Inheritance and Bequest

Inheritance and Bequest

From an early period women had extensive rights as heirs on intestacy. Their power of bequest, however, was severely restricted initially, the aim being to ensure the return of a family’s property to the agnates after the death of an heiress. In general, women acquired progressively greater control over the disposal of their property, while their inheritance rights were, from time to time, enhanced or decreased in certain respects, though still remaining substantial. Much of this change, however, was incidental to their sex and arose out of the needs of a changing and developing Roman society. While the agnate system of succession lost considerable ground to the cognate, the structure of the Roman family and of Roman property-holding remained essentially patriarchal throughout the classical period.1

In any discussion of ‘rights’ however, it is as well to bear certain things in mind. First, rights were mainly rights of inheritance on intestacy; those of heirs-at-law (of either sex) against a will were relatively limited and consisted essentially in the right, in certain specified circumstances, to make a claim which the praetor might or might not grant. Secondly, while rights on intestacy were initially established with reference to the agnatic familia (though weakened progressively in favour of the cognates) the actual testamentary habits of patres may have followed quite different principles. So, for example, a wife (if not in manu) ranked very low in the praetorian rules of intestate succession and had no claim at all in civil law; but in practice Roman men seem to have liked to leave their widows suitably provided for and capable of maintaining their standard of living. The interests of children might be safeguarded by leaving their mother only the usufruct of the property, but it was not uncommon for the mother to be made an heir.² A daughter, on the other hand, ranked very high among heirs on intestacy, but might sometimes receive a relatively small share of her father’s patrimony by will, specially if she had brothers, in consideration of her having received a dowry. A son, though, was likely to be the chief heir. Sometimes, the will made a son heir, while the daughter figured merely as recipient of a legacy, which might be specified as intended for her dowry. In one late Republican will (D. 33.1.22) an heir was instructed to pay the testator’s daughter a fixed sum every year, while she was unmarried. Several examples are cited in the Digest of wills in which male testators institute possible posthumous children as heirs; in each case the expected child is to receive a substantially larger proportion of the estate if it turns out to be a son than if a daughter is born. Two common patterns of bequest are attested from Roman Egypt. In one, the children, irrespective of sex, receive equal shares; in the other, some preference is shown to the oldest son.³

It will be convenient to look first at some of the rules governing testamentary disposal of property, and then at succession on intestacy.

Forms of Will

The little that is known about the early history of testamentary disposition in Roman law suggests that originally the making of a will was exceptional, confined to men and undertaken only when there were no sui heredes (immediate heirs on intestacy). Of the two earliest forms of will (both out of use by the later Republic) one, made at a special meeting of the comitia calata, an assembly of the Roman people, held for that purpose twice a year, amounted to the adoption by adrogatio of a named heir, to take effect at the testator’s death. The other, in procinctu, was essentially an emergency form of the same procedure, being executed before the mobilised army on the point of battle. Obviously, civilian emergencies were not adequately covered by these, and so, Gaius tells us (II. 102), a third form was developed, testamentum per aes et libram, by which, originally, the estate of someone threatened with sudden death was mancipated by formal sale to afamiliae emptor (buyer of the familia), who was both heir and executor.4

All these procedures were designed to supply the lack of sui heredes; if no will was made, the estate would revert to the agnates.

The first two procedures were available only to men, since women had no standing in the comitia and were not part of a citizen army. In any case, women had no sui heredes and, unlike men, did not become heads of familiae. It seems likely, then, that they had no means of testamentary bestowal before the institution of the testamentum per aes et libram.

This, originally an emergency procedure, became the regular and customary form of Roman will. The familiae emptor became a formal figure, distinct from the heir. The heir or heirs designated need not be the sui heredes, but a rule was applied, possibly from the first century B.C. onwards, that if the sui heredes were passed over without being specifically excluded, the will could be made void, and even if they were excluded, it might be challenged. Even without the formality of the mancipation, a written testamentary disposition sealed by seven witnesses could be accepted by the praetor, who could award ‘possession according to the terms of the will’ (bonorum possessio secundum tabulas), with the technical consequences, whose detailed effects need not be examined here, that the beneficiaries were not heredes.⁵

Although women could be witnesses in courts of law, they were excluded from being witnesses to wills—along with minors, slaves, deaf-mutes, lunatics, prodigals and convicted adulterers.⁶

The Right to Make a Will

A man sui iuris was a pater and, from as far back in the classical period as our knowledge extends, had the right to make a will. Women, however, acquired freedom to dispose of their own property after death only gradually. The reasons for this difference have already been indicated. The pater was the sole owner of the property of his familia in his lifetime and had absolute control over its disposal. If he made no testamentary provision, the sui heredes had first claim, and after them the agnates. Women headed no familia, and the agnates’ claim was primary. No doubt it would have been much simpler for the Romans to control the devolution of property and the maintenance of family status by confining succession entirely to the male line. They chose not to do this, and instead developed a system of property transmission in which important features were inheritance through both sons and daughters, dowries, arranged marriages, monogamy and agnate inheritance. This may have been a’strategy of heirship’ for a society which found it difficult to ensure the existence of male heirs.7

