4. TESTAMENTARY SUCCESSION

The complexity of the rules we have been considering must not blind us to the fact that intestacy was the exception, not the rule.

The Roman will is, however, even more remarkable for its precocious development. Elsewhere this development is late, but in Rome as early as the Twelve Tables there existed a will of some sort. This much is certain, but little else. In particular one must beware of assuming that the early will necessarily had all the characteristics of its mature successor. Of these the essential are four—(i) it appoints a universal successor, who may be someone other than the heres ab intestato\ (ii) it may make gifts

² See below, pp. 261 ff.

(legacies) out of the estate, and may make other particular dis­positions (appointment of tutors, manumission of slaves); (iii) it is ‘ambulatory’, or ‘speaks from death’ (e.g. the property dis­posed of is taken to include things which the testator acquired only after making the will, and a description such as ‘my nephews’ is taken to include those born since the will was made and to exclude those who have died since then (or rather their heirs)); (iv) it is revocable.

The early will.

(a) Testamentum comitiis calatis and testamentum in procinctu. Both these were obsolete some time before the end of the Republic. The testamentum in procinctu (‘in battle array’) is a soldier’s will, made before battle. The other is the more important. The comitia curiata met twice a year for religious business (probably in March and May and probably under the presidency of the pontifex maximus). This business included the making of adrogations (which may well have served, before any will existed, to provide a heres where no suus existed) and of wills, and when it was discharging the latter function it was styled comitia calata (‘called together’). The parallel oiadrogatio natur­ally suggests that the will was a legislative act, but there are signs that the function of the comitia may have been, or have become, limited to that of witnessing the act of the testator. Even if the former supposition is the right one, the important features of the will, granted the inability of a Roman assembly either to initiate or amend,¹ must have been its publicity and the control of the pontifex maximus, as guardian of the sacra. It was presumably ambulatory and revocable, and could apparently appoint tutors and manumit slaves, since the Twelve Tables refer to such acts as occurring by will. The parallel of adrogatio again suggests strongly that it could appoint a heres, though it has been argued that such interference with the customary rules of succession was as yet unthinkable and that the will was confined to par­ticular dispositions. This might explain the peculiar form taken by the third form of will.

(£) The mancipatory will {testamentum per aes et libram). This was apparently a product of pontifical interpretatio after the Twelve Tables, since Gaius speaks of the first two forms as having existed ‘from the beginning’ whereas this was added later, and for Gaius ‘the beginning’ would presumably be the Twelve Tables.

The testator first writes his will on wax tablets and then, having collected the usual five witnesses and a libripens, manci­pates his property (familia) to a person called familiae emptor. It is, however, a mancipation with a difference. The familiae emptor, instead of the usual assertion of title, declares that he has the custody of the property subject to the testator’s directions. Then follows a stage not found in the ordinary mancipation at all. The testator does not merely acquiesce but takes an active part by formally confirming the contents of his will and calling on the witnesses to bear witness that he has done so.

In Gaius’ time this mancipation is a pure formality, the only purpose of which is to give validity to the contents of the wax tablets. The familiae emptor is a man of straw. The will takes effect only on death, and then the person who succeeds to the rights and duties is not the familiae emptor but the heres named in the will. And the will is revocable, though only, it is true, by the making of another one.

In its developed form, therefore, the mancipatory will was preferable to that made comitiis calatis, certainly because it was secret and because it could be made at any time and at any place where seven Roman citizens could be found, and possibly (if the comitial will lacked this characteristic) because it could freely appoint a heres. But clearly it was originally not a will but a device for getting round the difficulty or impossibility of making one. The history of other systems shows that such a conveyance inter vivos is commonly the precursor of a will.

