§ 112. immoral conditions annexed to the institution of an heir are deemed unwritten (cp. supra, p. 226). § 113. ULPIAN. tit. 24 § 15: Ante heredis institutionem legari non potest, quoniam vis et potestas testamenti ab heredis institu- Eod. tit. 21: Heres institui recte potest his verbis: TITIUS Eod. tit. 22 § 4: Incerta persona heres institui non potest, velut § 9 I. de hered. inst. (2, 14): Heres et pure et sub condicione institui potest: ex certo tempore aut ad certum tempus non potest. 10: Impossibilis condicio in institutionibus et legatis, nec non in fideicommissis et libertatibus pro non scripto habetur. § 113. Succession by Necessity. This The original idea of a will is that it enables a person who has no son, to provide a son for himself (supra, p. 568) who shall 'perform all the duties of the testator, human and divine", that is, pay his debts and offer the funeral sacrifices. It follows, as a matter of course, that, where there are sons, there is no room for a will. is the condition of the law as we find it in ancient Attica, where the rule was that a man who had lawful sons could not make a will. At a later stage, however, a man who had sons was permitted to make a will in which he appointed his sons heirs, or to make a will for the event of his sons either not surviving him, or surviving him but 1 Compare the law of Gortyna in Crete, about 400 B.C., X. 42, 43. F. Bücheler and E. Zitelmann, Das Recht von Gortyn (1885), p. 134 dying before they attained majority'. There are strong grounds for § 113. believing that early Roman law passed through a very similar course of development. If a man had a filius suus, he could not, at the outset, make a will at all; if he had a daughter or a grandchild in his immediate paternal power, he could, it is true, make a will, but the rule was that he could only appoint a stranger testamentary heir together with the daughter or grandchild-a rule which probably contains the germ of the subsequent jus accrescendi of such sui (v. infra, p. 581). As against the sui, the power to make a will was excluded by the mere fact that, on the death of the father or grandfather, the suus heres was deemed ipso jure in possession of the inheritance (p. 546), so that, even if testamentary heirs were appointed, there was no possibility of their carrying out the 'cretio' and thereby obtaining possession of the estate. Nevertheless, as in Athens so in Rome, the right to make a will notwithstanding the existence of sui came ultimately to be acknowledged. In Rome the development of the law resulted in the adoption of the principle of an unrestricted testamentary power, and the effects of the material claims originally enjoyed by the sui were only traceable, at a later period, in certain requirements concerning the form of wills. It is probable that, as in other branches of the law, so here, the ultimate result was reached, not by any abrupt change, but by a series of intermediate modifications of the earlier doctrine. But however this may be, the outcome was that, with the obliteration of the notion of the sui being ipso jure in possession of the inheritance, 2 Schulin, op. cit. p. 15. In support of this supposition v. Schirmer, ZS. d. Sav. St. vol. ii. p. 170 ff.; Salkowski, ibid. vol. iii. pp. 201, 202. • The existence of such intermediate stages seems to be indicated more particularly by the system of 'pupillary substitutions' which was developed in Athens as well as in Rome. A 'pupillary' substitution-which is opposed to a 'simple' substitution (substitutio vulgaris, cp. supra, p. 577)-means, in the classical law, a will made by the testator for an impubes in his power and intended to take effect in case the latter intra pubertatem decesserit.' Originally, how- § 113. the ancient ideas of family ownership lost all practical influence and yielded to the stronger claims of patria potestas with its paramount power of disposition over the property and the members of the household. The evidence afforded by the rules of succession by necessity, as they actually appear in the historical records of Roman law, justifies us in believing that the leading ideas at work in the earliest phase of that branch of the law were such as we have just described. So far as the Roman law of succession by necessity can be historically authenticated, it presents a twofold aspect, either a formal or a material aspect. In its formal aspect it is concerned with the form of wills; in its material aspect it is concerned with the substance of wills. The formal law of succession by necessity requires that the heir by necessity shall be either instituted or disinherited in the will; in other words, that the testator shall make express mention of his heir by necessity. The material law of succession by necessity requires that every testator shall give his heir by necessity a certain portion-a 'statutory' portion-of his property; in other words, that he shall confer a material benefit on his heir by necessity. The earliest phase of development being closed, Roman law started, in historic times, with a purely formal law of succession by necessity, and then proceeded gradually to work out a material law of succession by necessity. Finally, under Justinian, a uniform system of succession by necessity was established by means of a fusion of the formal and the material