Whatever the reason, in the classical period women sui iuris could inherit, by will or otherwise, and could receive legacies. Their right to dispose of their property, however, was partially controlled by tutela. Gradually, women acquired increasing control over the disposal of their property, not only in their lifetimes but after death as well; at the same time, the succession rights of the agnates were progressively reduced. Most of the steps in the process have already been mentioned: appointment of a testamentary tutor; the availablity of coemptio (with tutor’s consent) for the purpose of making a will; the ins liberorum; the abolition of agnatic tutelage; the possibility of compelling the consent of tutors (save legitimi); the abolition of coemptio. The admission of a woman’s children, under the senatusconsultum Orphitianum, to her intestate succession further weakened the position of agnates.

However, the weakening of the rights of the agnates was not solely attributable to the increase in women’s control of property disposal; they lost ground also to the cognates, in a long historical process of movement away from the early collection of territorially based, exclusive agnatic familiae to the relatively fluid, interconnected urban society. Within this society, claims to family succession based on ‘family’ relationships in a wider sense, including blood ties through females, were also important as a means of maintaining social equilibrium by keeping wealth moving, especially between persons of the same or similar social status.⁸

Freedom of testamentary disposal had always been the prerogative of the Roman pater. Bequests to persons outside the family played, particularly in the late Republic and early empire, a very important role, both socially and politically, among the upper classes. The slackening of restrictions upon women’s rights of property disposal can be seen, not so much as a recognition of female ‘rights’, as something necessitated by the importance of maintaining this mobility of wealth. At the same time, since the structure of society remained essentially patriarchal, this freedom was not accorded to women in certain precise and closely defined circum-stances. The nature of these circumstances is significant; they were, essentially, those in which the woman’s nearest heir on intestacy was, not a cognate, not even an agnate, but a tutor legitimus, a (quasi) pater—that is, after Claudius, only a parens manumissor or a patron—whose consent to the making of a will could not be compelled. The ius liberorum interfered with this to some extent, but the principle remained in force for women without the ius (and, as we shall see, even for some freedwomen with it). We have already observed changes in some of the other features of the system: weakening in the authority of the pater over the marriage, maintenance of monogamy, but through easily dissoluble serial marriages, and changes in the concept and function of dowry.

It would not be particularly appropriate, then, to regard these changes as representing progressive stages in a deliberate process of female ‘emancipation’. The most important steps had already been taken, before the start of our detailed historical knowledge, by allowing women to be sui iuris and capable of owning property, and to be sui heredes.

No will was valid unless made by someone possessing testamentary capacity (testamentifactio). For this, certain conditions had to be fulfilled. Some affected both sexes equally: e.g., the testator must be free, of citizen status and capable of ‘attesting’ (testabilis)—so slaves, captives, lunatics, congenital deaf-mutes and those punished with infamia could not make wills.9

Some conditions, however, affected the sexes differently. Firstly, testators must be sui iuris. Not only men and women with a pater still living, but women married with manus would be excluded by this requirement. Secondly, they must have reached puberty. Here women had the advantage, being deemed to be of age at twelve years, men at 14.

However, the most important difference was the requirement of tutorial consent for a woman to make a will. This has already been discussed to some extent in Chapter 2, where the function of the tutela in maintaining some control over the woman’s property in the interests of the familia was described. Tutorial consent operated in two stages. First, until Hadrian abolished the requirement, it was compulsory for a freeborn woman to cut her links with her familia of origin by going through a coemptio, a change of status (capitis deminutio) involving a notional sale, before she could make a will. This cut the agnatic connection. On coemptio, she could be assigned to a tutor of her choice. Freedwomen were not required to make a coemptio, since they had already undergone a change of status (and had no agnates anyway), nor did daughters emancipated by a pater who was still living. If the woman was not exempt and had not made coemptio, the will was void.¹⁰

Secondly, all women had to obtain tutor’s consent to the actual making of the will, otherwise it was invalid. The only women exempted from this requirement were, under the Augustan legislation, those who had acquired the ius liberorum.¹¹

By the time of Gaius at the latest, as we have seen, the consent of tutors, except tutores legitimi, could be compelled; thus, many, and probably most, freeborn women had virtually free capacity of testamentary disposal. Two categories of women (three, until the Claudian abolition of the agnate tutela legitima) did not. One was the emancipated daughter, whose pater was her tutor legitimus; the praetorian rules maintained his right to intestate inheritance from his daughter, though a common motive for emancipation was, probably, to allow a daughter to make a will in her children’s favour. This right was not inherited by his heirs. In the other case, however, that of the freedwoman, not only her patron but the children as well (of a male patron), should he predecease her, had the right to secure intestate succession. A female patron, however, could not be a tutor, and the tutor assigned by the praetor was not a tutor legitimus. For women not in tutela legitima, obtaining tutor’s consent to a will became in effect a formality.