Bonorum possessio secundum tabulas; the ‘Praetorian will'. What­ever the original legal effects of the mancipatory will and whatever the course of events by which the familiae emptor was transformed into a man of straw, it is certain that the transforma-

tion was complete well before the end of the Republic. For we know from Cicero that already in his time the Praetor saw in the mancipation no more than an empty form: he promised bonorum possessio (secundum tabulas testamenti) to any person named heres in a written document sealed by seven witnesses, whether there had been a mancipation or not. His purpose in doing this was not to set up a new form of will but merely to recognize what had become the substance of the old. The five witnesses of the mancipation are now witnesses of the written will, and the familiae emptor and the libripens are added to their number. The will is written on one side only of two wax tablets which are then tied together, face inwards, and sealed by the testator and the seven witnesses. However this bonorum possessio was sine re until Antoninus Pius barred the claim of those entitled at civil law on intestacy.¹ Until then they had succeeded if they could show that the formalities of the mancipation had not been properly carried out. The practical effect of the Praetor’s inter­vention had thus until then been merely to reverse the burden of proof, though the discharge of the burden might not be easy, particularly if the will recorded, as all surviving examples in fact do, that a mancipation had taken place.

This is now commonly referred to as the Praetorian will, and from the time of Antoninus Pius it was for most practical purposes as good as a will. In the late law, however, though it makes an appearance in the pages of the Corpus luris, the Praetorian will had in fact been superseded by other forms.

Later forms of will.

end ensures that the document cannot be presented incomplete. In the Praetorian will the seals of the witnesses achieved all three purposes, the seal being both a ‘signature’ and a guarantee against the will being broken open. But if one asks why as many as seven witnesses were necessary, and why the testator himself did not have to authenticate the document with his seal, the answers can only be found in the original mancipatory will. The large ‘cast’ of a normal mancipation was probably needed not so much to prove that the act had taken place as to give it publicity, so that any defect of title (which in a small society would be likely to be known to the witnesses) could be investigated im­mediately. And the reason why the testator in the Praetorian will was not required to seal was presumably that in the manci­patory will he authenticated his act by the words he spoke. Neither answer makes sense, however, once the mancipation is no longer required. Moreover when papyrus or parchment began to replace wax tablets the seals no longer served so well to guarantee the completeness of the document. For this and other reasons there appears in documents of all kinds in addition to the old seal a newer mark of authenticity, the subscriptio. This is the nearest approach the ancient world made to, and is indeed the ancestor of, the modern signature. It is a sentence of varying length written ‘under’ (i.e. at the end of) a document, by the ‘subscriber’ in his own hand and consisting essentially of ‘I, Lucius Titius... have subscribed.’

The common will of the late Empire, introduced in a.d. 439, requires this newer form of authentication from both testator and witnesses, but still without jettisoning the inheritances from the past. It is in substance the Praetorian will with the addition of subscriptio by testator and witnesses. Justinian puts this in a more elaborate way when he says that the law from which the requirements of the new will were derived is ‘tripartite’—the will must be made in one operation and before witnesses (this comes from the civil law); the number of witnesses must be seven (Praetorian law, since by the civil law only five were witnesses); and to these requirements imperial legislation has added the subscriptio. From this analysis of its sources the will itself came to be called, not very happily, testamentum tripertitum.

The testamentum tripertitum has a direct descendant in the modern French ‘mystic’ (or secret) will, the subscriptio being replaced by its counterpart, the signature, and one of the wit­nesses by a notary. It is little used, however, and two other wills of the late Empire were of much greater subsequent importance —the public will and the holograph will. The public will (itself derived from the Greek law) was one which had been entered in the records of a court or deposited in the imperial archives. Its modern successor (the ‘authentic’ will) is made before a notary or a judge. The holograph will was simply written by the testator in his own hand. Justinian allowed it only in favour of children of the testator, but (with the additional requirement of the signature) it is in unrestricted use in Scotland and in Germany today, and is also recognized in France. It satisfies the layman’s desire to be free of what he sees as the lawyer’s artificial re­straints, but, quite apart from the dangers of any ‘home-made’ will, it is criticized as being open to forgery or to secret destruc­tion by the disappointed or the ill-disposed.