law. I. According to the civil law, the sui heredes were the only persons who had a right of succession by necessity. Their right was however a purely formal one. That is to say, it was one of the formal requirements of a valid will that the sui heredes should either be instituted or disinherited. If the testator wished to dispose of his estate in favour of other persons, he had first to expropriate those who were co-owners of his property (exheredes facere), because otherwise his property was encumbered and incapable of free disposition in favour of others. In this requirement of exheredatio we have a last trace or perhaps, in a certain sense, an acknowledgment of the rights of the family-the family of descendants, namely-as owners of the estate. In order to ex- § 113. tinguish the claims of the family and to convert the property into the free and unrestricted property of the holder, an exheredatio was indispensable. And the law required that a filius suus should be disinherited 'nominatim,' i. e. by special mention; in the case of daughters and grandchildren, an 'exheredatio inter ceteros' was sufficient, i. e. it was enough if the will contained the clause 'ceteri exheredes sunto,' a clause which every prudent Roman was in the habit of appending to the institution of the heir. If the formal requirements of the law as to the rights of heirs by necessity were not satisfied, in other words, if a suus was passed over-being neither instituted nor disinherited ('praeteritio') -the result was different, according as the suus in question was a filius suus or another suus. Where the suus passed over was a filius suus, the formal defect was fatal to the will and intestate succession took place. Where other sui-daughters or grandchildren-were passed over, the will remained valid, but the sui praeteriti were admitted to the succession together with the heirs appointed in the will (scriptis heredibus adcrescunt), the rule being that if the latter were extranei, the sui praeteriti should take together one moiety of the estate, but that if they were sui, each suus should take a portio virilis,' or equal share. Nor, again, was it sufficient, if the suus were conditionally disinherited or instituted (unless indeed the condition was one which merely depended for its fulfilment on the free option of the person conditionally instituted, a so-called 'condicio potestativa'): the suus had also to be instituted or disinherited for the event of the condition failing. If the testator had provided for the appointment of several 'degrees' of heirs by means of an institutio and substitutio (supra, p. 577), a suus who was not instituted had to be expressly disinherited 'ab omnibus gradibus,' i. e. the testator had to state his intention to disinherit him as against both the institutus and the substitutus for which purpose, amongst other modes, an exhere • Hölder says (ZS. d. Sav. St. vol. iii. p. 219): 'Like all acts of expropriation a testamentary exheredatio, being an act which extinguishes a private right in due legal form, implicitly re- § 118. datio at the beginning of the will (ante heredis institutionem) was sufficient. At the outset a difficulty arose in connexion with the postumi sui (supra, p. 531), who, being personae incertae (p. 577), could neither be instituted nor disinherited. With regard, however, to such postumi as were born after the death of the testator ('postumi legitimi'), traditional usage invested them with testamenti factio passiva, and a lex Junia Velleja conferred the same capacity on postumi born in the life-time of the testator, but after the execution of the will ('postumi Vellejani'). And prior to the last-named enactment, the praetor Gallus Aquilius had devised a formula by which it became possible effectually to institute a grandson by a son, who, though born after the execution of the will, became the suus heres of his grandfather in consequence of the death of his (the grandson's) father during the life-time of the testator. Postumi of this class are accordingly known as 'postumi Aquiliani.' It was thus that postumi sui gradually became capable of being effectually instituted and disinherited. If a testator desired to disinherit a postumus filius, an exheredatio 'by name' ('nominatim ')-in other words, an express exheredatio-was necessary. If he desired to disinherit other postumi (daughters or grandchildren), an exheredatio 'inter ceteros'—in other words, a tacit exheredatio not expressly mentioning the postumi in question—was sufficient. But as regards postumi of the latter kind, the insertion in the will of the general clause 'ceteri exheredes sunto' only operated as a valid disinherison, if the will contained legacies for the disinherited postumi, so that it was clear that in using the words 'ceteri exheredes sunto' the testator had these particular postumi in his mind. Hence it was thought more prudent to disinherit these other postumi in the same way as a postumus filius, viz. by express mention ('nominatim '). The praeteritio of a postumus suus-whether a son, a daughter, or a grandchild-resulted in all cases in what was called a 'ruptio testamenti,' i. e. the entire will became void and intestate succession took place. II. According to the praetorian law not only the sui, but all the liberi (supra, p. 556), had a formal right of succession by necessity, |