Even if the woman had neglected to obtain her tutor’s consent, the heirs named in the will could still succeed to the property by applying to the praetor for possession under the terms of the will. Gaius (II. 118) tells us that this would be granted so long as the will was properly sealed and witnessed and so long as there were no heirs under the civil law on intestate succession (i.e., agnates or patron).

The consent seems to have been required only to the making of the will; the tutor does not seem to have had any control over its contents. So, Gaius remarks (III. 43) that if a patron had given his consent and then found that his freedwoman had not made him her heir, then he had only himself to blame. This presumably means that, so long as the actual heir could prove that the patron had consented to the making of the will, the latter could not break it by appealing to the praetor, on the grounds of his rights as patron, for possession against the terms of the will (bonorum possessio contra tabulas). Since the patron had had the power to take everything, by preventing his freedwoman from making a will, his exclusion under a will could only be the result of his own act. In contrast, the praetor might grant possession, notwithstanding a will, to a patron of onehalf of the estate of a freedman, against any heirs other than sui heredes.12 In this connection, the ius liberorum was not an unmixed blessing for freedwomen; see further on this point below, in the discussion of patronal rights of succession.

The Institution of Heirs

A will had to provide for the disposal of an estate in its entirety; it was not possible to make bequests of part of it and leave the rest to be assigned by the rules of intestate succession. An heir or heirs must be nominated, among whom the whole estate was divided, in proportions designated by the testators. If a male testator did not want the sui heredes to be his heirs, he must expressly exclude them. Males must be disinherited individually and by name; for females, a general clause of disinherison was enough. (This was the praetorian ruling; the civil law insisted only on the naming of sons; grandsons could be passed over in a general clause). This indicates the relative importance in the eyes of the law of male heirs, who could head, and transmit property to, afamilia, and females, who could not. If there was no exclusion clause, the sui passed over could appeal to the praetor for bonorum possessio against the will. Specific legacies, to heirs or others, were either ‘prior’, i.e., deducted before the division of the estate, or were a charge on the estate, for which the heirs were liable in proportion to their shares. If the will lapsed through refusal of the heir or heirs, these legacies were protected by the praetor’s granting an action to the beneficiaries against the heirs on intestacy.¹³

Heirs were also responsible for maintenance of the family sacra (if any—they could not be transmitted by women) and for the debts—in their entirety—of the deceased. The latter requirement sometimes resulted in heirs refusing the inheritance. Sui heredes could not, strictly speaking, refuse, since they were automatically heirs on intestacy. However, they were allowed the ‘privilege of abstaining’. This would mean that the debt liability of the estate would be limited to the actual assets, which would be sold. A dutiful son or daughter might not want to bring the disgrace of bankruptcy on a dead father’s name in this way. The younger Pliny wrote (Ep. 2.4) to reassure one such daughter, his relative Calvina (to whose dowry, it will be remembered, he had contributed). We learn that Calvina’s father had died heavily in debt. Pliny, however, had paid off the other creditors so that he himself should be left as sole creditor, and he informed Calvina that he had given instructions for the debt to be written off. The intention was to encourage Calvina ‘to defend your father’s honour and reputation’ by accepting the inheritance; that is, her father’s bankruptcy was being hushed up. Pliny remarked: ‘If your father had been in debt to several people, or indeed to anyone other than me, you might have hesitated to accept an inheritance which even a man would have found a burden.’

Why ‘even a man’? Sui heredes of either sex would have been in potestate until the father’s death and unable to acquire property of their own. Extraneous heirs could be of either sex. Pliny probably had in mind certain sources of wealth for people of his class, for example, public service, especially provincial commands, and forensic activity, which were available to men and not to women.

The Voconian Law

More than once under the Republic it was found desirable to legislate in order to limit the proportion of estates which might be apportioned in legacies. The two most important laws were the lex Voconia of 169 B.C. and the lex Falcidia of 40 B.C.14 The lex Voconia provided that no legatee of either sex was to receive more than the heir or heirs taken together. That meant that it was possible to leave a single legacy, of almost half the estate. A device of this sort, known as the legatum partitionis, by which the heir was instructed to share the estate with a named legatee, giving the latter just under half the total value, was known at least as early as 133 B.C. and may even have predated the lex Voconia.¹⁵ One use was to enable someone to be effectively heir to half the estate without liability for sacra; it could also be used as a way of evading, partially at least, another restriction imposed by the lex Voconia, which forbade the institution of women as heirs. Murdia (FIRA III no. 70) made provision in her will for her daughter in this way, but made all her sons heirs.

This latter provision of the lex Voconia apparently applied only to citizens registered at the last census in the highest property class. The irregularity of the holding of the census in the late Republic will have impaired the effectiveness of the law and, although Gaius (II. 274) speaks of it in the present tense, it probably ceased to be applicable once the census itself became obsolete in Italy after the time of the Flavians. In any case, its provision on legacies was superseded by the lex Falcidia of 40 B.C., which provided that a maximum of three-quarters of the net estate could be taken up in legacies. Dowry, amongst other items, must be deducted from a man’s estate before the calculation was made.

