The Concept of Potestas
(1) The Nature of Potestas
The core concept of the Roman law of parent and child is the potestas (‘power’)[494] held by the senior male in the family over his descendants.
More specifically, this was power exercised over all descendants in the male line,[495] the term for which is ‘agnatic descendants’, as agnates are those related through the male line. In addition, potestas only arose in relation to the offspring of a valid Roman marriage.[496] Subject to this, a person was in the potestas of his or her oldest living male, agnatic, direct ascendant. The person exercising this potestas was known as the paterfamilias (‘father of the family’). A person subject to this potestas is generiÂcally known as a filiusfamilias (‘son of the family’), though the feminine form for a daughter would be filiafamilias. We saw in Chapter 5 the distinction between those who were legally dependent (alieni iuris) and those who were legally indeÂpendent (sui iuris).[497] The paterfamilias was sui iuris and those subject to his potestas were alieni iuris. When a person had no agnatic male ascendants, for example because of the death of the paterfamilias, that person was sui iuris and, if male, he was considered to be a paterfamilias.[498]It is important to understand that only a man could hold potestas over a descenÂdant. A woman could not be a paterfamilias,[499] though she could certainly be sui iuris just as much as a man could.[500] This does not, of course, mean that the mother was not an important authority figure in a child's upbringing. They clearly were.[501] Only the paterfamilias, though, exercised this special legal authority.
An example may illustrate how this worked. Suppose a man, whose parents and grandparents are all dead, and who was therefore a paterfamilias, had a son and a daughter.[502] The son and daughter each had a son.
The paterfamilias would have potestas over both his son and his daughter.[503] He would also have potestas over the son's child, but not over the daughter's child, because the latter is not an agnatic descendant. Instead, the daughter's child would be in thepotestas of his or her father (or of the father's own paterfamilias'). When the paterfamilias died, both of his children would become sui iuris. The potestas over the daughter, for example, did not in any sense pass down to anyone else, such as her brother. The son would now be his own son's paterfamilias.We shall see in this chapter what exactly the scope of this potestas was. One important limitation should be mentioned now, though, which is that it was only relevant for private law matters. The fact of being a filiusfamilias did not limit a person's capacity to act in public law matters, and such a person could for example hold a magistracy.[504]
It is unlikely that there has ever been a society in which the law did not give parents some degree of legal authority over their children. To us, the extent of the preference for fathers seems unusual, but such paternal preference was hardly uncommon in earlier societies. Where the Roman approach was unusual was in two of its features. The first was its scope. AsJustinian said: ‘there are no other people who have such rights over their children'.[505] Second, it was unusual in that it was in principle lifelong.
(2) Rights over the Person
The paterfamilias had extensive rights over the persons of those in his potestas. Indeed, in early times, these rights were so extensive that the paterfamilias could almost be charÂacterised as the owner of those in his potestas. Thus, for example, until the end of the classical period he could sell his children into a status analogous to that of a slave.[506] Such a child was said to be in mancipio (in bondage). The idea was revived in the post- classical period by Constantine: parents under pressure of poverty were permitted to sell their children.[507] Something very similar was permitted in classical law in the case where a son had committed a wrong: his paterfamilias could choose to surrender him to his victim in bondage instead of paying damages.
This process was called noxal surrender, and will be considered in Chapter 9.[508] Other than this, the paterfamilias had extensive rights, considered here. He had the power of life and death over those in his potestas; he could inflict punishments; he was in principle entitled to physical custody of his children, or to regulate their residence; he was entitled to support from them (as were they from him); and he had rights over their marriages.(a) Power of Life and Death
From the earliest period, part of the authority of the paterfamilias was the power of life and death over those in his potestas. The general view seems to be that this power was originally unfettered.[509] However, in the classical period this does not appear to have been the case, and a paterfamilias who abused this power could expect punishment if this came to the attention of the authorities:
It is said that, when in the course of a hunt a certain man killed his son, who had been committing adultery with his stepmother, Emperor Hadrian[510] deported him to an island, because he acted more like a robber in killing him than this being based on a father’s right. For paternal authority should consist in duty, not cruelty.[511]
Later, but still within the classical period, Ulpian tells us that a paterfamilias cannot kill someone in his power without first obtaining the approval of a magistrate.[512] Finally, in AD 318, Constantine provided that the killing of one’s own child, or of any other close relative, was to be criminally punished as parricide.[513]
A special case of the power of life and death applied to adulterous daughters. In terms of the lex Iulia de adulteriis coercendis, legislation passed in 17 BC under Augustus,[514] a paterfamilias could kill any married woman in his potestas who comÂmitted adultery.[515] The adulterous pair had to be caught in the act of intercourse, and had to be killed without delay.[516]
The final case of the power of life and death to be considered is the right of the paterfamilias to expose newborn children.