Internal requirements of a will. We have seen that the prime purpose of the will was to institute a heres, and that if it failed to do so, or the person instituted refused to enter, the whole will was void. As Gaius puts it, the institution of an heir is the source and foundation of the whole will. This being so, the institution had, to the formalistic mind of the early Roman lawyer, to be placed first in the wording of the will, and this reasoning was adhered to throughout the classical period. Out offavor testamenti, how­ever, failure to place the institution first was not allowed to invalidate the whole will but merely those provisions (e.g. legacies) which preceded it. There were of course other more substantial requirements also. We have seen that the institu­tion must be to the whole hereditas, though again favor testamenti led to a merely partial institution being construed as total. The institution must also be of an identifiable person (or persons). This made impossible, in particular, the institution of what we should now call corporate bodies, but which the Roman saw only as fluctuating groups of individuals (incertae personae). In this the lawyers were perhaps motivated by the fear of pro­perty accumulating in the hands of immortal bodies. That only Roman citizens (or peregrines with commercium) could be insti­tuted followed from the ‘personal’ character of this whole branch of the law.

Moreover the whole will must be in Latin, and the institution of the heir, like nearly all the other important provisions which a will might contain (appointment of tutors, manumission of slaves, disinherisons, legacies) must be expressed in formal words. The institution must contain the word heres and must be im­perative in form (‘Lucius Titius, be you my heir’ was the usual phrase, but ‘I order that Lucius Titius be my heir’ would do; not, however, *1 make Lucius Titius my heir’ or even ‘I institute Lucius Titius my heir’). Such distinctions, like the requirement that the institution came first, make pedantic reading and are obviously in detail difficult to support, but they have an under­lyingjustification. Just as the external formalities of a will serve to ensure its authenticity and completeness, so also these internal formalities are intended to minimize the possibilities of am­biguity. The lawyer is accustomed to precision of language, and he may argue that when the time comes to interpret the will the testator will not be there to explain what he means, and that if he is not prepared to consult a lawyer when he is making his will he should not expect a lawyer to face the unenviable task of deciding between the conflicting claims arising out of an ambiguous provision. Even when such technical terms are used there will be scope enough for ambiguity—in the description, for example, of what is included in a legacy or of the persons for whom the legacy is intended. The later Empire, however, was out of sympathy with such verbal formalities, and in a.d. 339 Con­stantine abolished the need for any particular form of words. The use of Greek was not universally allowed until a century later.

Substitutions, (i) Vulgar. The only essential of a will was the institution of an heir, but it would be an uncommon will which contained nothing else. If the instituted heir were an extraneus, the next provision would usually be a substitution, i.e. a direc­tion (in formal words) that if the instituted heir did not enter (and the wise testator would add a time-limit) some other person should be heir. There might follow other substitutions, ending usually with the substitution of a heres necessarius. The opening sentences of a will might therefore run: ‘Lucius Titius, be you my heir and do you make cretio¹ within 100 days of your knowing

¹ This was a formal (and therefore unambiguous) oral act of entry on the inheritance. Cf. above, p. 239.

New Roman">and being able to do so. If you do not thus make cretio_i be you disinherited. Then be you, Publius Maevius, my heir and do you make cretio, &c.... Then be you, my slave Stichus, free and my heir.’ These substitutions are called ‘vulgar’ (common, usual) and are in fact merely conditional institutions.

(ii) Pupillary. A child of any age could be heir, but not until the age of puberty could he himself make a will. This might present an unwelcome prospect to a testator. Suppose, for example, that he instituted his only son and that first he and then the son died before the son reached the age of puberty— the inheritance would then (since an impubes could neither have a suus heres of his own nor make a will) inevitably go to the nearest agnate. Apart from the general dislike of intestacy, the father might be reluctant to let the property go to the nearest agnate, and the lawyers therefore very early invented the device of the pupillary substitution to enable the testator to appoint an heir in case the son should die impubes. This, like the vulgar substitution, was in form a conditional institution (‘Be you, my son Titius, my heir. If my son shall be my heir and die before reaching puberty, be you, Seius, heir’), but it obviously results in a shifting of the inheritance, in violation of the principle ‘semel heres, semper heres’. First the son is heir and after him the pupillary substitute. It is probable that this was originally the limit of the breach with principle—that the substitute was looked on as the heir of the father and therefore took only that part of the son’s property which he had inherited from his father—but in the classical law the breach was far wider. The substitute succeeded to the whole property of the son wherever it had come from, and therefore a will containing a pupillary substitution was in effect two wills—one for the father and one for the son.