There may also have been another provision in the lex Voconia applying specifically to women. Paul (Sent. 4.8.20) remarks, on a provision of civil law restricting agnate succession by women to full sisters of the deceased, that this was done Voconiana ratione. If this is taken to mean ‘on the principle of the Voconian law’, the remark becomes evidence for Paul’s belief about the intention of the enactors of the law, but not for the date of this particular enactment. On the other hand, the meaning may be that this provision also was included in the lex Voconia, Its effect would be substantially to diminish the possibility of dispersal of property on intestacy outside the male line, and so into other familae.

However, the most notorious and controversial provision of the lex Voconia was that which forbade testators in the first census class to nominate women as heirs. Controversy is mainly concerned with the reasons for the passing of the law. A favourite explanation in modern writers is that the law was primarily aimed against the possession of large amounts of wealth by women; it is even called ‘anti-feminist’.¹⁶ This interpretation rests largely on the surviving fragment of Cato’s speech in support of the bill (quoted in Chapter 4 above), which is taken as evidence of male resentment of wealthy women. It may be evidence of Cato’s attitude, but not necessarily of that of the proposers or other supporters of the bill. Another version of this interpretation sees the law as essentially sumptuary, like the earlier lex Oppia, and designed to check the extravagance of women; it is not obvious how its provisions could have achieved that end.¹⁷

The ‘anti-feminist’ explanation also appears, in superficially more sophisticated dress, in Hallett (1984:92 if.). She suggests that the law was passed because wealthy heiresses were regarded as ‘a threat to the Roman patriarchal social structure’.18

The nature and content of this threat is unclear. Hallett says that such heiresses were ‘in a strong position to make decisions regarding their families’ private, and even public, expenditures—and thus to influence their families’ interactions with and image in Roman society.’ Her idea seems to be either that Roman heiresses were the sole source from which their kinsmen might hope to borrow, or that Roman upper-class men funded their personal life-style and political careers not from their own property and/or personal loans, but from some hypothetical common family kitty, on the disposal of which the largest contributors (here, the wealthy heiresses) had the chief say.

Consortium, or the joint ownership of estates, was not unknown to classical law, but it existed in two forms, neither of which is apposite to the lex Voconia. One was the consensual partnership in property, of a type which could exist between other persons than joint heirs, which could be set up by a process before the praetor. This, according to Gaius (III. 154a-b), was modelled on the other, an ancient community of property, which he speaks of as obsolete, known as ercto non cito and existing between sui heredes. It came into existence automatically on intestacy of the pater and existed until the heirs took steps for the division of the estate. Clearly, it originated in an early period before methods of testamentary disposal were introduced. There is no evidence for its being common practice in the classical period; the actio familiae erciscundae was commonly used for the division of the estate. In any case, it applied on intestacy, and so is irrelevant to a law limiting rights of testamentary disposal.¹⁹

The maintenance of joint families seems never to have been usual at Rome. If it had been, it would itself have threatened the patriarchal structure of Roman society, since the stability of the joint family would have depended upon acceptance of the headship of the eldest brother. In practice, brothers and sisters were equal heirs on intestacy. As many familiae were created as there were siblings (those of the sisters were end-stopped). Each brother, on the death of the pater, became pater of his ownfamilia, with absolute control, independent of his kin of either sex, of his own property and of the finances of those in his potestas. The Roman family system is uncompromisingly patriarchal’ (Zulueta 1953: II. 177). The joint ownership of property within a family is a voluntary association. It might originally have been created by bequest, but its maintenance was voluntary. In classical law, it is treated on a par with other types of voluntary partnership, and agreements on joint ownership, such as those arising from bequests of shares in specific items of property, are not relevant to this discussion.

The idea of a wealthy heiress directly controlling the purse-strings of her brothers and kinsmen is patently absurd. She could not do so in law. At most she might threaten their expectations by her will (that is, if she had a complaisant tutor) and so possibly reduce their borrowing power elsewhere—but that is a long way short of making decisions on their private and public expenditure. However, if a woman were the preferred heir, there would be less of an estate available for distribution among men.

Did married women, then, control their husbands’ public expenditure and so their careers? Borrowing by husband from wife could and did occur (see, e.g., Cato’s speech, mentioned above, and Sallust, Cat. 35, where Catiline expresses confidence in his wife’s willingness to help pay his debts), and a large dowry could be useful to a husband, as it was to Caesennia’s, but it is difficult to see any threat to the patriarchal structure in such purely personal and individual arrangements. The dowry, in any case, of a wife married while still in potestate would not have come out of her own property; and so long as the marriage lasted, it was the husband’s property.