While this is often considered sepaÂrately, it seems appropriate to consider it here, as the child's death would be the result of this unless the child was found and rescued. Such a child would often be raised and treated as a slave, though he or she technically remained free. This right of infant exposure persisted through the classical period and beyond, but was declared unlawful in AD 374.[517](b) Punishment
If the paterfamilias was entitled to kill those in his potestas, so much the more did he have the right to inflict punishments on them. This too, however, could attract adverse consequences if abused. Trajan,[518] for example, compelled a father who was maltreating his son to emancipate him.[519] After the son's death, the father was denied the normal rights of succession of an emancipating father.[520] In later law, the right of punishment was restricted to reasonable chastisement.[521]
(c) Custody
In principle, the paterfamilias was entitled to regulate the residence of those in his potesÂtas. A child in potestas was considered capable of being stolen, and could be recovered as stolen property.[522] In addition, the praetor would provide interdicts to compel the production of a filiusfamilias or filiafamilias who was in another's keeping,[523] or to proÂhibit anyone from preventing the paterfamilias recovering him or her.[524]
Actually, though, there was more control over the father's rights here than would at first appear. Where a paterfamilias sought to recover a prepubescent child, the judge could decide to defer a decision until the child reached puberty. The relative characters of the parties would be relevant here. If both parties were of bad character, it might even be considered appropriate to appoint a third party to take responsibility for the child's upbringing until puberty.[525] Similarly, the mother might be preferred to the father where the father was of poor moral character.[526]
(d) Support
Parents and children had mutual obligations of support.[527] This included chilÂdren born outside marriage[528] and those who had been released from potestas by emancipation.[529] An impoverished parent might, therefore, be able to obtain financial support from an emancipated son who had greater means.
The same would be true if the positions were reversed: the son could seek financial support from the parents.This obligation of support only existed to the extent it was necessary, however, so nobody was entitled to support who had sufficient means of his or her own.[530] Support could be refused on other grounds as well. For example, a father could refuse to support a son who had informed against him.[531]
(e) Rights over Children’s Marriages
The paterfamilias also had powers over the marriages of those within his potestas. Where the party to a marriage was alieni iuris·, the consent of his or her paterfaÂmilias was necessary to the validity of the marriage.[532] The paterfamilias could also compel those in his potestas to divorce.[533]
Legislation under Augustus had the effect of limiting the ability of the paterÂfamilias to abuse these powers. Those who wrongfully prevented those in their potestas marrying could be compelled by a magistrate to arrange a marriage.[534]
(3) Rights over Property
(a) The Legal Capacity of Those Subject to Patria Potestas
As far as legal capacity is concerned, the basic position of a filiusfamilias was not all that different from that of a slave. Though we shall see certain limited exceptions below, the position in principle was that nobody subject to another's potestas could own any property at all. Anything acquired by the filiusfamilias was acquired not for himself, but for the paterfamilias. Where the filiusfamilias acted on behalf of the paterfamilias, he could benefit him[535] but not bind him. Further, while the filiusfamilias (unlike a slave) could incur liabilities and be sued in his own name,[536] this was unlikely to be found particularly useful until the filiusfamilias became sui iuris and could have property of his own with which he could satisfy any judgment against him.
(b) Peculium
We saw in Chapter 6[537] that, although slaves could not own property, they could be allowed to manage a fund of money or property called a peculium.
This peculium remained the property of the master, but was for practical, day-to-day purposes treated as belonging to the slave. Much the same situation applied with free persons who were subject topatriapotestas. A filiusfamilias could also hold apeculium, which the paterfamilias still owned but which the filiusfamilias was allowed to treat as his own.Still, the inability of those subject topotestas to own property or to transact in their own right must have proved inconvenient. We shall see below that means were developed to hold the paterfamilias liable to a certain extent for the actions of the filiusfamilias. That was beneficial to third parties dealing with a filiusfamilias. From the point of view of a filiusfamilias who wanted to transact or to sue in his own name, however, it was still unsatisfactory.
An exception to the proprietary non-capacity of the filiusfamilias was introÂduced in the early Empire, by Augustus. This was the peculium castrense, and it was made up of property acquired by a filiusfamilias while on military service. This was owned by the filiusfamilias, who could transact freely with it and who could be held liable to its extent.45 Under Constantine, this was extended to any earnings in public service (the peculium quasi castrense).4
The peculium adventitium was another exception, also introduced by ConstanÂtine. This was made up of property acquired by afiliusfamilias on the death of the mother. Although the paterfamilias could use this property and retain any profits, he did not own it and could not diminish the capital.47 This category was progresÂsively extended, and under Justinian it included anything not acquired from the paterfamilias or forming part of apeculium castrense or quasi castrense.4
In the developed law, therefore, the position could be quite complex. There were now four types of peculium, with the original peculium owned by the paterfaÂmilias now confined to property and money derived from him, and called peculium profecticium to distinguish it from the other types. An individual might have any or all of these types of peculium.4
(c) The Liability of the Paterfamilias
The introduction of the peculia5' castrense and quasi castrense cannot, however, have solved the problem completely. In particular, it hardly seems fair on third parties dealing with a person for matters to depend so much on whether they were dealing with someone sui iuris or alieni iuris. After all, they could hardly have always known whether the person they were dealing with was subject to potestas.
Despite the inconvenience of the system, the Romans never got rid of it. Instead, in an example of the Romans' strange mixture of conservatism and inventiveness, they kept the system but devised ways of circumventing its conseÂquences. In simple terms, the paterfamilias could be held liable for the actings of those in his potestas up to the level of the peculium, or for anything he had authorised
45 D.14.6.2; D.49.17.10-11.
46 C.12.30.1(ad 326).
47 C.6.60.1 (ad 319). In other words, he had usufructuary rights, on which see Chapter 14.
48 C.6.61.8 (ad 531).
49 For an overview, including an account of the impact of the Roman law on South African law, see E Spiro, ‘The Law of Peculium in South Africa’ (1954) 17 THRHR 256.
50 This is the plural form of peculium.
the filiusfamilias to do. In addition, a special form of liability (called noxal liability) could be imposed on the paterfamilias for the wrongdoing of those in his potestas. We have seen in Chapter 6 that much the same approach was taken with slaves. The issues will be considered in Chapter 9, in more detail, for both categories of alieni iuris· persons.
B.