The ancient Greco-Roman economy had two distinct and contiguous characteristics: agriculture and the use of slave labor. These two features characterized the economy in general throughout antiquity in both urban and rural contexts. These characteristics were also fundamental for the economy of many households and the financial well-being of the family, and both adults and children were dependent on these elements for their continued existence.
The material comfort of the life of a child in antiquity was naturally closely linked to the family’s prosperity or poverty. The economy was a vital element in determining how many children a family could raise—if they had a choice. This chapter focuses primarily on how the family economy affected the life of a child and his or her family and also highlights other factors relevant to discussion of the economy, such as the social and legal status of the family, gender structures, and social practices. These factors all shaped the lives of children and their families in the ancient world. Much of our evidence for this topic comes from Athens in the fifth century b.c.e. and in Rome from the late republican period and in imperial times until about 200 c.e. The discussion and examples from the Greco-Roman world concentrate on these cities and these periods, but this chapter also embraces a longer period of time by bringing the discussion into the early Middle Ages. In late antiquity and the medieval period, Athens and Rome were no longer leading financial centers, and during these periods the evidence concerning children and the family economy comes mainly from other regions, reflecting the changing political landscape of the time and also the changing nature of the sources.
Children, particularly sons, were necessary for the continuity of the family line. The importance of children is demonstrated, for instance, by childlessness being an accepted reason for divorce in ancient Rome. Childless marriages were always considered the fault of the wife, and a man could divorce his allegedly sterile partner in order to remarry and have children with the intention of saving his family lineage for the future. According to Aulus Gellius (second century c.e.), the first divorce in Rome for this reason occurred around 230 b.c.e. when Spurius Carvilius Ruga divorced his wife because of her sterility.
The actual birth of a child, even if freeborn, did not automatically make the infant a member of the family of his or her parents. The final acceptance of a new child was made by the male head of the family—in Athenian society by the kyrios and in Rome by the paterfamilias. His decision would have been based on a number of factors such as gender of the child, the absence of any visible defects, and the size of the family. However, at the birth of a new child, the crucial concern must often have been if the household possessed the economic means to bring up another child. A child who survived the first critical years of life could at a later stage in the life course become an economic asset to the family. Thus, the economic well-being of the family would have depended on the number of children it could maintain.
Poorer parents might not have had the economic means to raise all children born to them, which may explain why exposure may have been used as a method of birth control. Child exposure and abandonment were practiced by the Greeks as well as the Romans. Exposure is a much-debated issue and is often believed to have been practiced more often on baby girls than newborn boys. It is clear from the ancient sources that abandoning children was not considered a criminal act and that it was not an uncommon practice, but it is impossible to determine precisely how frequently it occurred, both before and after the abandonment of children was prohibited by law in the late fourth century c.e. A poor family who could not afford to keep a new child might have been forced to have their children adopted by others or to sell a child, perhaps even into slavery.
A child who lived to be a teenager or young adult drew on the economic resources of the family in a number of ways. The costs involved were different and differently paced according to the gender of the child. There were costs connected to the upbringing of sons in terms of school education and socioritual events that, like many other ritual acts, involved the sacrifice of animals and large feasts. Thus, it is likely to have been something of an economic relief for a family when a son became financially self-supporting. Children living in a town or an urban center had more and probably better opportunities for education and training than children in the country. There was a greater stress on a boy’s education than a girl’s in both Greek and Roman society. Sparta was the only ancient state that prescribed a general education for girls. In light of the greater emphasis on male education, it has been questioned whether Greek girls and women in general could read and write. However, with the absence of a general school system, all children could neither hope for nor claim education as a right. A family’s social status and, probably most of all, its financial situation must have been decisive in the matter of formal academic education. Wealthy Roman families had their own educational system whereby both sons and daughters were often taught by a learned slave. The education of the young males of the family was likely to imply larger costs than for daughters, since a son’s education was more extensive, often including time abroad and preparing for a public career. A daughter in a wealthy Roman family was provided some years of education, thus making her a literate and educated person—a factor in her high status in adulthood.