Since the son might die before the father, a properly drafted will would include both a vulgar and a pupillary substitution, usually of the same person (‘Be you, my son Titius, my heir. If my son Titius shall not be my heir or shall be my heir and die before reaching puberty, be you, Seius, heir’). If such a provision were included in the ordinary way in the father’s will it would obviously provide a temptation to the substitute to ensure that the child never did reach puberty. The proper course, Gaius tells us, was to put the substitution in a separate sealed document, not to be opened unless the child died im­pubes.

Restrictions on the power of testation. The underlying principle of the Roman law, as of the modern English law, is that the testator may do as he likes with his own, but this freedom was restricted in two ways.

(a) Formal—exheredatio. As far back as our knowledge goes the rule existed that a suus must be either instituted or expressly disinherited. If this were not done, the will was void. This was a purely formal requirement. It did not prevent the testator from instituting whomever he chose (provided, of course, he was by the general law capable of being heres}; it merely required him, if there was a suus, to make his intention doubly clear, positively by instituting the heres of his choice and negatively by disinheriting the suus. The original reason for the rule lay probably in the ‘vested’ character of the interest of the suus'. since he was in a sense already owner, he could only be deprived of his title by an express provision. In the classical law, however, the rule was an irrational survival which served mainly to provide pitfalls for the testator.

If we use the word ‘disinherit’ we must not accord to it its English overtones. The father was not necessarily motivated by disapproval or dislike, and the disinherited suus was not neces­sarily ‘cut out’ of his father’s will—he might well be a substantial legatee. It would often be wise in this way to prevent the par­celling up of the family estate—to give the whole to, say, the eldest son, subject to provision for his brothers and sisters. More­over even if the father instituted all the sui he need not institute them to equal shares.

This part of the law is unedifying not only because it is an irrational survival, but also because its details are excessively complicated. The Praetor intervened by the grant of bonorum possessio contra tabulas (against the will) but the scope of his reform was very limited, his object being only to bring this branch of the law into line with his rules of intestate succession by requiring testators to institute or disinherit not only sui but all liberi, and to modify in detail the consequences of failure to do so. His rules were somewhat simpler than those of the civil law, but they still provided many pitfalls. One asks why the Praetor did not ‘correct’ the law more drastically. The answer lies mainly, no doubt, in the conservatism of the lawyers, but it may also have been thought that these rules did require a man, in however clumsy and inadequate a way, at least to consider his obligations to his children. This purpose could however be very much more effectively served by the newer device of the querela inofficiosi testamenti which we must now consider; but nevertheless the rules of exheredatio survived into the later years of Justinian’s reign when they were finally confused with those of the querela.¹

(b) Substantial—querela inofficiosi testamenti. The ‘complaint of an unduteous will’ is peculiar among the institutions of Roman law in being, in its early development at any rate, the product of court practice. It apparently originated, perhaps in the late Republic, in the centumviral court, a tribunal of perhaps some thirty or forty lay judges which provided an alternative forum to that of the single iudex for the hearing of, in particular, cases concerning inheritances. The relative stability of its composition and the narrower field of its jurisdiction probably made the development of a consistent practice easier for this court than for the multiplicity of single iudices whose cases concerned all branches of the law. The claim seems originally to have been made, as was a similar proceeding in Greek law, under the cloak of an assertion that the testator must have been insane, but the ground of complaint came to be simply that the will, by disinheriting the complainant or making inadequate provision for him, offended against the qfficium pietatis—the moral duty which a man owed to his family. It was thus probably in origin an appeal to standards more akin to those upheld by the Censor than to those enforced by courts of law. In one respect it long retained this discretionary character: the court had to decide whether the testator had a just ground for excluding the com­plainant, and no attempt was made to define a just ground until 542, when Justinian, with the lawgiver’s dislike for unregulated discretion, promulgated a long list. In other respects, however, it gradually acquired the precise outlines of a legal institution. These must now be briefly considered.

The classes of person who could bring the querela were even­tually settled to be (in this order): descendants (including ¹ See below, p. 263.

descendants of a woman); ascendants; brothers and sisters (though only if a ‘base person’ had been instituted). These persons could only claim, however, if in the particular case they were entitled on intestacy by either the civil or the Praetorian law, and then only if they had no other remedy. For example, the suus heres who had been passed over could proceed under the rules of exheredatio or could obtain bonorum possessio contra tabulas, and was therefore denied the querela.

The complainant must show that the testator had without just ground made no adequate provision for him. What con­stituted adequate provision came, under the influence of the lex Falcidia,¹ to be fixed at one quarter of what the complainant would have been entitled to on intestacy. This was later called his legitima pars or legitima portio—the modern ‘legitim’. The provision need not have been made by instituting the com­plainant heir. If he had received his legitima portio by a gift out of the estate in any form his querela would fail.

The results of a successful querela varied according to the cir­cumstances. As has already been said, the ostensible ground may in the beginning have been the insanity of the testator. If this was so, the result should have been to invalidate the will and produce an intestacy. And in the simplest case—where there was only one possible claimant and only one heir instituted —this was the result. The claimant took the whole inheritance (not just his legitima portio) and all subsidiary provisions of the will, such as legacies, were void. On the other hand, if there were several instituted heirs, the querela might be brought against only one of them, and then the will failed only to the extent of that one’s share, legacies &c. being only proportionately invali­dated. Thelogicof intestacy was thus abandoned, and the querela was treated as a suit against a particular heir. The principle that a man could not die partly testate was in effect abrogated.

The law was thus arbitrary and unsatisfactory. It appeared to say that certain persons had a right to a legitima portio, and yet the result of the testator’s failing in his duty would usually be to give them a great deal more. Moreover it would often be difficult for the testator to make sure of complying with the law—the value of the inheritance might increase rapidly just before his death, or some members of a class might die and ¹ See below, p. 266.

thereby increase the legitimae portiones of the survivors. And the consequence of a small miscalculation of this kind might be to destroy the entire will. Finally, since the querela was only a last resort, there still existed all the complications of the law of exheredatio.

These defects were largely removed by interventions of Justinian. His earliest and most significant reform was to restrict the querela to cases where the complainant had received nothing at all under the will. If he had received something but it fell short of the legitima portio, he could only bring an actio ad sup­plendam legitimam to have the deficiency made good. The will was otherwise unaffected. In this way the testator was protected from the possibility that a miscalculation of his might enable the complainant to destroy the whole will and obtain far more than his legitima portio. Justinian later provided that any descendant or ascendant who was entitled to a legitima portio must be instituted heir. If he was instituted, but received (whether as heir or in any other way under the will) less than the legitima portio, he could only bring the actio ad supplendam legitimam. If he was not instituted, then the querela lay, but its effect was now only to replace the heirs instituted in the will by the successful claimant or claimants. In effect there was a will without an instituted heres. From this last reform it would seem likely that Justinian intended at last to combine the principles of the legitima portio with those of exheredatio and bonorum possessio contra tabulas, but it is not clear that this was in fact the result. Justinian also altered the legitima portio of children. If there were four or less, they were entitled to a third of their intestate share; if there were more than four, the fraction was a quarter.

Most modern Civil law systems have rules, derived sometimes from the querela inofficiosi testamenti, sometimes from the customary law, which are intended to strike a balance between the freedom of the testator and the claims of his family. These take the form either (e.g. in France or Scotland) of denying altogether to the testator the power to dispose of a certain proportion of his estate, which is therefore ‘reserved’ to his close relatives, or (e.g. in Germany) of placing no restriction on his power of disposal but allowing the relatives, if they wish, to claim a share. In either case the share is fixed and corresponds to the legitima portio of Roman law. By contrast, the English testator could in modern times give his fortune to his mistress and leave his wife and children destitute until, in 1938, Parliament allowed the wife (or husband) and children (but excluding adult sons and married daughters unless incapable of maintaining themselves) to apply to the court, not for a fixed share, but for ‘reasonable provision’.

This contrast between the Common law and the Civil law not only helped to make intestacy far more common on the Con­tinent than in England but also led, particularly in France, to a progressive parcelling up of family property and to the absence of that English figure, ‘the younger son’.