                                    —— . 2003. Children and Childhood in Roman Italy. Oxford: Oxford University Press.Find this resource:

                                                                                                      Saller, R. P. 1988. “Pietas, Obligation and Authority in the Roman Family.” In Kneissl and Losemann 1988: 393–410.Find this resource:

                                                                                                        —— . 1991. “Corporal Punishment, Authority, and Obedience in the Roman Household.” In Rawson 1991: 144–65.Find this resource:

                                                                                                          —— . 1994. Patriarchy, Property and Death in the Roman Family. Cambridge: Cambridge University Press.Find this resource:

                                                                                                            Shaw, B. 1991. “The Cultural Meaning of Death: Age and Gender in the Roman Family.” In Kertzer and Saller 1991: 66–90.Find this resource:

                                                                                                              Thomas, Y. 1982. “Droit domestique et droit politique à Rome. Remarques sur le pécule et les honores des fils de famille.” Mélanges d'Archéologie et d'Histoire de l'école Française de Rome, Antiquité 94: 527–80.Find this resource:

                                                                                                                —— . 1984. “Vitae necisque potestas. Le père, la cité, la mort.” In Du châtiment dans la cité. Supplices corporels et peine de mort dans le monde antique. Rome: École Française de Rome. 499–548.Find this resource:

                                                                                                                  Wacke, A. 1980. “Elterliche Gewalt im Wandel der Jahrtausende. Zum Sorgerecht der geschiedenen Mutter nach römischem Recht.” In W. Eck, H. Galsterer, and H. Wolff (eds.), Studien zur antiken Sozialgeschichte. Festschrift Friedrich Vittinghoff. Cologne and Vienna: Böhlau. 417–34.Find this resource:

                                                                                                                    Watson, P. A. 1995. Ancient Stepmothers. Myth, Misogyny and Reality. Leiden: E. J. Brill.Find this resource:

                                                                                                                      (p. 642) Wiedemann, Th. 1989. Adults and Children in the Roman Empire. New Haven: Yale University Press.Find this resource:

                                                                                                                        Wierschowski, L. 1988. “Die Alimentarinstitutionen Nervas und Traians. Ein Programm für die Armen? ” In Kneissl and Losemann 1988: 756–83.Find this resource:

                                                                                                                          Woolf, G. 1990. “Food, Poverty and Patronage. The Significance of the Epigraphy of the Roman Alimentary Schemes in Early Imperial Italy.” Papers of the British School in Rome 58: 197–228.Find this resource:


                                                                                                                            (1.) And for the abandonment of disabled children, see Stahl in this volume.

                                                                                                                            (2.) On Roman testamentary habits generally, see Champlin 1991.

                                                                                                                            (3.) Laes 2007: 37. What Laes demonstrates overall, however, is that this particular ‘literary’ or ‘intellectual’ discourse on children was radically different from the kind of talk (and emotion) that appears on the tombstones of Roman children. He posits, at least in part, a difference in attitude owing to the difference in genres of writing/communication. Note that there are several other articles in this volume on children and the perception of them in the Roman world (H. Sigismund-Nielsen, J. Huskinson, and R. Redfern).