A girl’s life in general was much more focused on a future of marriage and family life, and if she had any education, her school days were probably over by the time she approached marriageable age (normally the age of twelve). An important part of bringing up a daughter, as well as a son, was training for adult life. For a daughter, such training implied preparation for her future role as wife in her husband’s household. For this role she would probably have been trained by older females, either family members or slaves. Thus, the part considered most essential in a girl’s training for life as an adult (i.e., as a wife capable of running a household) mostly took place in her own home. Financially, this training would have involved hardly any major cost for the family and did not incur the expense of engaging a teacher. If a girl’s education involved lower costs for the family than educating a son, marrying off a daughter was likely to have been more expensive. The preparations for the girl’s marriage may have begun several years before the wedding by looking for an appropriate husband, preferably of an equal, or perhaps higher, social and economic status to the daughter. The choice of husband could be followed by a betrothal. For a first marriage, the girl was probably still living in the house where she had grown up, and when the day of the wedding had been set, preparations for the wedding ceremony, her bridal outfit, and the dowry began. The jewelry a bride wore at her wedding represented part of the wealth of the household, but the largest cost for her parents was her dowry.
In contrast to a daughter, a son was normally older at the time of his first marriage and by that time was likely to have already established a household of this own. In the capacity of master of his own home, a man in his early or midtwenties would, at least on a daily basis, have been economically less dependent on the older generation than a daughter living with her parents or other adults or relatives until the day she married.
From an economic point of view, a daughter may not have contributed much to a family’s financial well-being. Instead, she may have been seen as a burden to the household because a dowry would strain the family economy. In most European preindustrial societies, a dowry was the means for a daughter to claim a share of the father’s estate, but the size as well as the legal and practical arrangements concerning dowries varied considerably over time. In classical Greece, the legal system and practices on dowry varied from polis to polis as is demonstrated from those states where evidence survives. In the law code from Gortyn in Crete (fifth century b.c.e.), a daughter had the right to one-half of the share of a son, and it could be used either as a dowry by the time of her marriage or inherited at the death of her father. Had a daughter’s share been used for a dowry in a marriage that ended by divorce, the woman could keep what she had brought to the marriage. Contemporary with the law code from Gortyn, in classical Athens, a dowry was essential for a girl’s marriage since the marriage could be considered illegal without it. The importance of a dowry was demonstrated by the fact that if a father could not by economic means provide a dowry for his daughters, it would have to be produced by relatives or even by the city of Athens itself. The size of an Athenian dowry was related to the prosperity of the girl’s father. Thus, a man would perhaps not have the option of raising more daughters than he could afford to furnish with dowries. In Rome, however, there were no legal obligations of a dowry to grant validity to a marriage, but it was a widespread practice. A substantial dowry could increase a girl’s status as a marriage partner, and vice versa—the lack of a dowry might ruin her chances of marriage. The dowry could be composed of money and/or actual property that immediately became the full legal property of the husband, or his paterfamilias, once it was handed over to him.
There was also some variation in the size of dowries in antiquity. Most of our information from both Greek and Roman contexts concerns dowries for daughters of elite families, but a broad pattern can be discerned. Dowries in Athens were in general moderate compared to those in Rome. This is especially true from the second century b.c.e., when the Roman aristocracy allowed very substantial dowries to become important economic tools. The more generous dowries coincide in time with a general increase in wealth among aristocratic Roman families, partly through republican warfare that resulted in war booty that made already wealthy elite families even richer. Even an Athenian dowry could occasionally represent a substantial share of the patrimony, taking up to about twenty-five percent of the whole estate. But still, even the most substantial dowries in the ancient Greco-Roman world were much lower than dowries in the wealthy families of early modern Europe.
Regardless of whether a dowry consisted of money, as was customary in classical Athens, or if it involved actual property, as it often did in wealthy Roman families, its overall purpose was to contribute to the maintenance of the wife during marriage. The size of the dowry and its composition was a reflection of the economic means of the bride’s family. In Rome, the dowry could be reclaimed by her family in case of the death of the woman or the husband or in the case of divorce. If the dowry was reclaimed, it was often with the intention of using it again in the formation of another marriage. Roman law of the early imperial period included a relatively liberal divorce policy, and children from a marriage ending in divorce remained with their father and were still under his control. However, a mother might have had some financial responsibility for the children after a divorce.
The liberal view of divorce changed in later Roman law, from the fourth century c.e., when divorce, as well as the possibility of remarriage, was more restricted than in earlier periods. During the reign of the emperor Constantine (305–337 c.e.), laws were enacted to restrict divorce, especially for a wife who wanted to divorce her husband. In general, Christian leaders opposed divorce except if a wife was involved in adultery. It is difficult to know to what extent the laws of late antiquity in combination with the diffusion of Christian ideology may have affected popular opinion on divorce, but it seems that at least among the Roman aristocracy, Christian or pagan, divorce was extremely rare by the fifth century c.e.
In many ways, an ancient marriage was an alliance between two families—that of the bride and that of the groom—but it was not primarily an economic pact. In the interest of both families, the properties of a Roman husband and his wife were legally kept apart by restrictions in relation to gifts, inheritance, and financial commitments. During the period of betrothal, gifts between the partners were allowed, but gifts between husband and wife once married were not permitted in Roman law. The intention of these marital restrictions was to secure and keep separate the property of the family of the husband from that of the wife and to prevent any reduction in economic value in either familia. Consequently, inheritance between spouses was not allowed unless it was the type of marriage where the husband had been appointed guardian of the wife (i.e. a marriage cum manu, rare after 100 b.c.e.). From the first century c.e., the protection of separate property was taken a step further by the senatus consultum Vellenaium—husband and wife could no longer stand surety for each other’s debts. In late antiquity, the legal view of property within a marriage changed radically. The laws of the fifth century c.e. were less interested in keeping property between spouses apart, and, as a consequence, husband and wife could now inherit from each other and were even looked upon as each other’s natural heirs if there were no children.
All families, rich or poor, might have had an interest in limiting the number of children—either due to the costs of maintaining a child or with the intention of maintaining the inherited wealth of the familia. Sons and daughters tended not to inherit equal shares of an estate. In the laws of , discussed earlier, a daughter inherited half the share of a son; in classical Athens, a daughter did not inherit anything while all sons inherited equally, although the eldest son might have had some privileges in relation to his brother(s), such as claiming the family name as well as having first choice when landed property was divided between brothers. The division of land, of course, could cause problems if the original estate was very small because very small plots of land were insufficient to support several new families. This problem is well attested in ancient sources. The inheritance laws of Athens reflect the importance of having a son because only males could inherit; thus, the economic value of a child was clearly related to gender. Even in the context of high child mortality and the potential problems caused by multiple heirs, many families still raised more than one son for insurance against another’s death.
If there were no sons, or if a man was unmarried, the adoption of a male could be the solution to perpetuate the oikos. Because girls could not inherit property or wealth from their father, they were rarely adopted. The purpose of adoption was not to care for an orphan or to provide childless couples with children. Instead, the intention was primarily to perpetuate the oikos and the family line through a male heir. An adopted son became the heir to his adoptive father’s estate and was known by the patronymic of his father by adoption. Because of this, small children were very seldom adopted; instead, an adult relative was preferred. Logically, the adopted son lost his right to inherit his natural father. In classical Athens, a daughter would gain her share of her family’s estate mainly through her dowry, and normally she would not inherit on the death of her parents. In certain circumstances a daughter could, however, technically become an heiress. She would then be an epikleros—“attached to the family property” (see chapters 1 and 7).
Roman law presents a different perspective on male and female rights of inheritance since both legitimate sons and daughters inherited from the father. Yet female inheritance in Rome was, by tradition, much more restricted than male inheritance, and a son was more likely to be the primary heir. In 169 b.c.e., the Lex Voconia limited the rights of inheritance for Roman females. From then on, a man could no longer make a daughter heir to more than half of his property. Roman sons and daughters could inherit through a mother’s will, and, from the second century c.e., intestate inheritance was also possible from a mother. A child of either sex in a well-to-do Roman family who lost one or both parents early in life could, thus, be wealthy at a very young age. Such a child would normally be under the guardianship of a tutor, an adult responsible for the child’s economic assets and transactions. An example that may illustrate a Roman daughter’s right of inheritance from a father is the case of Antonia Minor (36 b.c.e.–37 c.e.), the youngest daughter of Mark Antony (83–30 b.c.e.). At his death in 30 b.c.e., all his children inherited from his estate. Antonia was Antony’s last legitimate child and only six years old when her father died. She probably never even met her father, but, nevertheless, she inherited a share of his estate, including large estates in the eastern Roman provinces, and became very wealthy by age six. In such a situation—the death of a father—a guardian would normally be appointed for the younger children to protect the property left to the children; a mother, if still alive, would be responsible for the upbringing and welfare of her children. In the late fourth century c.e., a widowed mother who declared she would not remarry could herself be the guardian of her own children. In the later Roman Empire, the idea of marriage shifted from a general civic duty to a more personal choice or a choice of the family. Within marriage, inheritance remained a vital aspect, and parents were prohibited from disinheriting their children. The shift in attitude and law on marriage also affected unmarried persons. In Augustan legislation, unmarried men and women experienced restrictions on inheritance rights, but changes in late antiquity implied improved rights for unmarried persons to inherit and leave property not only to family members but also to unrelated persons. For example, unmarried Christians may have preferred to give some or all of their property to the church.
The financial well-being of a family depended on its social status and wealth, which could derive from work, marriage, and/or inheritance. The head of the family (a kyrios or paterfamilias) had overall responsibility for the household and its economy. In an ideal world, the male citizen would have been the owner of productive land that made him financially independent and also enabled him to devote his time to the public duties of a citizen. However, few could match this ideal. Family finances, however, were not solely a male preserve; a husband’s wife was also an important partner in the household economy. Once married, the woman was expected to run the household in a responsible way and oversee the work of slaves, duties for which she was trained prior to marriage. The family economy was partly dependent on how well the housewife ran the household. Some insight into what was expected from a housewife can be found in Xenophon’s book Oeconomica from the fourth century b.c.e. It is clear from this book and other sources that female work, by free women or slaves, was mostly confined to indoor activities, while it was more common to expect men to work outdoors.
Some of a girl’s premarital training, whatever her social status, would have been devoted to wool working. The homespun bridal dress was a very public proof of a girl’s aptitude in textile production. Throughout her life, from cradle to grave, a girl was associated with symbols derived from the working of wool. In the restricted lives of Athenian women, textile production was part of their daily routine, and from the early Roman republic (fifth century b.c.e.), textiles were produced by the women of the family as important contributions to the household economy. In addition to its economic value, textile work was a strong and persistent symbol of female moral virtue from antiquity into the early Middle Ages. The symbolic value of wool working in the Roman period is illustrated by the story of the emperor Augustus wearing clothes made by the women of his household—this was of little economic consequence but said much about the moral values associated with the emperor’s household. The symbolic values attached to wool working do not rule out the possibility that many households economically benefited from textile work undertaken by the women in the household. The intermingling of two sets of values—one economic and the other symbolic—associated with domestic wool production presents difficulties of interpretation. It is possible from the same set of evidence to conclude, on the one hand, that wool-working women were a cliché and that upper-class Roman women did not do wool work at all and, on the other hand, that even women of the nobility continued to produce clothes worn by members of their families.
Ideally, the Roman household was a self-sufficient economic unit whose needs were supplied from its own produce and by its own labor force. However, with the sophisticated monetary economy of the Hellenistic and Roman world, few households maintained this ideal of self-sufficiency. Several household products seem, in fact, to have been cheaper to buy at the marketplace than if they were produced within the household. An example of how household production may have become outdated within the changing economic conditions of the early second century b.c.e. is reflected in the writings of Cato the Elder, renowned for his parsimonious nature. Among his advice to wealthy landowners is a recommendation to buy clothing for slaves at the market instead of producing it on their estates. It was simply cheaper to buy some ready-made products, and there was also a thriving secondhand market in Rome. As demonstrated by the advice of Cato, textiles continued to be part of the basics of life and maintained their economic importance even if some textile items could be bought more cheaply outside the household. The ideology of the landowning, economically self-sufficient Roman citizen comprised the ideal of otium (learned scholarship)—a wealthy man who would at times withdraw from public duties and politics. However, in reality, otium and business were closely linked, and, from late republican times, otium was sometimes used as an excuse to withdraw from public duties in the city to go the country estates and see to business. However, the same ideology caused many to distance themselves in public from any sort of occupation considered unworthy. This is most forcefully expressed by Cicero in his work on occupations and duties, De Officiis (On Duties). Cicero, consciously echoing Aristotle, considered physical work and many occupations to be of low social standing and unworthy of a Roman gentleman. However, ideals and reality are often divergent, and it is well-known that many aristocratic Roman families were involved in a wide variety of trades and businesses deemed by Cicero as unworthy of their status. In order not to appear as businessmen of low social standing, they often engaged their freed slaves as front men or managers while they as proprietors of the business enjoyed a profit.
A characteristic feature of the ancient economy was slave labor. Warfare and piracy were the major sources of slaves, and slave trade was an established form of commerce in antiquity. Work in the ancient Greco-Roman world was frequently pursued by slaves of almost any age, and slaves worked in nearly all economic sectors. Many families, even those of modest wealth, had slaves who were an essential feature of the household economy. The price of a slave bought from a slave market was based on age, gender, ethnicity, and skills. Information on prices paid for individual slaves are rare, but it appears that prices varied in the ancient Greco-Roman world. Walter Scheidel has concluded that in relation to daily wages slave prices were low in classical Athens. This was probably also the case in Italy during the Roman republic. However, in the Roman Empire of the first two centuries c.e., the cost of slaves seems to have been fairly high. For the individual purchaser of a slave, the outlay was an investment expected to render the owner future income through work. Slaves had no legal right to marriage or families of their own, but it was an established custom among slave owners to encourage them to have families. The children of slaves were thought to provide stability in a household and were the property of the owner of their parents (i.e., slave children could be acquired without having to buy new ones). Such children could be developed as an economic resource by their owner: they could be sold at any age, or they could be trained for a job that would result in future income for the family. Slaves could also be hired out to work for someone else and still create income for the family.
It was not only slaves who worked but also male and female adults and children in many families of freeborn and freed slaves. With the exception of the elite families, it seems plausible that work of all sorts must have constituted an important part of the everyday lives of all members of the family, regardless of age or gender. In antiquity and the early Middle Ages, agriculture formed the economic basis of society. With agrarian production such a pervasive element in the economy, the financial well-being and living conditions of many families were closely related to agricultural production and animal husbandry. Many people worked in agriculture or related areas, especially those living in the countryside or agro-towns as land owners, slaves, or hired workers. The number and status of people involved in agricultural work is a matter of some controversy, but obviously it embraced both adults and children of both sexes. According to Varro, writing in the first century b.c.e., many poor Roman families relied on their children for agricultural work. Other Roman writers express similar views: children were capable of helping with farm work, especially unskilled tasks that needed little or no training and could be undertaken from a very early age. A parallel to this practice can be found in modern developing countries, where many owners of small farms rely on the work of the whole family, including their children.
Over the course of the Hellenistic and Roman periods, a higher degree of specialization in production was developed. Very specific job titles attest to this development. Job titles were used primarily by Roman slaves and freedmen and appear in the epigraphic evidence for the first time in the third century b.c.e. In this period, a monetary economy was being established in Rome. The occurrence of job titles is particularly noticeable in urban contexts of Roman Italy, especially in the city of Rome itself. The evidence for job specialization increases over time and involves a variety of occupations including the production of various kinds of goods, tradesmen and shopkeepers, jobs in the service sector, and so forth by the first and second centuries c.e. Roman occupations are best known to us from inscriptions, often found on tombstones, and they mirror a wide spectrum of work and occupations pursued in the Roman world. Adults of both sexes as well as children appear in inscriptions with job titles, although male workers occur in greater numbers and in a greater variety of jobs than both women and children.
In spite of the disdainful attitude toward physical work displayed by the upper classes, for other social groups all kinds of work played a central role in their lives and in the family economy. Funerary inscriptions and visual images of work, especially on funerary monuments, reflect this view. Success in the field of work was the basis for many families’ economic status, and it probably also provided social and financial advancements for members of these families. Both men and women worked, but it is not clear if Roman women pursued the same kind of occupations as men and to the same extent. In public, women’s primary roles as mothers and housewives were more essential than their occupation or work, and wifely virtues are recurrently stressed in inscriptions while men are more often presented as family breadwinners. But, in reality, husbands and wives seem to have been involved in family businesses where children could also have a place.
In a rich family, children would have experienced material wealth and would not have worked. All other families were dependent on their children to work in order to make a living, both in rural and urban settings. Children were looked upon as the private possessions of the household and contributed to the household economy by working in the fields, in workshops, or caring for younger siblings. In this way, childhood as a stage of life involving preparation for adulthood functioned in a different way from that of the upper class. Among working families, children were part of the labor force in the ancient world, either involved in household work or participating in or learning a professional trade with a craftsman from outside the family. Evidence for apprenticeship contracts has been found among the surviving papyri in Egypt and dated to the Roman period. These contracts are concerned mostly with jobs in textile production, including several examples of weavers and some who specialized in weaving linen; there is evidence for other jobs as well, such as training for builders, musicians, and other crafts. Training for a profession normally began at an early age—probably younger than fourteen—although the exact age when a child’s training started is unclear from the contracts that survive. Training could, however, last from a couple of months or a year to several years with a maximum of between five or six years.
Funerary inscriptions provide further evidence of children’s work and display a variety of occupations pursued by children, especially boys. The documentation of girl’s jobs is more scant and usually involved domestic tasks—probably reflecting gender roles in Roman society. The same pattern can be seen in jobs performed by adult men and women, where female occupations are fewer in number and relate more to the domestic sphere. Some children worked as entertainers and could apparently start their careers early in life. Entertainers were, according to Cicero, among the lowest of the low in the Roman social hierarchy. Children working as entertainers were most likely to have been slaves or slave descendents, working for either private owners or theater companies. It would mean hard work, even for small children, and the prospect of a short life. There is evidence for child labor in antiquity that reveals the brevity of children’s lives; among examples are dancers who died between the ages of ten and twelve and a mime artist who died at the age of twelve. But there are many examples of children in other occupations who also died young, such as the gold spinner, Viccentia, who died at age nine.
The evidence of widespread child labor in antiquity has formed part of the now-controversial argument for an absence of the concept of childhood in antiquity. In his now classic work on childhood, Philippe Ariès argued that childhood as a separate stage in life was invented in the Middle Ages. However, most now recognize that childhood was seen in the ancient world as a distinct period in the life course, at least for the Roman period, although for many children it implied work, sometimes very hard work.
The political crisis in the fourth and fifth centuries c.e. led to the fall of the western Roman Empire, while in the east a more stable political system remained in place for several more centuries. In the Latin west, many novelties appeared: a new political structure with groups from outside the Roman world holding power and new economic conditions for people living in the former western provinces. New social structures were created, but some continuity from Roman times remained.
Marriage was still important for individuals and for the family to transfer family property. Unmarried and married women were still subject to fathers and husbands, and children continued to inherit from their parents. The inheritance rights of both sons and daughters that characterized Roman legislation continued into later periods in some regions of the post-Roman world but in some parts with a more restricted inheritance for females.
The later Roman Empire saw a decline of large cities that continued into the Middle Ages. Many people, in economic sectors other than farming, became more tied to their work in the late- and post-Roman period since sons were forced to follow in the footsteps of their fathers in a wide range of occupations. Thus, a job in this period may be seen as part of the family inheritance, especially for sons. In the fragmented society of the post-Roman world, the documentation of work is generally less frequent both for adults and children than for the high Roman Empire. Both iconographic work scenes and job titles in inscriptions gradually became fewer until they finally vanish from our sources, but for most people work must have continued more or less uninterrupted throughout these periods as it was still fundamental to the economic standard of most families. The gendered view of male and female work that is significant in Roman society continued in this later period. Women were still mainly supposed to engage in traditional indoor female occupations—opera muliebra, such as running the household and textile work—while men worked outdoors.
In the post-Roman world, the economic situation as a whole was very complex. However, farming and animal husbandry continued to form the economic basis of society, and many families were still engaged in and dependent on this kind of work as their financial platform. Nuclear families seem to have been normative both in late antiquity and in the early Middle Ages. The fall of the former political organization also affected the economic system and led to the collapse of the system of long-distance trade offered by the Roman Empire. In the countryside, economic power was concentrated in large farming estates—latifundiae. These were economic centers owned by wealthy families who also formed a strong political group and whose counterparts in earlier periods were the senatorial aristocracy for whom otium had been a hallmark. In the west, the former secular elite disappeared, and in the circle of latifundiae owners a new, more utilitarian ideology was created to replace the ideal of otium. The constantly increasing inflation of the period put a heavy financial pressure on many small landowning farming families who could be forced into serfdom and, thus, tied to their farming lots. Peasants who lived close to the frontiers had to produce food for the legions, and mighty landowners could force the sons of peasants to serve in the Roman army, which would affect the family economy negatively if one or several sons had to leave the farm for a longer period of time, especially during harvest. If peasant farmers could not supply the required goods or pay taxes, they became coloni—peasant farmers who in practice were serfs of the rich and powerful landowners and had to work on the latifundiae. However, the proportions of slave labor, free wage labor, and tenants on rural estates is somewhat unclear during this period, partly due to regional differences with variations from one region to another.
To conclude, in the post-Roman world the family remained the basic social and economic unit, and children were still important to the family, whose economy shaped the life of the child. As social structures in both ancient and medieval times mostly favored patrilinear systems, a family normally depended on a son for its continuity. A child was proof of the fulfillment of a marriage, and a son would give hope for the future and family continuity. Throughout this period there is scant material—both written and archaeological—connecting economic conditions to the lives of ordinary people, not least in the Roman west. This situation causes problems for the interpretation of the relationship between economic conditions and the lives of families and children during the fifth to the eighth centuries c.e., but it is still possible to conclude that marriage, family life, and work continued to be vital agents in shaping the lives and economy of both adults and children.
 Aulus Gellius Attic Nights 17.21.44.
 On the age of marriage for girls see Shaw 1987; Harlow and Laurence 2002: 90–91, 95–99; Evans Grubbs 2002: 88–91. On betrothals see Evans Grubbs 2002: 88; Harlow and Laurence 2002: 58–60; and Harlow and Laurence, in press, with an exhaustive discussion on betrothals in Roman imperial times.
 Gardner 1986: 46–47, 74–75; Treggiari 1991: chapter 11; Evans Grubbs 2002: 98–100 with examples from Roman law on these issues; Harlow and Laurence in press. For late antiquity see Kuefler 2007: 357–359 with further references.
 Cato On Agriculture 135.1.
 For children’s work see especially Bradley 1991: 106–109, Table 5.1. On Roman work in general see Cicero On Duties 1.42; Joshel 1992; Dixon 2001b. The gold spinner Viccentia, deceased at age nine, is documented by a funerary inscription, Corpus Inscriptionum Latinarum 6.9213. For further examples of working children who died young see Bradley 1991: 115.