Let us return to Cicero’s statement. The law was passed for the sake of the practical advantage of men, but it is full of injustice to women.’ The injustice which chiefly concerns him is apparently that involved in the extreme case of the daughter and only child, who cannot be made her father’s heir by will if he has been registered in the first census class. It was perhaps excessive concentration on this type of case which led some modern scholars (references in Vigneron 1983:144, notes 20–3) to the supposition that the intention was to strengthen the agnatic tutela. The effect of the law, it was pointed out, could be evaded by intestacy, but the heiress would then automatically come under agnatic tutela. As Vigneron observes, this hypothesis is altogether too tortuous, and in any case applies only to a proportion of testators. It takes no account of various other possibilities. A testator might wish to make his heir a woman who was not in his potestas—for example, his wife, married without manus, or his sister, already sui iuris. He would have no control over the designation of her tutor. Moreover, some testators would themselves be women. A daughter might not be the only child. Even if she were, it would still be perfectly legal for the father to appoint outsiders as heirs and leave the daughter a legacy of up to half the property. Daughters in a large family stood to lose relatively little, since each individually could receive as much as the heir or heirs together in legacies; on intestacy, they would each get only a fraction of the heir’s portion, if the latter were not sui, and might even get less themselves, if there were several heirs, who were sui also.

If the intention of the proposer of the law, Q.Voconius Saxa, had been to prevent large patrimonies falling into the hands of women, he would presumably also have forbidden women’s right to be sui heredes;20 intestacy could come about in other ways than by the intention of the deceased person. Women’s right of agnate inheritance was partially restricted but, as we have seen, this may not have been under the law itself but at some later date.

The ‘practical advantage’ for men of which Cicero speaks surely refers to the importance of wealth for social and political status in the increasingly competitive society of the period in which the law was passed. In order to participate in the competition, men must have wealth available to them. Wealth in the hands of women was, in a sense, out of circulation for such purposes, and might, indeed, remain so, if women, in turn, made other women their heirs. Moreover, the new wealth coming in from foreign wars and conquest probably consisted mainly, in the first instance, of res nee mancipi, i.e., not land or buildings, though possibly some livestock, very probably some slaves, but mostly money (Shatzman 1975:63 ff. gives some examples of the fortunes made). Some of this wealth might be converted fairly quickly into land, but there would be much in the form of liquid assets. Once in female hands, these would not be subject to tutorial control and would be particularly vulnerable to dispersal.

There is little evidence from which to determine whether upperclass Romans at the period of the passing of this law tended, where there were heirs of both sexes, to leave land to the males and property other than res mancipi to the females. L.Aemilius Paullus Macedonicus, for example, had two sons, and one or two daughters, married to Q.Aelius Tubero and a son of Cato. What provision he made for his daughter(s) in his will is not known; but it is perhaps significant that at his death his sons had to sell res mancipi, land and slaves, to raise cash towards repayment of the dowry of 25 talents to his widow.²¹

Later, and at a lower level in society, wills from Roman Egypt reveal, according to Hobson (1983:320–1) ‘a tendency to leave real estate to male children and household furnishings to females’, though this is not universal. However, in the absence of male children the desire to keep property in the immediate family, so far as possible, must sometimes have resulted in females inheriting. Hobson also observes that a wife was not usually bequeathed her husband’s real property, perhaps because she had property of her own, from her dowry or from her own family.

The evidence of the Digest indicates that children of both sexes were often left as legacies land and real estate, i.e., houses with their furnishings, farms with their equipment, business premises and so on. These, of course, would have to be specified separately as legacies, since they could not be arithmetically divided. Dowry was certainly sometimes left as cash. Boyer (1965:358–9) has collected a number of instances in the Digest in which a daughter was not made heir, but was left a cash sum, chargeable upon the heir, as dowry. Mancipia (slaves) are included in one instance, but it is specified that these are due only if the girl marries within the familia. Obviously, for practical reasons, it would be more convenient to make provision for eventual payment of dowry for a daughter, unmarried at the time of the testator’s death, in cash. This does not exclude the possibility of other bequests to the daughter, for her immediate possession and use, and daughters married without manus could, of course, also be beneficiaries.

There is at least a fair possibility that upper-class Romans in Cato’s time tended, where there were children of both sexes, to leave their ‘old’ wealth, the res mancipi, to the males as heirs and the res nee mancipi, the more liquid assets, to the females, who had already formed, or were likely to, marriage ties with other familiae. Where sons were lacking, the estate was quite likely to go to the daughters, either in entirety, as heirs, or in bulk, as legatees, with a remoter male relative as nominal heir, to preserve the sacra. It might even go to the widow: Aemilia, wife of Scipio Africanus, benefited greatly in this way.22 Voconius’ law may have been meant to ensure that at least a sizeable portion of large patrimonies would always pass directly into male hands and be available for the purposes of male public life, though it did not, apparently, go so far as to prescribe the nature of the bequests which might be made to women.

We do not know enough about fertility and mortality and their effects in the first half of the second century B.C. to determine the extent, if any, to which the maintenance of social and political status by marginally upper-class males was felt to be imperilled.²³ However, the desire to ensure an adequate financial backing for these male activities, rather than absolute hostility to women’s ownership of wealth per se, seems a plausible motive for the law. The opportunities for women to acquire wealth were still substantial. Women could not be heirs of men in the top census class; but they could receive up to half the value of an estate in legacies, the restriction on legacies affected both sexes, and the (possibly later) restriction on agnate inheritance left the rights on intestacy of daughters, wives in manu and sisters unaffected. Incidentally, the prevalence or otherwise of manus-marriage by the time the law was passed is unlikely to have made much difference overall to women’s independence, given that daughters and wives were the most likely heirs. With manus, the effects would be that wives and unmarried daughters became sui iuris, but were in the tutela of the deceased man’s family or of a tutor designated by him, while married daughters would be in the manus of their husbands. Without manus, the wife would either be in potestate or already sui iuris in the tutela of her own family or as designated by her pater, while daughters, married or unmarried, would be in tutela as determined by the deceased, or agnate tutela. In both situations, it was open to the pater to determine, in two out of three cases, the nature of the tutelary control he wished to impose.²⁴ As for the disposal of property in their possession, except for the ban on appointing other women as heirs, and the restriction (applying to men also) on legacies, women’s rights were the same after the passing of the law as before it.

It was not too difficult to find ways around the law. The curious story of the ‘dowry’ which, as mentioned in Chapter 4 above, Scipio Aemilianus, as heir, was required to pay to his adoptive aunts on their mother’s death may have concealed legacies which amounted to more than his share. More definite examples are found in Cicero. Someone who (whether deliberately or not) had not been registered in the census was not liable to the law. So, he says, ‘no law forbade’ P.Annius Asellus to follow the dictates of nature and make his daughter his heir, and, indeed, legal equity, praetors’ edicts and current legal practice all tended the same way. However, Verres, the corrupt praetor in Sicily, declared that under the terms of his own edict the law did apply, and, Cicero says, he took a bribe from the man designated heir in second place. Later, however, says Cicero, many such wills, of persons not in the census, were successful. He mentions particularly a wealthy woman, Annaea, who made another woman her heir.25

Less successful was an attempt, also related by Cicero (defin. 2.55), by a certain Q.Fadius Gallus to bind his heir P.Sextilius Rufus by a fideicommissum to pass the entire inheritance over to his, Fadius’, daughter. Fideicommissa were not legally enforceable during the Republic; they became so under the Principate, with gradually increasing protection for the heirs against external claims on the estate, and the use of the fideicommissum, according to Gaius, seems to have become the routine way of avoiding this restriction.²⁶ Sextilius, however, was not bound by the fideicommissum, and he accepted the inheritance and denied the trust. Many eminent persons, says Cicero, thought the daughter should be given no more than she was entitled to under the lex Voconia (he may mean one-half, as a legacy, rather than nothing); but if right-thinking men had had their way, Sextilius would not have touched a penny.

Augustan Legislation on Inheritance

As part of his programme to encourage marriage and the procreation of legitimate children, Augustus, by the leges Juliae et Papiae, imposed certain limitations on the ability of the unmarried (caelibes) and childless (orbi) to receive under a will.²⁷ These rules did not apply to persons who had a right of intestate inheritance at civil law, that is, to relatives up to the sixth degree. Some relatives by marriage were also empowered to receive bequests—not only husbands and wives, but parents-, sons- and daughters-in-law, stepchildren and step-parents. We have already seen that divorcées and widows could receive under the wills of others only within the ‘period of grace’ after the end of their marriages.

The rules applied more or less equally to both sexes. They applied only within certain age-limits, namely 20–50 for women and 25–60 for men, related to the capacity for reproduction. They did not affect the majority of bequests, which were between relatives. The leaving of bequests to friends outside the family was conventional mainly among the wealthier classes.

Although husband and wife could receive bequests from each other, these were limited to one-tenth of the estate, unless they were outside the prescribed age-limits, or related within six degrees, or had the ins liberorum, or the husband had been away on public service. One living child of the marriage, or one that had survived to puberty, sufficed to give capacity to inherit. They also received some ‘credit’ on account of children of other marriages.

The rules were complicated and inevitably produced anomalies and contained loopholes, some of which were removed by later emperors. A series of senatusconsulta tried to cope with the problem of the person who did not marry or have children at all within the age-limit and who, it was felt, should not automatically regain capacity after reaching it. If a husband aged over 60 married a wife under 50, then they benefited. Apparently, however, some women over 50 (and so very unlikely to bear children) had been marrying themselves to men under 60, and hoping to become eligible to inherit. Claudius closed that loophole. Nero took action against fraudulent adoptions.²⁸

Domitian banned women of immoral life (probrosae) from receiving legacies; previously, even if married, they had had a limited capacity, being allowed to receive only one-quarter.29 (He also forbade them the use of litters—a more direct and frequent inconvenience.) So, convicted adulteresses, prostitutes, actresses and the like would not benefit, regardless of how many children they had. Nor, presumably, would concubines, not because they were regarded as immoral, but because they were unmarried and their children were illegitimate.

The rules on intestate inheritance by female patrons were also affected; these will be described below.

Some people attempted to evade the ban on receipt of legacies by passing on money in other ways. Under Vespasian, gifts under fideicommissa were brought under the same rules as legacies, and by the time of Hadrian gifts mortis causa, including those between husband and wife, were allowed only in so far as the recipient had capacity to receive legacies.³⁰

Wills and Social History

As well as the exhaustive, and exhausting, discussion of the minutiae of testamentary law in legal sources, a certain amount of literary and epigraphic evidence of actual wills survives. Cicero and the younger Pliny³¹ are notable among literary sources. Egyptian papyri have been the subject of special studies, notably those of Montevecchi (1935) and Hobson (1983), and epigraphic records of charitable foundations and other bequests by wealthy men and women feature prominently among the evidence drawn upon for the masterly study of the economic context and social application of wealth in the Roman world by Duncan-Jones (1982). There is also a useful collection of wills from various sources by Amelotti (1966).

Several wills have already been mentioned in the course of this book. Cicero and Pliny give us glimpses of the arrangements made in upper-class Roman society, in which legacies to one’s friends and social equals were clearly a normal and, indeed, expected provision. Pliny (Ep, 10.94) specifically mentions, in support of his request for the ius liberorum for his childless friend Suetonius, that this would enable the latter’s friends to express their consciousness of his merit (i.e., by making legacies which he would then be qualified to receive).

A wealthy banker from Puteoli left an estate to Cicero and at the same time a legacy of 50,000 sesterces to Terentia, which, Cicero noted with satisfaction, was to be deducted from someone else’s share of the estate, not from his own (ad Att. 13.46). In 50 B.C. a certain Livia made Dolabella heir to one-ninth of her estate, on condition of changing his name (i.e., having himself adopted into her gens). In Cicero’s view (ad Att. 7.8), there are two main-ations to be balanced by Dolabella in deciding whether or not to accept. One is the possible effect on a young nobleman’s political career of his changing his name because of a woman’s will. The other is the value of the bequest. That there is any hesitation suggests that Livia must have been very wealthy.

There is considerable evidence under the empire of wealthy testators choosing to perpetuate their memories by setting up alimentary foundations in towns in which they were interested, or by providing towns, collegia or other bodies with which they were associated, usually as patrons, with endowments for such things as regular distribution of wine and oil, or annual banquets. Several women figure among these benefactors.32

The commonest recipients of specific legacies, however, at all levels of society, were members of the testator’s immediate family and household. Sons and daughters were quite likely to receive the family house, or a part-share in it (shared houses appear frequently in papyri), with its furniture, and perhaps farms and shops, with all their stock and appurtenances (instructa), as a going concern. This could also be a way of providing a ‘pension’ for the old family retainers, i.e., freedmen and freedwomen. Like the children, they might individually be left houses or estates, or a share in them; the death of a patron or owner could mean loss of their livingaccommodation. For instance, a third-century will from Egypt (FIRA III no. 10) manumitted a thirteen-year-old girl, freeing her from all patronal obligations, and making her a gift of her peculium. As it was apparently not contemplated that she should live with the testator’s son, she was also bequeathed the fourth-part of a dwellinghouse, with a few sticks of furniture, and the proviso that she should sublet to no one other than her own brother. Provision might also be made, individually or collectively, for them to receive food and clothing (cibaria and vestiaria), either out of the revenue of a sum of money, or as usufruct of an estate. Pliny actually gave a small farm during his own lifetime to his old nurse, to provide for her support, but was obliged to arrange for someone to work it on her behalf, and presumably give her an income from the proceeds (Ep. 6.3). A whole title of the Digest (34.1) is devoted to bequests of this sort. Sometimes a tract of land was left to the freedmen and freedwomen collectively, with the specification that it remain in perpetuity in the possession of them and their descendants. One of the best-known examples is the bequest by Junia Libertas at Ostia of an enclosed area, including houses, gardens (i.e., produce-gardens) and shops, left for the use of her freedmen and freedwomen and their descendants inalienably; if the familia should die out, the property was to revert to the community of Ostia.³³

As we have already seen, widows commonly received, not only personal effects provided for them by their husbands, but also legacies of maintenance, either directly (penus) or as the usufruct of part of the estate, either for life or for a specified period, and the details of such bequests afford us glimpses of the style and standard of living of the parties involved.

Wills are potentially a useful source of evidence for the study of social and economic history. Taken in isolation, they may be misleading or relatively uninformative. (1935:73) is pessimistic about the value for statistical studies or as social documents of the wills in the papyri and indicates some of their drawbacks. For instance, even if the document itself is complete, we cannot always be certain that all members of a family have been specified; or the bequest of an estate in toto to one person may leave us uninformed about the number of slaves, the real estate and other property included, etc. Hobson (1983), however, shows that it is possible to some extent, by combining wills with other kinds of evidence, such as census declarations, tax lists and receipts of sale, to draw some tentative conclusions about the types of property mainly owned by women in Egypt and its sources, and about the nature and level of their economic activity. Making a special study of Socnopaiou Nesos, a commercial rather than an agricultural centre, she concluded that women’s involvement in the economic life of the community consisted mainly (though not entirely) in the acquisition of land and houses by inheritance, and also partly by way of dowry, from their families, and its disposal by sale or by will. Her conclusions are not necessarily valid for the whole of Egypt, let alone for the Roman world, and her sample is a small one; nevertheless the approach has value.

A man’s will, as we have seen, could be challenged if he had failed either to institute or expressly to disinherit his sui heredes. The praetor’s rules also allowed other liberi to make a claim on this account, which would allow not only other emancipated children, but perhaps also former Vestals, to claim. Whether the latter, or flamines, could claim while still in office is uncertain; the relevant text (D. 37.4.1.6) says only that liberi, whether they have been emancipated or passed out of potestas in some other way, are admitted to succession by the praetor. Omission of a son, without his being specifically disinherited, made the will void. A daughter omitted had merely the right of accretion, i.e., she could share equally with other sui but she could not exclude extraneous heirs; against them she had a right only to half. While it was for much of the classical period possible for liberi in fact to exclude extraneous heirs entirely, since the praetor might grant the latter possession only sine re (in title only, without effect against civil law heirs), Antoninus Pius issued a rescript allowing women heirs to receive no more than they would by the rule of accretion.34 A woman’s will could be challenged by her parens manumissor or her patron, if the necessary consent had not been obtained.

Provision for Posthumous Children

A man’s will could also be broken if no provision was made for postumi, born within ten months of the father’s death. An expectant mother could make a claim to the estate on behalf of the venter, the child in the womb, both in these circumstances and on intestacy. She was allowed maintenance from the estate; this was not reclaimable if she had a miscarriage. Customarily she asked for one or more curatores to be appointed, both for the child and the property—usually from the relatives and friends of the deceased. Since the curator of the property was personally responsible to any creditors, he must be solvent.³⁵

If there were other heirs standing in the same grade of priority as the expected child there might be problems about determining the proportion of the estate due to each of them in the meantime, since there was the possibility of multiple births. Paul, though sceptical, cites some remarkable examples—seven at one birth (to ‘many’ women in Egypt). One woman was brought specially from Alexandria to be shown to Hadrian. She had quintuplets, of whom four were born together, and a fifth 40 days later. Paul himself recommends that the praetor strike a happy medium and make an interim award on the basis of possible triplets, since that could happen; the Horatii were a famous example.³⁶

Once possession had been granted to the venter, Hadrian advised in a rescript, any accusation of adultery against the woman should be postponed, to avoid prejudice to the child’s interests. The possibility of fraud by women was recognised, whether by feigning pregnancy, or arranging to transfer the property to another, and the praetor’s edict specified what actions could be brought (D. 25.5 and 6). The woman could be punished with infamia.37 Rules were introduced, as we have seen, under Hadrian requiring the expectant mother to notify interested parties and to allow her pregnancy to be monitored. Poor Petronilla suffered under this edict. However, it did allow the widow some protection against harassment. If the family had not taken advantage of the procedure, then presumably they would be unable subsequently to challenge the child’s claim to the estate.

A child, posthumous or otherwise, who was challenged as being supposititious or illegitimate could be allowed possession by the decree of the praetor, under the Carbonian edict.³⁸ The mother could demand a decision from the praetor. The edict was available only to children under the age of puberty, and the claim must be made within a year. Unless the praetor was satisfied that the allegations were proved, he was expected, in the interests of the child, to award possession and defer a final decision until the child reached puberty. Petronilla’s in-laws may have been trying to oust not only the expected baby but also her existing son from the succession, by such a challenge.

‘Unduteous’ Will

These rules did no more, essentially, than protect the liberi, if their father had failed to provide for them. Of much wider scope was the procedure, existing in some form at least by the middle of the first century B.C., known in classical law as the querela inofficiosi testamenti (‘complaint of unduteous will’).³⁹ This was available against the wills both of men and of women, and was limited to certain heirs on intestacy. The persons with a claim were the close relatives of the deceased—children, parents, brothers and sisters. Ulpian commented that it was a waste of time for cognates remoter than brothers to claim, and Diocletian gave a similar response.⁴⁰

The grounds for breaking the will were not technical, such as failure expressly to disinherit sui heredes, but moral. A claim could be made on the grounds that the will was unjust, by sui who had received less than the Talcidian fourth’. If the petition was accepted (details of the procedure are controversial and need not be discussed here), by a legal fiction it was deemed that the testator must have been insane (color insaniae).

Complaints by sui heredes of either sex could be made against the wills of fathers. The elder Seneca quotes⁴¹ from the maiden speech before the centumviral court of the prolix orator of Tiberius’ reign, Votienus Montanus. The case concerned Galla Numisia, who had been made heir to only one-twelfth of her father’s estate. She was being accused of poisoning him. Votienus had at least four attempts at making the same point: