The Guardianship of Women

The Guardianship of Women

A legitimate child was, from birth, subject to the control (potestas) of the father,1 either as filiusfamilias (son) or filiafamilias (daughter). The father (pater) was head of the familia, the basic Roman social and property-owning unit. The familia under his control consisted of his children, whether living with him or not; his sons’ children, if any; his wife, if married with manus (see p.11); and his slaves. The pater, therefore, could be the grandfather or even great-grandfather of some of the persons in his potestas; nevertheless, for convenience, ‘pater’ and ‘father’ will be used interchangeably. The familia, obviously, could include several nuclear families, living apart (those of the married sons), as well as daughters married and living in families belonging to other familiae. At the death of the pater, the children (and wife) ceased to be alieni iuris (subject to another’s control) and became sui iuris (independent). Each adult son became a paterfamilias; no woman ever did—materfamilias in Latin was merely the term used to designate the wife, or strictly the wife in manus, of a paterfamilias. ² A woman’s children, if legitimate, belonged to the familia of their father; if illegitimate, they were sui iuris.

The powers of the pater were extensive,³ and they lasted over his sons and their children as long as he lived, and over his daughters likewise, unless they previously had passed into the manus of a husband. Some of these powers, originating in a very primitive stage of Roman society when protection of the group rested on self-help rather than the rule of law, had become in their extreme form rather an embarrassment by the classical period. This applies particularly to the power of life and death (ius vitae necisque) and the powers of sale or surrender.

It was the father’s right to refuse to rear the newborn child, and the mother had no legal power to prevent this. Child exposure was practised, and was not made illegal until A.D. 374, although the evidence does not allow us to determine whether there was any discrimination against girl babies. The father had also the right, as mentioned in the Twelve Tables and in the formula of adoption by adrogatio, to punish his children up to and including the infliction of the death penalty.4 This was finally abolished in the reign of Valentinian and Valens. The authority of the pater over his children remained almost intact throughout the classical period. Although the authorities from time to time intervened to check abuses of disciplinary powers, and although the pater was expected to consult a council of family or friends before exercising severe discipline, no legal restrictions were introduced until later imperial times.⁵

Women condemned by the judgment of the state were sometimes handed over to their families for private punishment, as, for example, those condemned in the suppression of the Bacchanalia in 186 B.C. In A.D. 57 the Senate referred Pomponia Graecina, accused of ‘foreign superstition’, to her husband’s judgment. He took the advice of relatives, who acquitted her. In 154 B.C., when Publilia and Licinia, the wives of two consulars, were accused of poisoning their husbands, their relatives took matters into their own hands. Giving bail to the praetor, they judged and condemned the women and carried out their execution by strangulation. Some of these women may have been sui iuris. In any case, what these examples reveal is the state’s recognition of the continuing separate identity and authority of the family, in the wide sense. Men, once they were independent, were subject to the state’s justice; sons in potestate and women were regarded as being still to some extent the responsibility of the family.⁶

Recorded instances of fathers actually putting their sons and daughters to death are few. In the case of daughters, unchastity was typically the offence felt to merit the penalty. Valerius Maximus reports two instances. Pontius Aufidianus killed both his daughter, who had lost her virtue to her paedagogus Fannius Saturninus, and also her seducer, ‘so as not to have to celebrate her shameful nuptials’. A certain Atilius, himself a prostitute in his youth, killed his daughter because she had fouled herself with stuprum (sexual immorality). The daughter of P.Maenius, who was merely guilty of kissing her father’s freedman, got off more lightly. Her father punished the freedman, as a warning to her to save herself for a husband.⁷

Augustus’ lex Julia de adulteriis (18 B.C.) specifically allowed a father to impose summary justice on a daughter caught in the act of adultery in his or his son-in-law’s house; but, as it must be imposed immediately and as he was obliged to kill the adulterer as well and must not kill either without the other, this in effect constituted a restriction on the ius vitae necisque, and it is possible that the intention was in practice to discourage such killing.⁸

Included in the power over the child’s person was the right of sale or surrender. Originally this included the right to sell a child into actual slavery, but this was obsolete by the end of the Republic, except for noxal surrender. The paterfamilias was legally liable for the actions of his children, both male and female, as well as his slaves, and if one of these committed a delict, the pater must either make himself responsible in court for the damages, or surrender the guilty person. In order to terminate potestas, the surrender had to be accompanied by the formal procedure of mancipation, which took the form of a notional sale, repeated three times in the case of a son, while once sufficed for a slave. There was originally no distinction of the sexes in noxal surrender, though the classical jurists use the masculine, and it is assumed by some moderns that surrender of daughters had become obsolete before the end of the Republic. Justinian, formally abolishing noxal surrender except for slaves, says:

The implication is that surrender had in practice been abandoned earlier for both sexes.9

Notional sale was used classically in certain situations where it was desired to terminate or create potestas—that is, emancipation, adoption and the various applications of coemptio (see pp.12 and 17). Only men could adopt, since only men could have potestas. Adoptio in the strict sense, i.e., of someone in the potestas of another, involved the abolition of the potestas by a notional sale, which had to be performed three times for a son, but only once for a daughter. The adopter then claimed the child as his. The other form of adoption, adrogatio, was used only in the case of persons already sui iuris. In form, it was a legislative act, carried out by thirty magisterial lictors, representing the curiate assembly of the Roman people and summoned by the Pontifex Maximus. The formula used is preserved by Gellius. It was accepted that women could not be adopted by this method; there would have been little point, indeed, as the procedure, involving as it did the destruction of one familia or potential familia (that headed by the sui iuris) not lightly to be undertaken, and was intended for use when it was urgently needed in order to save a familia and also its domestic worship (sacra) by providing an heir. Women could have no direct legal heirs, in this sense, and could not found a familia.¹⁰

There is some evidence suggesting that ex-slave married couples sometimes secured the enfranchisement of children born to them in slavery. Whether they adopted them is unknown—in inscriptions, such children would not necessarily be differentiated in designation from freeborn children. If the parents did wish to adopt them, however, adrogatio was probably the only available method, since it is uncertain whether adoption of slaves was permitted. The ineligibility of women for adrogation meant, then, that the slave-born daughters of such couples probably could not be adopted into their natural families. However, adoption conferred little legal gain. The father’s patron (former owner) had a claim against adoptive children of up to half the estate on intestacy, and until A.D. 178 intestate succession to a woman’s estate did not go in the first instance to children, and after that, both legitimate and illegitimate children had a claim (by the senatusconsultum Orphitianum).¹¹

The fictitious sale was also used to emancipate the son or daughter from potestas. Again, a son was ‘sold’ three times, a daughter once. One purpose of emancipation was to allow the making of a will, which was not possible for someone under potestas. Another common reason was to fulfil the conditions of an inheritance left to the child. Persons alieni iuris had no legal ownership over property, and so any bequests to them would simply be absorbed in the father’s property. Sometimes a testator specified that this was not to happen. Pliny describes an instance. Domitia, the daughter of Domitius Lucanus, had been made the heiress of her maternal grandfather, Curtilius Mancia, on condition that her father emancipated her. Mancia evidently disliked and distrusted his son-in-law and wanted to prevent his taking over the inheritance. However, Mancia’s wishes were initially frustrated, for the child was promptly adopted by her father’s brother Tullus, an elderly and childless man, and since the brothers were operating the family property jointly, the girl’s inheritance came under Lucanus’ control after all and only subsequently into her possession when she became Tullus’ heir.12

Persons in potestate could own no property. Anything given or bequeathed to them belonged to the pater. The principle, despite its manifest inconveniences, and indeed absurdities, remained valid throughout the classical period. A son might be a grown man, with an active commercial or professional career, active in public life, even a leading magistrate, married and with children, and yet legally own nothing. A daughter might be married and a mother—even, like Cicero’s daughter Tullia, who predeceased him, have had several marriages. Ways round the difficulty were devised. The son was, like the slave entrusted with business as his master’s agent, given control over a sum of money or some property, a peculium.¹³ Soldiers were even, from Augustan times, given testamentary rights over it. Probably as a result of the existence of peculium, sons and slaves were allowed to undertake contractual obligations. Daughters were not. Does this mean that they had no peculium? Ulpian interprets the masculine gender as covering both sexes, in a passage of the praetorian edict granting actions depeculio. Pomponius speaks of a situation in which a woman draws upon her peculium to provide herself with a dowry. Gaius says the action is granted ‘especially when the woman (whether daughter or slave) is sarcinatrix (clothesmaker or clothes-mender) or weaver or engaged in any common trade.’¹⁴

This, the sort of work that dependent women might do to help the family income, could perhaps involve the need of some capital for materials and stock, if the women were working separately and not as part of a family enterprise. However, the Egyptian and Pompeian evidence for weavers shows employment, mostly, though not entirely, of slaves, in a ‘factory’ situation, with the worker supplying only the labour.¹⁵ As we shall see later, for the most part the evidence for working women seems to concern family firms, or freedwomen, who are sui iuris, or the provision of labour and services. In other words, the situation of the dependent daughter or female slave operating a business and needing a peculium would be much less common than that of the son or male slave. A married son would also need access to some income for running his house-hold; the married daughter, on the other hand, would normally be equipped with a dowry, which had its own set of legal regulations.

Real life is never so tidy as the law, and in most households, even if no specific grant of peculium was made, there must have been a certain amount of hard cash handed over by the father for the women’s personal purchases, as well as an accepted treating of various items and commodities as common household property. This sort of situation is reflected, for married life, in the detailed discussion of lawyers as to what did or did not count as a gift between husband and wife.¹⁶

The pater’s consent was necessary to the marriage of sons or daughters. In early law, their consent may not have been needed, but in the classical period it seems that a father could not force his son to marry. The daughter’s situation is less clear. Ulpian17 indicates that non-objection on her part is taken as consent, but goes on to limit her right of refusal apparently to cases where the groom is morally undesirable. Since the legal minimum age for the marriage of girls was twelve and betrothal could happen even earlier, their consent, for a first marriage at any rate, may often have been formal. It is clear nevertheless both from legal and non-legal texts that in practice older sons and daughters often took the initiative in matrimonial matters—Cicero’s daughter Tullia is perhaps the best-known instance. From the time of Augustus, they could appeal to a magistrate if their father refused to permit the marriage.¹⁸

Sons and daughters in ‘free’ marriage, remained subject to the father’s potestas after marriage.¹⁹ Until the time of Marcus Aurelius, a father could dissolve his children’s marriages even against their will. Thereafter, he was prohibited from breaking up a happy marriage (bene concordans matrimonium).²⁰

The situations of sons and daughters wishing to divorce were not symmetrical. A son married in free marriage could probably divorce irrespective of his father’s wishes. If he had married with mantis, his wife was technically in his father’s potestas and the latter must be involved. A daughter married with manus was out of her father’s control. One married in free marriage could, at least until late in the classical period, divorce only through the pater.²¹

The father’s death terminated potestas. Both sons and daughters had equal rights of intestate succession. Both became sui iuris at his death, but whereas the adult son now became fully capable of independent legal action, including the right of testamentary disposition, and acquired the powers of a paterfamilias, a woman had no familia, or, rather, ‘she is both the source and the end of her own familia’,²² since she had no potestas over her children. Her legal capacity was limited by the requirement of having a tutor, whose authorisation was necessary for a wide range of legal transactions. Furthermore, until the time of Hadrian, in order to make a will she must also go through a form of coemptio (see below).

Wives and Manus

Manus (literally ‘hand’) meant a relationship in which the wife stood in the power of the husband. She was regarded as being filiae loco, in the situation of a daughter, in relation to her husband. She had the same rights of intestate succession as her husband’s children. His power over her, though, was more restricted than that over his children. He did not have the right of life and death over her, nor of noxal surrender or sale (other than the fictitious one in a fiduciary coemptio). She could possess no property of her own; everything was vested in her husband or in the latter’s father, while he lived, and anything accruing to her by gift or bequest or in any other way during the marriage was absorbed into her husband’s property.

However, once widowed, the wife married cum manu had two important advantages; it was possible, at least by 186 B.C., for a husband to give his wife in his will the right to choose her own tutor;23 and as by entering into manus she had undergone capitis deminutio, a change of status,²⁴ she could make a will without the need of a further coemptio. These advantages might be regarded as offset by the likelihood that, husbands being generally younger than fathers,²⁵ she would have to wait longer to enter into independence than a woman married without manus. She would also be unable to invoke the protection of her pater, as she had passed out of his familia, nor, of course, did she retain any rights of intestate succession in her family of origin.

Manus could come into existence in three ways, of which two, confarreatio and coemptio, were procedures usually gone through at the time of marrying, while the third, usus, became effective only after a year.²⁶

Confarreatio²⁷ took its name, we are told, from the use of a cake made of spelt (far) in a sacrifice made to Jupiter. Its survival into the Empire was ensured by the fact that it was essential for the maintenance of the state religion, since the principal flamines and the rex sacrorum had to be born of parents so married and the priesthood must themselves marry in this way. It may have been confined to patricians

In A.D.23, because of a shortage of candidates for the office of flamen Dialis, a law was passed, based on a senatusconsultum of 11 B.C., to the effect that the wife of the flamen should come under her husband’s manus only so far as religious rites were concerned; in all else, she was to have the status of a woman in free marriage.²⁸ Tacitus cites as reasons for the unpopularity of this form of marriage distaste for the difficulties of the ceremonial and the ‘negligence’ (incuria) of both men and women. This probably refers to their lack of interest in the maintenance of the priesthood. Further reasons he mentions are not only that it involved the wife’s entering into manus but also that the flamen himself was removed from paternal jurisdiction. We should not be justified, therefore, in supposing that the apathy or antipathy lay mainly among women.

Coemptio²⁹ for matrimonial purposes (to be distinguished from coemptio fiduciae causa)³⁰ was a form of notional sale of the woman. If she were sui iuris, the consent of ‘all her tutors’ was needed, since all her property would pass with her; responsibility for her existing debts, however, remained with her.³¹ Two instances of coemptio in one family in the first century B.C. are known from the so-called Laudatio Turiae.³² One of these, between the parents of Turia’, was apparently contracted some considerable time after the marriage. Gaius, in the second century A.D., speaks of it as a living institution, but it is likely to have been rare even then. Paul speaks of conventio in manum in relation to Augustus’ laws on adultery; Ulpian mentions it in relation to a pronouncement of Antoninus and Commodus; for Servius in the fourth century it is already a practice of the past.³³.

Usus involved no ceremony. After one year of marriage, a wife passed into the manus of her husband, unless, as provided in the Twelve Tables, she stayed away for three nights, repeating the manoeuvre every year. This method of creating manus was already obsolete by Gaius’ time; known to Cicero, it was possibly no longer automatic in his day. It was abolished by statute, possibly under Augustus.³⁴

Watson (1967:21–23) suggests that the three-nights’ rule finally disappeared around the end of the first century B.C., being replaced by a requirement (based perhaps on an interpretation of another clause in the Twelve Tables about the usucaption of a woman’s property) for the authority of the tutor legitimus to be given for usus, and that this requirement was then extended to all tutors and to patres as well. In other words, instead of contracting out of manus deriving from usus, it would be necessary to contract in.

The avoidance of manus, then, is attested from the time of the Twelve Tables—too early for it to be attributed to a ‘humanistic’ trend in Roman family law, still less to feminine rebellion.35 Like so much family law, both in Rome and in other ancient societies, it has to do with the transmission of property—in this case, probably with a desire to try to keep the property of the familia as intact as possible.³⁶ Dowry was probably not recoverable at the time of the Twelve Tables,³⁷ whether a marriage ended in death or divorce, nor was any legacy bequeathed to a wife in rnanu. Even if left sui iuris by her husband’s death, she would be unable to make a will without the consent of her tutor, probably a close relative of her husband. In free marriage, however, even if dowry was not yet, at the time of the Twelve Tables, returnable, the wife’s pater assumed any property accruing to her during his lifetime, and after his death, though retaining her rights of intestate succession in her family of origin, and henceforth possessing property in her own right and independently of her husband, she was in the tutorship of her agnates (unless her father had made a will and provided otherwise) and unable without their consent to make a will which might bequeath her property away from her family of origin (e.g., to her children).³⁸

The ascription of motives, however, in anything to do with the transmission of property through Roman women is always dangerous, because, private sentiments apart, the interests of the man as father tended to conflict with those he had as husband. To have a wife in manu would secure him more property for the familia; to have a daughter in free marriage would tend to prevent property going out of the familia, at least in the early state of Roman law. As we shall see later, sentiment gradually gained ground, though never entirely ousting the claims of the familia.

Women and Tutors

All children with no pater were required to have a guardian, tutor impuberis. For boys, this tutelage ended at the age of fourteen, and they became legally independent. For girls, the tutela impuberis ended at the age of twelve, only to be replaced by the tutela mulieris. An adult woman (mulier) who became independent on the death of her father or husband was also required to have a tutor.³⁹

Tutors were appointed in a number of ways, the three principal being the intestacy of the father or husband, by will and by magistrate’s appointment. The oldest type attested, and the most significant as an indication of the original purpose of the institution, was the tutor legitimus. Where the father or husband had made no provision in his will, the tutela was assigned to the male agnates, either all or the one nearest. For the daughter, this would usually be her brother or paternal uncle, or even her cousin in the paternal line. If she had been married with manus, the most likely would be her husband’s brother, or even her own son. Attested in the Twelve Tables, this rule was not abrogated until the reign of the emperor Claudius.⁴⁰ If there were no agnates, the gens could claim the tutela, and did as late as the middle of the first century B.C. The woman commemorated in the Laudatio Turiae had resisted a fraudulent claim of this sort.41

A freedwoman had no agnates, as she had no pater, but on manumission her patron (former owner) became her tutor legitimus. A woman emancipated by her pater could have him as her tutor. Strictly, he was her tutor fiduciarius (see below), but as the ‘manumit-ting parent’ (parens manumissor)⁴² he was regarded by the jurists as having tutela legitima, similar to that of a patron, with the difference that the patron’s male descendants inherited a tutela legitima, while that of the parens manumissor became fiduciary in the next generation. Otherwise, the tutor legitimus normally appeared only in cases of intestacy.

A tutor testamentarius, or dativus, was one appointed by the will of the father or husband. The latter (though apparently not the former) could allow the woman to choose her own tutor, a right first attested for the year 186 B.C. This option could be limited or unlimited—i.e., the woman could, according to Gaius, unless specifically limited, change tutors as often as she wished.⁴³ ‘Obviously,’ says Schulz, ‘a woman chose only a person whom she could rely on to raise no difficulties about giving his auctoritas’⁴⁴ The question must be asked: why, in that case, should Roman men be willing to allow such unfettered control and disposal of property to women married with manus while withholding it from daughters, married or unmarried, who had been in their fathers’ potestas, and from freedwomen? Were all women in the latter categories regarded as less sensible and responsible than manus-widows? In any case, with the progressive decline in manus-marriages, the women able to benefit, even supposing that all husbands in that category gave them the choice, would become a diminishing, and ultimately negligible, proportion of the whole, and so even the unlimited possibility of changing tutors, mentioned by Gaius, would represent no very substantial accession to the ‘emancipation’ of women in general.

The original motivation of the provision that a widow might be allowed choice of tutor (bearing in mind that we do not know how long before 186 B.C. it had been available) may be connected, like the decline of manus, with the preservation of the property of the familia. A woman in manu acquired no property—it was absorbed in that of her husband—and, except for certain rights over her dowry, what she received by his will depended on his generosity. A remedy for nearest relatives who felt unjustly treated in a will, the querela inofficiosi testamenti (‘complaint of unduteous will’)⁴⁵ was available to her, but this, though established by the time of Trajan, may not have originated until almost the end of the Republic. So, in a sense, the husband’s familia stood to lose only what he had decided he was willing to spare. As to the risk of the woman choosing a pliable tutor, one should not overlook the ties of affection with her family of origin, or the possible pressures exerted by it, or, indeed, the likelihood that, for many widows, specially the younger ones, the men best known to them would be those of their own original families (and not of their husbands’). The choice of tutor, then, might often fall on one of the woman’s natural relatives. As husband, the Roman took a limited risk of the familia losing; but as father, brother, uncle, etc., he probably expected it to gain, especially as the woman could now acquire further property in her own right and since if he as tutor consented to her making a will he might avoid the property’s all going back eventually to the husband’s family.

A decree of the Senate passed in the time of Marcus Aurelius and Commodus (A.D.I75–180) prohibited marriage between a female ward and her tutor or his male descendants, to prevent any concealment of mishandling of the property. An exception was made where the girl’s father had betrothed her to her guardian or expressed a desire for the match in his will. Whether, conversely, a husband was banned from being tutor to his wife, except as appointed in the father’s will, is not known. Examples of husbands as tutors are found in Roman Egypt, both before and after the date of the decree, but these may have been influenced by Greek practice.46

If a tutor had been appointed in neither of the above ways, the lex Atilia (c.210 B.C.) provided for appointment at Rome by the urban praetor and a majority of the tribunes of the plebs. A freedwoman with a woman patron had a tutor appointed in this way. From the time of Claudius, the consul also could appoint, and later emperors extended the function to other magistrates in Italian towns and Latin towns and colonies. In the provinces the governors could make appointments under the leges Juliae et Titiae (mid-first century B.C.). Egyptian examples indicate that the woman herself might suggest a suitable candidate: To Claudius Valerius Firmus, prefect of Egypt, from Aurelia Ammonarion. I ask you, lord, to give me as tutor Aurelius Plutammon, according to the Julian and Titian laws and the Senate’s decree.’ The prefect confirms the grant, adding ‘providing that this is not to the exclusion of a just tutor’ (meaning, presumably, a tutor legitimus).⁴¹

Women requiring to be assigned a tutor by this method would include those whose fathers or husbands had made no testamentary provision for a tutor and (before Claudius) who had no agnates; also freedwomen whose patron had died with no male issue;⁴⁸ and women whose tutors had died or undergone capitis deminutio, by captivity or in some other way.

A magistrate could also appoint a replacement tutor,⁴⁹ e.g., when a lawsuit was being brought against an existing tutor, or when the temporary absence of a tutor was impeding the transaction of legal business. Replacements were not allowed for tutores legitimi e.g., for agnates (before Claudius), manumitting parents, patrons or their sons. Freedwomen would be the largest group of women affected by this ban, and they were allowed even temporary replacements only for important matters affecting their property, such as the acceptance of an inheritance or the creation of a dowry. The grounds for the ban, as for certain other exceptions in their favour, was probably that the interests of tutores legitimi were involved, since they had succession rights. Since no lower limit seems to have been set for the distance (and so the duration) of the absence, this magisterial replacement is usually interpreted as a mere device for change of tutor, underlining the unreality of the tutela in classical law.

Other methods of changing tutor were by cessio and by coemptio fiduciae causa, both of which represented a real surrender of the succession rights of tutores legitimi. Cessio was open only to the tutores legitimi of women, not to those of minors. They were allowed to make formal surrender of the tutorship to another person; on his death, or that of the original tutor, the tutela reverted, in the former instance, to the original tutor, in the latter to the person next in degree to the original tutor.⁵⁰ In the interim, however, the woman might have obtained the new tutor’s consent to such radical action as, e.g., the making of a will.

Coemptio fiduciae causa51 was a notional sale of the woman, with the tutor’s consent, to a man of her choice, who then manumitted her and became her tutor fiduciarius. As we have seen, when this means was used to end potestas, with a slightly more involved procedure, ending up with the pater as tutor, he was regarded as a tutor legitimus, and so protected the family’s rights of intestate succession. Before the time of Hadrian, it was a necessary condition, for any woman who had not otherwise undergone capitis deminutio (e.g., by manus-marriage or manumission), in order to be qualified to make a will. It was not a sufficient condition, however; the tutor’s consent was still necessary for the making of the will, and could be compelled—but only if the tutor was not a tutor legitimus. The wider application of coemptio fiduciae causa was simply as a device for changing tutors.

While the duties of a minor’s tutor could include the administering of the ward’s property, those of an adult woman’s tutor consisted of interposing his auctoritas, i.e., giving or withholding consent to certain of her actions. This was required if the action were of a kind that might diminish the property—alienation (including manumission of slaves), undertaking contractual obligations, promising a dowry, marrying with entry into manus, accepting an inheritance (since that could involve liabilities) and making a will.⁵²

Authority was required for alienating only the type of property classed as res mancipi, namely, slaves, oxen, horses, mules, asses, land in Italy both urban and rural (which included also any buildings on the land)⁵³ and rustic servitudes—that is, the land itself and the power of man and animal needed to work it, which were the bases of production in the peasant economy.⁵⁴ Anything else, a woman could dispose of freely, and she was free to purchase what she could.⁵⁵ She could sell sheep, goats, poultry, jewellery, clothes, furniture, houses and land outside Italy—in short, everything that was res nee mancipi—and she could lend money. Faustilla, who appears several times in graffiti at Pompeii, seems to have been a pawnbroker, receiving such items as earrings and cloaks as pledges for small loans.⁵⁶

Obviously, the well-to-do would be most affected by the need for tutor’s authorisation, especially in a society in which land was the principal form of wealth. Of the landowners owning clay-yards in the Roman area in the second and third centuries of our era thirty per cent were women, of whom ten (from a total of about fifty) were members of imperial families and seven possibly of senatorial rank. As many women as men proprietors are known from the time of Antoninus Pius, and inheritance from father to daughter is common.⁵⁷ Domitia Lucilla, of whom we have already heard, inherited a clay-yard in A.D. 108 and bequeathed it in due course to her daughter, who became the mother of Marcus Aurelius.⁵⁸ In three recently published papyri from Oxyrhynchus we find three sisters, apparently sharing family estates with their brother, a local gymnasiarch and banker, and contracting with a potter to make jars in their workshops, from materials supplied by them, apparently as containers for the wine produced on their estates.⁵⁹

The requirement of tutorial consent for marriage and for the creation of a dowry reveals the original concern of the Romans to control movement of property between familiae. Consent was apparently not needed for marriage without manus even in the early Republic, and by Cicero’s time consent was needed for the establishment of manus both by usus and coemptio.⁶⁰.

As stated above, a valid will could not be made without a tutor’s authorisation, and until the time of Hadrian it was also necessary for the woman to undergo capitis deminutio. The reason for the requirement of capitis deminutio is obvious, namely to break agnatic ties. Buckland and Kaser both derive it from the primitive tutela of the agnates, at the period when they necessarily inherited the property. Watson, however, points out that it can scarcely predate the introduction of the will made per aes et libram (again, a kind of notional sale), the only form originally open to women, and the only one of the three ancient procedures of will-making surviving into the classical period.61 Gaius believes that the procedure per aes et libram was introduced later than the other two forms.⁶² It is not mentioned in the Twelve Tables, and views vary on the date of its introduction, yet the agnates’ monopoly of the tutela (of minors as well) was already broken by the time of the Twelve Tables, which provided for testamentary tutors.

The combination of a tutor from outside the group of the agnates and an available testamentary procedure would make it easier for women to leave property away from agnates. It must be remembered that, for a widow who had been married with manus, the agnates in question would be those acquired through her husband. Watson’s view⁶³ has much to recommend it, that the original purpose was to allow women who had married with manus, at a time when this was still common, to leave property back to their natural relatives, but that, as the rule was expressed in terms of capitis deminutio, it was open to other women to satisfy the qualifications by coemptio. From a humane point of view, this would, for example, enable the wife married without manus to make bequests to her husband and children—and, incidentally, also to her own family, since coemptio, other than to her father, would have extinguished the agnatic succession.

The tutor’s consent was also required for any actions tending to diminish the property. This was originally intended to restrict the movement of property away from the familia. By Gaius’ time,⁶⁴ however, tutors, except for tutores legitimi, could be compelled to give their consent. The latter could not be compelled, because their interests were affected. They were the heirs on intestacy (and they had it in their power to ensure intestacy). Most of the women affected would be freedwomen; for them, the tutor’s authority continued to be a real and active restraint. The law operated in the interests of patrons. Other Roman women, intestacy being the exception rather than the rule among their menfolk,⁶⁵ did not usually have tutores legitimi, and for them the main advance towards independence in controlling their property came, not with Claudius’ abolition of the tutela of their agnates, but much earlier, firstly with the supersession of control by the agnates by provision for testamentary tutors and later—tantalisingly, we do not know how much later—by acquisition of the means to compel tutors⁷ consent through application to the praetor. The principle of discouraging dispersal of family property was maintained in other ways than through the tutela.

Augustus’ social legislation drove a coach and horses through the concept of tutela. By the provisions of the lex Julia (18 B.C.) and the lex Papia Poppaea (A.D.9) women were released from the necessity of having a tutor if they had three children (four, for a freedwoman). The richer and more influential could, by imperial dispensation, obtain the relief without the children.⁶⁶ Given the primitive state of medicine and postnatal care,⁶⁷ one may speculate upon whether women thought the bargain a good one, specially freedwomen, whose children born before their manumission did not count. Where the patron was also the husband, the tutorship might not be irksome, so long as the marriage went well. Daughters with a father living had to wait to be orphaned or emancipated before they could benefit from their fecundity.

Nevertheless, some women did achieve the ‘privilege of children’ (ius liberorum), and it is frequently cited in papyri from Egypt. One third-century document survives from which we gather that women claiming the right to be exempt from tutelage had to apply to the prefect’s office (or, presumably, that of the appropriate official elsewhere) to have their claim placed on record.68 Aurelia Thaisus, also known as Lolliane, says: