CHAPTER XXI. MANUMISSION DURING THE EMPIRE (cont.). MANUMISSION

BY WILL. DIES, CONDITIO, INSTITUTION.

A gift of liberty by will is not necessarily absolute and immediate: it may be subject to a condition or deferred to a future day. Pending the event the man is a statuliber: we have already considered his position^[MDCCLXXX] and have now to discuss the other questions affecting these modalities.

Where the liberty is deferred to a certain future time, it is said to be subject to dies certus. If the words ad annum are added, e.g. ad annum liber esto, they are construed as meaning “ at the end of a year².” If the words are ad annos decern, they are treated as super­vacua*. A gift of freedom intra annum post mortem entitles the donee to liberty at once. The rule is attributed to Labeo, and is declared to be justified by him as an inference from the rule that where the gift is: Let him be free si heredi intra decimum annum decern dederit, the man is free if he pays at once⁴. It is plain that this does not justify the rule. The one rule says merely that to impose a time within which the condition must be satisfied is not to impose dies in addition to the condition: it leaves the choice of time within a certain limit to the slave himself. The other does not: it does not say who is to have the choice of time, and the actual rule is a case of favor libertatis. We saw that ad annum meant at the end of the year. The text adds that this is to be reckoned from the death, but that if the words are such as to require the time to run from the date of the will, and the testator dies within the time the gift is not void⁴, but the time must be waited for. The same rule applies no doubt to all cases in which that con­struction is given. It is not too plain why anyone should have thought 480 Manumission by Will: Dies Incertus [pt. n it void. The context suggests that Julian is simply emphasising the fact that it differs from a case in which the testator does not die within the year.

Dies incertus is on a different footing. Dies incertus both an and quando is a condition and will be considered later’. As to dies certus an, incertus quando, of which cum T moreretur is the type usually cited, we are told by Papinian in a famous text⁴, that dies incertus conditionem in testamento fadt, and the proposition is confirmed in many texts’. It has, however, been the subject of much discussion in recent times. The existence of the rule itself has been doubted; the view being held that the dies referred to by Papinian is a dies incertus an et quando. But in view of the emptiness of the remark in that sense, the generality of his text and the content of the other texts cited, it is not necessary to do more than advert to this view“. But the admission of the rule leaves, still, a number of doubts, which the texts do not clear up. The main effect of the rule is to prevent transmissibility of the legacy to which the modality is attached, and that this is probably due to intuitus personae, i.e. a recognition of the testator’s presumed intention to benefit a particular person who may not be alive at the time of the event, appears from the language of at least one of the texts⁷, and from the fact that the rule was not applied to a gift cum legatarius morietur, since here the point could not arise: the testator’s intent was clear⁸. It is not easy to see why the rule was not applied to dies certus, in which the same uncertainty would arise.

That the rule of Papinian which is our immediate concern was in fact extended from institutions to other gifts appears from the fact that all the jurists who lay it down in general terms or apply it to gifts other than institutions seem to be rather late: Pomponius is, it appears, the earliest³. Others are Papinian⁴, Ulpian⁸ and Paul⁶. On the other hand, Julian, while he admits the rule in the case of insti­tutions⁷, denies its applicability to a gift of liberty⁸. He is discussing the coexistence of a simple legacy of the slave and a gift of liberty to the slave at the death of the legatee, and he is reported by Gaius as considering the legacy void, on a principle already discussed⁸, by reason of the existence of a gift of liberty subject only to dies.

This view, that the idea is carried over from the case of institution, though it is strongly suggested by the foregoing case, and is supported by the existence of analogous extensions, e.g. the treatment of unlawful and impossible conditions¹¹, has, however, to face some difficulties. It may be said that if this rule was carried over, the rule excluding dies certus ought to have come over too. The answer seems to be that this last rule is not one of interpretation: there is not the logical reason which exists in our case. The question whether certain words are to operate as a condition or not cannot depend on the kind of gift to which they are attached even though they may have been declared conditions for ulterior reasons which apply only to institutions.

It may also be objected that Labeo is found applying the rule to legacies¹². But Labeo is speaking of a gift si moritur, and though the difference between cum and si would not be conclusive to a classical

¹ See Dernburg, Pandekten, 3. § 82, n. 1, and literature there cited; Bufnoir, Conditions, 12.

482 Manumission by Will: Dies Incertus [pt. n jurist¹, it is by no means clear that it would not have been so to Labeo. It may be noticed that he speaks of it simply as a conditional gift, and not as one which has for this purpose the effect of a condition, as Papinian says for our case².

It may also be said that the Code contains an enactment of Dio­cletian, extraneum quum morietur heredem scribi placuit³, which seems to shew that the rule in this case differs from that in legacy, which thus cannot have been simply borrowed.

It has been pointed out that the chief text, one of Papinian, does not say that dies incertus quando, certus an, is a condition, but only that it fadt conditionem, i.e. has the effect of a condition’. The dis­tinction is exact as a matter of words, and Ulpian has no doubt the same point in mind when he says that such a modality appellatur conditio³. A condition is essentially incertus an. Elsewhere, Papinian, Ulpian and Julian ignore this distinction’, though it recurs as late as Justinian, who carefully distinguishes the cases of dies incertus and conditio, though he gives both of them the effect of postponing dies cedens¹³. It does not seem indeed that the distinction involves any difference of effect¹¹.

Condition is a somewhat complex matter. A condition is a future

^I See 45. 1. 45. 8. ² 85. 2. 75. 8 C. 6. 24. 9.

* Ante, p. 480. ⁶ So Accarias, Precis, § 324 (cp. § 384) following Machelard.

⁶ loc. cit. 1 35. 1. 75. Brunetti, op. cit. 22, 37.

8 30. 30. 4. 9 go. 104. 6; Ulp. 24. 31; D. 35. 1. 79. 1. « C. 6. 51.1. 7.

^II Brunetti, op. cit. 37, explains by it the view taken by Julian in a text already discussed ch. xxij Manumission by Will: Conditions 483 and uncertain event. It seems that every restriction which makes the event depend on an occurrence which may not happen is a condition even though it be such that if it occur it must occur at a certain time —incertus an, certus quando.

A gift of liberty is not conditional, and therefore delayed, unless it is clear that this was the intent. Some provisions which look at first like conditions may be only directory: it is a matter of construction, to be decided for each case. Thus where a man is to be free ita ut rationes reddat, this is not a condition: it is a direction. He must carry it out; indeed every slave who has administered has to render his accounts whether such a direction is given or not¹. Where slaves are to be free if they attend in alternate months to the sollennia of the testator’s tomb, this is a direction: it is to be carried out after liberty is attained, and they can be compelled officio iudids, to do the duty. But the liberty is not conditional². Where one is to be free “sic tamen ut" he stays with the heir so long as the latter is a iuvenis, and, if he does not, iure servitutis teneatur, this too is only a direction to be carried out after the man is free⁸. On the construction of the whole gift, even the strong words at the end are not allowed to limit what the jurist thinks to be meant as an immediate gift of liberty.

Another type of case is represented by such gifts as Sticha cum liberis libera esto. This is not a condition: she is free, apart from evidence of intention, though she have no children or they cannot from any cause be freed*. So where the gift was “ Let S and P be free if they are mine when I die,” either may take though the other has been alienated : there is no condition⁸.

A condition involves future uncertainty, and thus a gift which is expressed in conditional form, but the event is one which must neces­sarily be determined by the time the gift operates, is not really conditional. The donee can never be a statuliber under it. Thus, " to be free when I die’,” or “to be free if I do not veto it by codicil⁷” or the like: these are not conditional⁸. “ To be free if he pays what I owe (30. 68. 3) as to legacy of a slave with a gift of liberty on the death of the legatee. Bat we have seen (ante, p. 469) by comparing the text of Papinian (31. 65. 3) that the trae explanation is historical. And this latter text makes it clear that the treatment of dies incertus as*a condition produced other effects than the postponement of the vesting of a transmissible right. Such an effect would have no meaning in the case of gift of liberty discussed in the texts.

i 40. 4. 17.1; 40. 5. 37. » 40. 4. 44.

⁸ 40. 4. 52. See the Testamentnm Dasumii (Girard, Textes, 767) where testator directs a slave to be free cum contubemali sua ita ut earn in matrimonium habeat jidele.

4 40. 4. 13. 3. The jurist is helped to the conclusion by the construction put on the Edict: ventrem cum liberis in possessione esse iubebo, 37. 9. 1.pr.\ cp. 35. 1. 81; 40. 7. 31.

«28.7.2.1. «40.4.18.1. 7 h. t. 28.

8 «’To be free if he has managed my affairs welT’ is not conditional, he is free if the testator having long survived the making of the will has made no complaint, 40. 5. 18; “ if I have no son when I die” is no condition: the gift is good or bad according to the event, 40.4. 7; cum moriar being understood to include postumi, there might be a lapse of time. to T ” is no condition if there was no debt to T^[MDCCLXXXII]. If there was a debt and the testator has paid it, the condition is treated as having failed: it was in fact an ademption². It is clear that in all these cases where there is an express hypothesis on which the gift is made to depend, there may be need for enquiry on the result of which the fate of the gift will hang. In common speech they would be called conditional, and they are in fact sometimes so called in the texts. This is the case with the gift “if mine when I die’,” and qui sine offensa fuerunt liberi sunto*. Although, as the rest of this text shews, the words refer entirely to the past, the gift is called conditional, but the man is not a statu­liber even though there be some delay, before it is clear whether he is free or not⁶. The description of such gifts as conditional is not correct: the practical point is that when the matter is settled they are free and have been so from the time of aditio*. Thus, for instance, their interim acquisitions are their own.

Just as a gift, on the face of it conditional, may be pura, so, con­versely, a gift on the face of it pura may be in fact conditional. There may be tacit conditions. Thus in the case of divorce a woman who frees within the 60 days of the lea lulia makes the man a statuliber⁷. A slave freed in fraud of creditors is a statuliber till it is certain whether they will avoid the gift or not⁸.

Impossible conditions are struck out—a Sabinian extension of the rules as to institutions, accepted in later law’, and it may be assumed that the same was true as to illegal or immoral conditions, though texts seem silent¹⁰. Impossible conditions are those impossible in the nature of things^{[MDCCLXXXIII]}. These cases, where impossibility is patent on the gift, create no difficulty, but there are other types. Impossibility to the person concerned is no objection to the condition, and the texts put on the same level gifts on which a condition or dies is imposed such that, though it is not contrary to the nature of things, it is practically certain not to occur. Such for instance is a condition of paying a vast sum of money or living a hundred years¹². Such a gift is on a level with one cum morieturℱ, and is treated as illusory and void. It is clear that the line between these two is shifting, and probably the matter is not thoroughly thought out by the Romans¹⁴. Some conditions so treated

ŃĐœ. xxi] Impossible Conditions. Negative Conditions 485 seem out of place. Thus where a slave is to be free, “if my heir alienate him,” this is held illusory, since it cannot operate till the slave is alienus. There must, one would think, have been some other evidence of intention here; if not, the interpretation seems not very consistent with favor libertatis¹. The view is Paul’s, who comes to the same unfavourable conclusion, rather more rationally, on the words, liber esto, si heredis esse desierit². Another case is that of a condition, possible on the face of it, but already impossible at the death owing to circumstances. This case, and the analogous one of supervening impos­sibility, will be discussed later in connexion with the topic of satisfaction of the condition³.

Some conditions would doubtless avoid the gift: in general they are those which would avoid an institution⁴. There are, however, cases in which a condition is allowed in manumissions which would be differ­ently treated in legacies. Ulpian tells us that a manumission at the discretion of a third party is valid³, while we learn elsewhere that gifts by will in such a form are void⁶, though they can be effectively made in a disguised form, e.g. si Maevius Capitolium ascenderit⁷. But there is perhaps no distinction here, since Ulpian seems to have rejected this rather absurd differentiation, and to have considered all such gifts valid whether disguised or not⁸. As in case of legacy the gift might be at the discretion of the donee himself⁸

A case of more importance is that of negative conditions. Liberty on such a condition, if taken literally, is nugatory, for till the death of the man it cannot be said that he will not do the thing, and the cautio Muciana, by which this difficulty is avoided in ordinary legacies, has no application here. Liberty once effective is irrevocable: it is incon­ceivable that the man should become a slave again on failing to observe the condition. Moreover, as liberty is inestimable it is impossible to give security for it. Thus such gifts seem to have been made in a derisive way, and Pomponius, taking the case, si Capitolium non ascen­derit, says that Julian holds that if it appears that the testator meant the gift not to take effect till death it is a nullity, as a gift cum moreretur would be¹⁶. On the other hand if there was no such intent, the words

і 40. 4. 39. An analogous institutio was better treated. Where a man was to be heres if he freed a serous hereditarius, the mere act of manumission was held by Labeo to be a satisfaction of the condition though it was void: verum est eum manumisisse, 28. 7. 20. 1.

‱ 40. 4. 39. in Jin.

⁸ Post, p. 489. See De Ruggiero, Dies Impossibilis, 29 sqq.·, Babel, Aus Rom. und Burg. Recht, 193 sqq.

⁴ Accarias, loc. cit.

⁸ 40. 5. 46. 2. The text deals with a fideicommissum, but its reasoning applies to direct gifts.

⁸ 28. 5. 32. pr.; 35. 1. 52. 7 28. 5. 69; 35. 1. 52.

8 80. 43. 2; 31. 1. pr.; 40. 5. 46. 2. Bufnoir, op. cit. Ydbeqq. As to such discretion in fideicommissa, post, pp. 516 sq.

9 40. 5. 46. 1. This maybe true only of fideicommissa. Apart from condition he could not yarn ĐŸĐč

10 40. 4. 61. pr.; 40. 7. 4. 1.

486 Manumission by Will: Conditio Iurisiurandi [pt. ii were construed also by Julian favors libertatis, as if they were “if he does not do it at the first opportunity¹,” so that if at any time he was able to do the thing and abstained, he was free. It is plain that in one of these cases the gift is treated as puerile and empty, while in the other the condition is put on that level.

But there are traces of another way of looking at the matter. It is obvious that a patron may have reasons for wishing a thing not to be done by the freedman, and it is not unreasonable that he should have his way here, as in legacies. The way seems to have been found by allowing conditiones iurisiurandi. We are told that where liberty is given on the condition of taking an oath, this cannot be remitted by the Praetor, since to remit the condition is to bar the gift, as the liberty cannot be attained aliter quam si paritum fuerit condicioni*. Such conditions are invalid in institutions and legacies⁹, and our text adds that if the liberty is coupled with a legacy, and the same condition is applied to both, he does not get the legacy unless he swears⁴. Clearly not, for he is not free, and if it were struck out from the legacy alone, the legacy would be void⁹. If the legacy is under a condition of swear­ing, and the liberty simple, there is no difficulty: the condition is remitted’. The validity of such conditions in manumission applies to any act⁷, but its importance is most obvious in reference to negative conditions, and it seems possible that it is on their account that the condition is allowed. Where a condition of swearing is remitted and the thing sworn is not improper, the beneficiary is not entitled to the gift until he does the thing’, or, no doubt, in negative conditions, gives the cautio Muciana. Now, in manumission, if the condition of swearing not to do were remitted, it would be quite impossible to give the cautio: it is inapplicable. Accordingly, as the text says, to remit the oath is to bar the gift* The oath is no great security since at least in later law it is not binding¹⁰, though Venuleius¹¹ records an earlier doubt, but he speaks of those who have taken such an oath as religions adstrictiℱ.

We have already seen that a condition imposed may be adeemed, even in the same will, notwithstanding the technical point that it is, in strictness, not datum but adscriptum¹*. Apart from this, the slave is,

¹ 40. 4. 17. pr.; cp. 35. 1. 29. » 40. 4. 12. pr. ’ 28. 7. 8.

â€? 40. 4. 12. 1. · Ante, p. 471. » 40. 4. 12. 2.

’ 40. 4. 36 ; 40. 7. 13. 3. ⁸ 28. 7. 8. 6, 7. » 40. 4.12. pr.

¹⁸ 38. 1. 7. pr., 2; 40. 4. 36. ¹¹ 40. 12. 44. pr.

¹² Bodemeyer, op. tit. 54, cites many attempts at explaining the allowance of conditiones iurisiurandi. His own requires impedire to mean “make more difficult.” Bufnoir, op. tit. 47, following Vangerow, Band. § 434, thinks the right conferred by the Praetor in remitting the penalty is only praetorian, and as there is no true praetorian liberty, the freedom could be acquired only by doing the thing. Here too impedire is given the above sense. See also Karlowa, R. R. G. 2. 137. But Pernice has shewn (Labeo, 3. 54) that the effect of release of the condition is not to give a merely praetorian title.

>» 34. 4. 3. 9; 35. 1. 53.

till the condition is wholly satisfied, the slave of the heir¹. He may not know at once when it is satisfied, and it is provided that if, in ignorance of the arrival of the condition, he enters on an inheritance at the order of the heres who holds him, he is not personally bound’. But if he is in doubt as to whether it is satisfied or not, Paul cites Julian as holding that he is, so to speak, “ put upon enquiry,” and so is bound’.

There remains for discussion, in relation to conditions in general, the question, what amounts to fulfilment of the condition ? The rule is that it must be actually and completely fulfilled⁴, though in obscura voluntate manumittentis favendum est libertati⁶.

The meaning of a condition “ to serve my heres (or Titius) for a certain time ” is the subject of discussion. Servire and operas dare are equivalent, neither need involve slavery and thus they may be done to a third person⁸. The service must be personally rendered⁷, and 100 operae means 100 days’ work⁸. They should be rendered continuously, but days during which the man is prevented from working by illness or other good cause are credited to him as days of work’.

Satisfaction of the condition is all that is needed¹⁸ If the condition is a promise or oath to do something, the promise or oath fulfils the condition and the freedom is gained, though, as we have seen, the promise is void as having been made by a slave¹¹. Where the condition was “ to be free on handing over the peculium" and the man gave up everything, he was free though he owed debts to his owner¹â€™. In answering the question whether the condition is satisfied or not there is a general favor libertatis. Thus where the condition of liberty given to a woman was “ if her first child be a male,” and she has twins, one of each sex, there is a presumption, apart from actual knowledge, that the male is the elder, and thus the daughter is ingenuaℱ. If two are made free on the same condition, it is applied separately to each, if this is possible and will save the gift, e.g. “if they are mine at my death¹⁴.”

¹ Ulp. 2. 2; D. 40. 7. 9. pr. Ante, p. 286.

² 29. 2. 74. 3. As to any right of bonae fidei possessor, ante, p. 346.

⁸ 29. 2. 74. 4. < 40. 7. 3. 12.

⁸ 50. 17.179. “ To be free when my debts are paid.” They must be paid: it is immaterial that the heres is rich, 40. 7. 39.1. If the heres wilfully delays, the rules as to prevention may come into operation, 40. 5. 41. 1. “ To be free if he goes to Capua.” He must go though the journey be otherwise aimless, 40. 4. 61. 1. The element of time may be material here. See also 35.1. 44. 10.

⁸ 40. 7. 4. 4, 41. pr. The text (4. 4) adds that the will may shew that the testator meant “ be a slave to.” See post, p. 493.

⁷ 40. 4. 13.pr.; 40. 7. 20. 5, 39. 5. ⁸ 40. 7. 20. 5.

⁸ 40. 7. 4. 5. This does not turn on impossibility, but on the notion that such a prevention is an ordinary incident of continuous service. Days in which the man is in fuga, or raising a claim of liberty, or suffering punishment for crime, do not count, and other days must be served in their stead, 40. 7. 4. 8, 14. 1, 39. 3.

¹⁰ But cum darepoterit means when he does pay, 40. 7. 4. 12.

40. 7. 13. 3, 21, 41. 1. ¹² 40. 7. 40. 1. ¹⁸ 34. 5.10. 1.

¹⁴ 32. 29. 4. The rule is the same in legacy. Where two were to be free on paying 10, either was free on paying 5. As to legacies, cp. 35. 1. 112. 1.

So where the condition was si rationes reddiderint, unless it had been a joint administration, in which case neither was free till the whole was adjusted¹. If the act is indivisible, both are free if one has done it’. If there are two different gifts the statuliber may choose the easier* (which he may not in legacy), or, as it is put in other texts, the levis­sima scriptura applies, and that is lemssima through which liberty is attained⁴, i.e. he has not to make a choice, but may take the benefit of that which occurs first. This applies only if the two gifts are distinct; if they are coniunctim, i.e. “if he do this and that,” he must satisfy both*.

If a statuliber is alienated, and the condition is an act to be done by him in relation to someone else, the question arises: to whom must he satisfy the condition ? This will be discussed in relation to two specially important conditions,pecuniam dare and rationes reddere, which will need separate discussion. Here it is enough to lay down the general rule. Any condition which admits of it may be done to the acquirer⁸, but per­sonal service, such as to teach his child, remains with the heres though the man be assigned⁷. But it must be an alienation of the dominium: to give a usufruct to a third person does not enable the payment or other act to be rendered to the usufructuary⁸. It should be added that if the condition is doing work for anyone, the slave himself must do it, but money may be paid by anyone".

There were, however, circumstances under which the liberty took effect though the condition was not in fact satisfied. Before entering on these it is desirable to point out that dies and conditio are some­times intermingled in a way which makes two observations necessary.

(i) The same modality may be construed as dies in one case and as conditio in another, the decision turning sometimes on construction of the testator’s language and in others on extraneous considerations such as favor libertatis. This is most commonly illustrated by cases in which a gift is to take effect “ when X is 20,” and the like. Here if X dies under 20 it may be said that the condition has failed, or it may be said that the words were a mere way of describing a certain date which has in fact arrived. Both views are found¹⁰.

>40.4.13.2. *h.l.pr.

» 35. 1. 51. pr., 87—89. ⁴ 35. 1. 35 ; 40. 4. 5..

⁶ 40. 4. 45. In one case the words were “ Let S be free when he is 30. Let S be not free if he does not give 10.” These are not alternative gifts, one ex die and the other conditional. The second is in form an ademption, which puts the gift under a contrary condition, so that he is to be free at 30 if he nays 10, D. 40. 7. 13. 5. Cp. 34. 4.14. pr.; 40. 4. 59. 2.

⁶ 40. 7. 6. 3. The rule is as old as the Xn Tables, h. t. 29.1. See Appleton, Le Testament Romain, 86.

⁷ 40. 7. 6. 7, which shews a more complex rule in relation to the condition of rendering accounts.

⁸ 40. 7. 7. f h. t. 39. 5.

¹⁰ e.g. 36. 2. 22. pr.; C.6.53.5; 7.2.8. Brunetti, Dies Incertus, 160; port, p. 491.

oh. xxi] Impossibility as a Release 489

(ii) A condition may, indeed it usually will, include dies. Here, though from any cause the donee be released from the condition, the gift will not take effect till the time has elapsed. This is illustrated in a number of texts¹ and must be borne in mind when these cases of release are under discussion. The plain reason is given, by Paul, that it is absurd that the gift should take effect before it would have done so had the condition been satisfied in the ordinary way², the testator having imposed both condition and dies, and the former alone having been released. If, however, there is no express dies and the condition is invalid ab initio, the dies is not considered².

In discussing the circumstances under which a condition is released it is necessary to look somewhat closely at the Roman conception of Impossibility. We know that where a condition is impossible in the nature of things⁴ it is struck out in all gifts by will. But it is not easy to say what is impossible in the nature of things. What seemed incon­ceivable to a Roman might be an everyday event now⁵. But the exact position of the line is not important: the point to note is that it commonly means a condition which is on the face of it inconceivable’. This opens up the question how the jurists looked at impossibility on the facts—latent impossibility, either existing at the time of the will or supervening. It may be said at once that, at least in the case of wills, they do not seem in general to have applied the notion of impos­sibility to cases of supervening impossibility—it is indeed difficult to treat a provision as non scriptum by reason of an event subsequent to the making of the document. It is in fact casus rather than impossi­bility⁷. There are evidences of doubt as to the treatment of cases in which the condition becomes, or is, impossible in fact, before the aditio: with these we shall shortly deal. Another point of interest is that the illustrations of patent impossibility always seem to be cases in which there was something to be done or left undone by the donee. Such a condition as “if there shall be 370 days in the year” is never taken as an illustration. In one text which contains some dispute, and traces of more, the condition is si filia et mater mea vivent and one of these is dead at the time of the will⁸. This is a case of latent impossibility existing at the time of the will, and the condition does not contemplate action by the donee. It is dear that it gives difficulty and Pomponius, or perhaps Tribonian, describes it as a case of “ quasi-impossibility.”

¹ e.g. 40. 4. 41. 1 (corrupt); 40. 6. 41.pr.; 40. 7. 3. 15, 4. 2, 20. 5. Cp. 34. 1.18. 2; 45. 1. 8.

² 40. 7. 4. 2. The final words in 40. 4. 41. 1 are an interpolation.

⁸ Cp. 45.1. 8. An analogous rule in case of prevention, post, p. 493.

⁴ P. 3. 4 b. 1. See ante, p. 484.

⁸ A man was instituted “ if he build a monument in three days.” Even this the lawyers, after hesitation, decide to be quasi impossible, 28. 7. 6.

⁶ There were forms of such impossibility in institutions which could not have arisen in gifts erf liberty, 28. 7. 4. nr., 10, 20. pr.

⁷ Cp. C. 6. 46. 6. ⁸ 35. 1. 6. 1, in jin.

We can now consider the different cases which present themselves in the texts.

There is a well marked type of case in which the gift is made to depend on the attainment of a certain age by X—the man is to be free if (or when) X is 20. This is a case of dies incertus an, certus quando. Here, in the texts which speak of gifts of liberty either fiduciary or direct, the death of X under 20 is treated as immaterial¹. On the other hand some of the texts make it clear, by way of contrast, that in the case of legacy on the same condition, the death under the age would be regarded as amounting to failure of a condition^{[1784] [1785] [1786]. This of itself would shew that the rule has nothing to do with impossibility, since, if it were treated as impossible both sorts of gift would be valid. We are, however, left in no doubt: we are told that it is an exceptional rule laid down favore libertatis, the modality being treated as dies in this case’.}

The case of ordinary condition to be performed by the donee is treated in the same way, where impossibility supervenes after aditio. In the case of legacy, the condition is regarded as having failed^{[1787] [1788] [1789]: in gifts of liberty the other view is taken, favore libertatis, and the gift takes effect’. In the case in which the condition is the payment of money to X and X dies before it is paid, there may be the further complication that the man was not ready to pay the money at the time of the death. But Julian observes that the whole favourable rule is a matter of constitutor», ius resulting from favor libertatis and not resting on any logical principle of interpretation, and that thus the man will be free if at any time he has the money[1790].}

In the case of latent impossibility arising before the aditio there is more difficulty. Where the thing not only is impossible, but always was, i.e. where it assumes a state of facts which never existed, it is a falsa conditio and the gift whether of property or liberty is good^[1791]. The same seems to be the case where part of the condition is on that footing, the rest being separable, e.g. a gift of hereditas “ if my wife and my daughter X survive,” and the testator never had a daughter: the

ch. xxi] Latent Impossibility existing at Aditio 491 gift is good, the reference to the daughter being ignored¹. Where the impossibility is one which arises before the will is opened, whether before or after it is made, the matter is complicated by questions of interpretation, themselves affected by the existence of provisions inde­pendent of the purpose of the gift. But, so far as can be made out, the view of the later classics, accepted in the Digest, seems to be that in case of liberty the gift takes effect, though it is clear that this is favore libertatis, and independent of logic³. In case of legacy the gift fails’, so that it is no question of impossibility. But there are some texts which seem at least to contradict these conclusions and which need to be carefully looked at.

In one text Paul makes the liberty fail*. The man is to be free if a usufruct in him given to X by the will ceases to exist. In point of fact it never arose, the fructuary not having survived to take it. On such facts the gift fails, says Paul. The condition fails, since that which never began cannot have ended. The rule, which is from Neratius, ignores favor libertatis, and is moreover a piece of literal interpretation, which quite disregards the plain intent of the testator. The gift is in effect one at the death of X.

Where a woman is to have a legacy if she marries arbitratu Seii she will take the legacy though she marry without his consent, and even though he be dead, vivo testators¹. This is due to the fact that such a condition is in practice one that she shall not marry without his consent, which is void’. The reason assigned for allowing the gift to be good though S be dead is given, by Papinian in the words quia suspensa quoque pro nihilo foret, words which shew that some special reason was needed and that in ordinary cases the gift would have failed. Pomponius, however, while dealing with this case⁷ applies the same rule to a legacy, si sum manumisisses, where the man died vivo testators. His reasoning uses language appropriate to prevention, quia per te non stetit quominus perveniat ad libertatem, a principle which, as we shall shortly see, ought not to be applied with this breadth to legacies. Javolenus applies it to a case of impossibility arising before aditio in a case of liberty, but expressly observes that it is favore libertatis³.

> 28.5. 46. « 40. 7. 28. pr., 89. 4. »35.1.31.

* h. t. 96. pr. & 30. 54. 1, 2 ; 35. 1. 72. 4, 28. pr.

⁶ Other texts, Pothier ad D. 35. 1, § xxxviii.

? 30.54.2. Pothier (ad h. I.) supposes this rule to apply where it is a condition not involving cooperation of another person.

⁸ 40. 7. 28. pr. Bufnoir, op. cit. 91, shews that Ulpian in 9. 2. 23. 2 holds that the condition has failed in such a case. He cites and refutes reconciliations by Vangerow and others. B. himself holds that 30. 54. 2 deals with modus. But on its terms it does not. But, where one is to he heres if he swears to free 8, and S dies vivo testators, the gift is good, though in such cases the condition is remitted and actual freeing substituted. This confirms our rule, since the substituted act is to be done after the acquisition. It is modus, not condition: this, as appears from the hypotheses cited in support, is why the gift is good, 28. 7. 8. 7. Ante, p. 483. Bufnoir, op. cit. 90.

The view stated above as that of the later classics was not accepted without dispute, of which the texts shew traces. Thus we are told that Labeo and Ofilius declared a gift of liberty to fail, if the person to whom money was to be paid as a condition died, vivo testatore, and that Trebatius held the same view if the death was before the will was made¹. This last view may, of course, possibly mean merely that Trebatius has a derisory gift in mind^{[1792] [1793]. Pomponius[1794] discusses a case of gift of hereditas on condition of freeing certain slaves, some of whom were in fact dead when the will was made. He cites Neratius, (whom we have just seen holding a severe view[1795],) as thinking that the con­dition has failed, and that the notion of impossibility is not applicable. But he cites Labeo and Servius as holding that in a similar case, where however the condition was not something to be done[1796] [1797] but the survival of two persons of whom one was dead at the time the will was made, the condition has not failed, and Sabinus and Cassius as holding that this is a case of quasi-impossibility’. This text involves two points, i.e. that the distinction between legacy and liberty where the impossibility occurs vivo testatore must be limited to the case post testamentum factum, and that where a condition has separable parts, each part has to be considered by itself for the purpose of these rules. But this can hardly be an adequate account of the law where impossibility had arisen before the will was made. The text tells us too little. Probably the state of the testator’s knowledge was material, and no doubt this text is an indication of far-reaching differences of opinion[1798].}

Prevention of fulfilment brings other distinctions into prominence. Just as it was difficult to set qxact limits to the notion “ impossible,” so it is not quite easy to say exactly what is meant by “ prevention,” but for the purpose of the rules now to be stated, two points must be made clear. Prevention is essentially an interference with the action of the donee. Hence the rules do not apply to conditions which have to be satisfied without the cooperation of the donee. This appears, apart from the specific rules which express the distinction, in the language commonly used in expressing the principle generally. The texts usually speak of his satisfaction of the condition being prevented^[1799]. Further, the thing done does not amount to prevention, unless it was

ch. xxi] Conditions: Prevention of FuljUment 493 done with a view to prevention¹; at least where the prevention takes the form of prohibition and not of rendering the thing impossible. On the other hand if it is a definite act of prevention it is presumed to have been so intended rather than as a normal exercise of right’.

The general rule on the matter is that, where the statuliber is prevented from doing the act which’ is a fulfilment of the condition, he is placed in the same position as if the act had been done. It is immaterial for this purpose whether the person who prevents is one like the heres, interested in its non-performance⁴, or one to whom or with whose cooperation it is to be done’, or any third person: it is enough that someone prevented fulfilment⁸. It may be noted that the rule in hereditas and legacy is not so wide : it releases only in the first two cases⁷. On the other hand where the condition is to be fulfilled without the cooperation of the slave and the person to do it refuses, or fails to do it, there is no relief, any more than there would be in any other form of gift⁸.

Prevention has the effect of putting the man in the same position as if he had not been prevented. If even apart from the prevention he could not have satisfied the condition, the prevention does not make him free: it is not true that non per eum stat. Thus where the heres refuses to receive the accounts but the statuliber is in arrear and has not the means to pay up the balance due from him he is not free*. In the same way, if the act would necessarily take time, e.g. to do so many days’ work for the heres or another¹⁸, to go to Capua, to go to Spain and gather in the crops¹¹, or to make a series of periodical payments¹², the man is not freed by refusal to let him begin work, or start on the journey, or make the first payment. These rules, however, seem to apply only to such prevention as takes the form of prohibition or refusal: if it takes the form of making the thing impossible, the man seems to be free at once. Thus where he was directed servire heredi for a time and was manumitted or sold by the heres, he was free ex tes- tamento¹³ at once¹⁴. The word servire here means “ be slave to ” : in the one case this is made impossible : in the other it is made impossible for the statuliber to do it. It is not clear why the time was not required

¹ 40. 7. 38. ² A. t. 3. 3.

⁸ With or without the cooperation of another person.

< Ulp. 2. 5; Festus, loc. ctt.; D. 35. 1. 24, 57, 78; 40. 7. 3. 13, 4. 16, 17, 23. 1; 50. 17. 161.

8 Or his guardian, Ulp. 2. 6; D. 35. 1. 78; 40. 7. 3. 10, etc.

« 40. 5. 55; 40. 7. 3.pr.

⁷ 28. 7. 3, 11; 30. 92. 1; 35. 1. 14, 21, 31; 36. 2. 5. 5; 50. 17. 161.

⁶ Consistently with all this, if one heres prevents a payment on which liberty is conditioned, the statuliber is free, 40. 7. 3. 4, while if he was to pay to two and one refuses he is released only pro parte, but can tender the same money to the other, 40. 7. 4. 3. See post, p. 503.

9 40. 7. 34. 1. w 34. 1. 57; 40. 7. 3. 15, 4. 4, 20. 5.

i¹ Ibid.; 40. 7. 34. 1. ¹⁸ As to a difficulty in 40. 4. 41. l,post, p. 499,

is 40. 7. 3.15, 17.

¹⁴ In classical law the manumission would commonly make him only a latin: in no case would it make him libertus orcinus.

494 Manumission by Will: Rendering accounts [pt. n to elapse. One text purports to give the reason*, but all this amounts to is that refusal is only prevention of a part: the logical result would be that in the case where it was made impossible the man is free at the expiry of the time with no further tender, not that he is free at once.

An enactment of Justinian’s² deals with a case which he declares to have divided the jurists. A slave was to be free on paying money to the heir. He started to travel with it to the heir, but was robbed of the money on the way. The question was, did this suffice under the condition ? Clearly there was no impossibility, though on the facts it had become impossible to the man. The question seems to mean: was this prevention, i.e., could it be said per eum non stare, since it was his going on the journey which made it possible to rob him ? Justinian settles it by deciding that in such cases per eum stat only when he intentionally does not fulfil the condition: in all other cases of preven­tion by persons or casus he is to be free, but to remain liable for the value of the render, except in so far as it has been repudiated by the person to whom it was to have been made.

There remain for consideration two conditions of exceptional importance.

(a) Rationes reddere. The importance of this condition is shewn by the frequency of its appearance in the Digest. It is found also in a surviving roman will⁸. Though, as we have seen, all freed slaves may be called on to render an account⁴ of what they hold, the importance of making it a condition is that less risk is run, since he must make the statement and render before he is free. Moreover all that can be required of him after the freedom, apart from the condition, is that he hand over the accounts and the property of the testator which he holds. No personal action will lie against him for anything done while he was a slave’.

The condition practically means that he must state and account for all moneys that he has had to administer’. He must make his account in good faith and with due care, though a mere mistake, even negligent, is not a breach of the condition⁷. His account must shew his gestio to have been in good faith, at least to the extent that he must debit himself with anything he has wrongfully taken away, and of course there must be no false credits⁸. He must state all needful details, going over his account books, and giving a proper account of matters not in writing⁹. The condition covers the whole field of administration,

» 40. 7. 3. 15. ² C. 6. 46. 6.

⁸ Testamentum Dasumii, Girard, Textes, 767.

* 40. 4. 17.1; 40. 5. 37. � Ibid.; C. 7. 2. 4. Post, eh. xxrx.

⁸ If he has never administered it is a void condition, 40. 7. 26. 1.

’ 35. 1. 32, 112. 3; 40. 4. 22. s ₈₅. i. _1U. _{c 7}. _{2 4}

⁸ 40. 7. 26.pr.

ch. xxi] Condition of Rendering accounts: Interpretation 495 not merely trading¹. If he has given credit, he must shew, not indeed that the debtor is solvent, but that at the time of the transaction he was such that a bonus paterfamilias might reasonably have such dealings with him². He has not merely to render an account: he must hand over the reliqua, i.e. all property of the estate in his posses­sion³. This implies that he must get in all debts now recoverable, rents due and so forth, and must make good anything he has made away with and bad debts incurred through his negligence⁴, but not losses resulting from casus³. What is required is a true and just account and render: the law prescribes no exact steps’. He must account for receipts since the death, and, if this seems to be intended, give the same detailed account of his administration since, as before⁷. The account must be rendered where the person is, to whom it is to be rendered, at least if that person is away on public business, but in other cases reasonable arrangements may be made to suit the case⁸, and a deputy appointed to receive the account³. Each heir is entitled to his share of the reliqua, and though this may be excluded by apt words, it will not be by the naming of some of the heirs in the condition¹³.

The condition adds, as we have seen, to the obligation on the slave¹¹, but a good deal turns on the wording of it. If it was merely reliqua reddereℱ, this is satisfied by paying over the balance without giving the full means of examining the accounts which was needed under the condition rationes reddere¹³. If the condition is that he is so to render accounts as to satisfy X, he must do this: even satisfaction of the curator of X apud iudicem will not suffice, unless X is present and assenting¹⁴. If the condition is si rationes diligenier tractasset, this involves rationes reddere, and proof that his diligence has been exer­cised in the interest of the master, and not in his own¹³. Where it is si rationes diligenter tractasse videbitur this means videri poterit¹â€™. If a time is set within which the account is to be rendered, and by his fault it is not done at the expiry of the time, he has not satisfied the condition¹⁷. As we have seen, if the condition is to account within

i 35. 1. Ill; 40. 5. 41.11 omne quod quoquo genere actum jidemque senoi respiceret. a 35. 1. Ill; 40. 5. 41. 17.

a 35.1. Ill; 40. 7. 31; C. 7. 2. 4. Liquid assets must be given at once and security for the rest, 40. 7. 5. pr. ⁴ 40. 7. 40. 4—8.

6 40. 4. 22; 40. 5. 41. 7. He is responsible for adiutores if their malversation is in any way due to his negligence, 40. 7. 40. 4.

fl Thus the fact that where peculium is left to him he takes it away before he has rendered his account does not bar liberty, h. I. 6. Nor does the fact that the testator, having been Hl, has not signed the accounts for a long time, h. I. 3.

7 h. I. pr.; 40. 5. 41. 10. If needed an arbiter will be appointed to settle disputed points, 35. 1. 50; 40. 1. 5. 1; 40. 7. 21. pr.

» 35. 1. 112. 3. ⁹ 40. 7. 4.pr. ¹⁰ h. t. 12; post, p. 500.

n 3. 5. 16—18. 1, 44. 1., ¹³ 35.1. 82.

¹⁹ Ibid. A condition ne rationes reddat is not a gift of peculium, but an absolution from the duty of rendering strict account, the liberty being unconditional. He must hand over any balance and account for dolus but there will be no enquiry into negligence, 30.119.

i* 40. 4. 53. ¹⁶ h. t. 8; 40. 7. 21. pr.

¹⁶ i.e. to an arbiter, 40. 7. 21. pr. ¹⁷ 40. 5. 41.12.

496 Condition of Rendering accounts: Prevention [pt. n

30 days from the death, and the aditio is not till later, the liberty does not fail, as it is not his fault¹. Titius by his will² left certain servi adores to different persons, si rationes heredi reddiderint, and in another place said, “ All the slaves whom I have legated or freed, I wish to render their accounts within four months, and to be handed over to those to whom I have left them.” Later in the will he freed other adores, with the condition si rationes heredi reddiderint. The time passed and without any fault of the heir³, the accounts were not rendered. Were the men barred or could they still claim their liberty by satisfying the condition later? The answer given by Scaevola is that it is for the person before whom the case comes to consider whether this is intended as a condition limiting the time given to the slaves or whether it is really intended to impose speediness on the heirs, by preventing them from dawdling in the matter. In the former case the claim is barred : in the latter it is not⁴.

The ordinary rules apply as to prevention. If the slave is prevented by the heres from paying over the reliqua after the account has been adjusted and the res peculiares have been sold, he is free as if he had paid⁵. So if the heres is in mora in receiving the accounts, the slave is free if he tenders them and the balance^[1800]. The text adds that it is for the arbiter to decide which party is in mora, and to determine accordingly, and further that declaration of waiver of right to the balance, by the heir, satisfies the condition.

Some cases in which the difficulties are really of construction are discussed^[1801]. In one of these the heres is impubes. S is freed ration# reddita, and he agrees with the tutor to take some of the money due and divide it. The tutor certifies the account as correct. The man is not free, for though the rule is that the reliqua may be paid to the tutor of the heres, and his prevention of payment has the same effect as the heir’s, this applies only where the statuliber and the tutor are not fraudulent. Here, the text says, as in alienation of property, the tutor can deprive the pupillus only where there is no collusive fraud^[1802].

Money paid under such a condition is a mortis causa capio, though it is not a donatio, no matter by whom or to whom it is paid^[1803]. A question of some difficulty arises where the lex Falcidia and similar legislation comes into operation. The point of importance seems to be that though it is a mortis causa capio, it is not necessarily acquired hereditaria iure². If, however, it is ex bonis mortui (which certainly covers the peculium which the statuliber had at the time of the death) and is paid to the heir, it is acquired iure hereditaria and must be debited to him, when his quarter is being made up for the purpose of the Falcidia³. So too it counts towards the half that an orbus may take⁴. If there are two or more heirs, and the payment is to be made to one, only that part of it which corresponds to his share in the hereditas is acquired hereditaria iure, and counts towards making up bis quarter : the rest is an independent mortis causa capio⁶. If it is paid from outside—not ex bonis mortui, the lex Falcidia has no application to it⁸. There are, however, cases in which it seems to be doubtful whether it is ex bonis mortui or not. According to Ulpian, if the slave acquires the money only after the death, it cannot be said to be ex bonis mortui, and so will not be imputable⁷. Papinian, however, holds that even though it were given to the statuliber to be paid to the heres it be­comes part of the peculium, and even if it were handed direct by the extraneus in the presence of the libertus. It is only if it is handed over by a third person without the presence of the libertus that the taker holds it really aliunde, so that it is not acquired hereditaria iure⁶. Apparently Ulpian’s view would exclude all acquisitions to the estate after the death—which certainly was not the law for his time⁹.

Money validly paid under the condition is of course irrecoverable. But if it was not due it can be recovered, like an indebitum^w. The remedy will be condictio or vindicatio according as the transaction has or has not vested ownership in the heres. This is illustrated by many texts. There can be no recovery if it was paid in full knowledge of the facts^[1804]. It is not strictly an indebitum in any case, and it is treated mainly under the head, possibly a creation of the compilers, of condictio causa data causa non secuta¹². If there is no condition and he is really free, and he has paid it out of his own money he can condict it¹. But if he has given it out of peculium to the heres, thinking, as he would, that the money already belonged to the heres, there has been no transfer of dominium, and if it came, for instance, from a part of the peculium which he has acquired after he is free, the ownership remains with him and he can vindicate’. And supposing the payment was not to the heres but to an extraneus, here too if it was a res peculiaris, as he was not authorised so to deal with it, the ownership will not pass and the heres can vindicate. But if a third person has paid it to the extraneus, or the man has paid it himself, after he is really free, out of his property, the dominium will pass and the proper remedy is condictio^s.

Proculus lays down the same rule for the case where the will is not valid : money paid ex peculio to an outsider can be vindicated^{[1805] [1806]. Scaevola discusses another case. A man really free but supposed to be a slave, receives a gift of liberty from his supposed dominus on a condition of 10 annual payments to the heres. After paying 8 he discovers that he is an ingenuus. He can recover by condictio if he has paid out of what he has acquired otherwise than ex operis or ex re possessoris. If it did come from that, he was merely giving the heres what was his already’. In all these cases there is no real divergence of opinion, but in another case there is. In a will a slave receives liberty on condition of payment to the heres. By a codicil he receives an unconditional gift. Before he hears of the codicil he pays the money. Can he recover it? Celsus pater thinks he cannot. Celsus filius, on grounds of equity, says that he can, and Ulpian adopts this view[1807]. Nothing is said as to the reason for the view of the elder Celsus, or as to the source from which the money came, and each omission increases the difficulty of repairing the other. For the question to arise at all the money must have been paid from something which was not at the time of payment the property of the heres. The point is perhaps that the condition not having been adeemed, perhaps not having been adeemable[1808], exists, and he has acted under the wrong gift[1809].}

As to what amounts to fulfilment of the condition, the ordinary principles apply, but some special rules need mention. A condition to give is satisfied by payment by a third person either with or without

ch. xxi] Condition: Payment of Money: Fulfilment 499 the presence of the actual slave¹. The whole must be paid^{[1810] [1811]. Further, as the payment is in satisfaction and not under an obligation, there is no alienation at all, till all is paid: up to that time the owner of it can vindicate it, and the alienation does not relate back for any purpose[1812]. If a time is fixed, the payment must be within the time[1813], which runs from aditio if the will is not explicitÂź, and even “thirty days from the death ” is reckoned from the aditio, at least if it is necessary so to do to save the gift[1814]. Where a man was to be free on paying 10 a month for 5 years, he is not free unless he pays it every month[1815]. The heir’s refusal of one payment does not release the statu­liber from the others, though, as it does from that one, and the heres may not change his mind, the same money may be offered when the next pay-day comes8. These texts point out, as we have already seen, that such directions involve dies as well as conditio. It is, therefore, surprising to find two texts which say that if he offers all future pay­ments at the date fixed for the first he will be free. In one of these texts the rule is justified by the consideration that the earlier loss of the slave is compensated by the earlier receipt of the money[1816]. As the interest of the money bears no necessary relation to the value of the slave’s services, the argument is not strong, and the form of the remarks strongly suggests Tribonian. In the other10 it is given as a benignior rule, and we are told that both benefit, the one by earlier freedom, the other by earlier payment. The point is the same, with the added suggestive fact that this line of argument is one which Justinian employs elsewhere11. Altogether it is difficult to credit this view to the classical law.}

The giving must make the alienee owner of the money: thus it may not be stolen money¹². The transfer must not be merely illusory. Where the heres gave the man the money, “ to pay me with,” and he returned it, he was not free, though he would have been if the gift to him had been absolute¹³. It need not take the form of an actual traditio to the heres. We have just seen that release of the payment sufficed¹⁴. If at the death of the heres the man is found to have enriched the hereditas to the required amount, e.g. by payment to creditors, pro­vision of stores or the like, he is free¹Âź.

If it is payable to the heres, the commonest case, since it is payable

to him if the will does not say to whom it is payable¹, we are told in several texts that it may be paid to heres heredis, the rule being due to Hadrian^{[1817] [1818]. The point is that the personality of the payee is not imported into the condition, favore libertatis, for in legacy it is[1819]. If the heres has died leaving no successor, we are told by Hermogenianus, that, constitute iure, the man is free, without paying at all. It is not clear whether he must have the money at the time of failure of heirs: this is suggested by the earlier part of the text[1820] [1821], though the contrary suggestion is found in a remark of Julian reported by Ulpian, already considered[1822] [1823]. The rule is the same even if the heres is mentioned by name6. He cannot pay it to a pupillus heres without the auctoritas of the tutor’’, a rule laid down rather on grounds of analogy than on strict principle, for it is not the payment of a debt[1824] [1825] [1826] [1827]. If the words are h'eredibus date, they take pro rata, but if they are mentioned by name they take partes viriles”. The text seems to indicate that if they are called heredes and also named they take pro rata1”. If one heres renounces the institution the payment is to be made wholly to the other11, though if having actually entered he refuses the money, there is no accrual to the other, who is only entitled to his share, which may be satisfied with the same money[1828].}

If it is to be paid to an extraneus the rules are much the same. If the payee is dead, the money can be paid to his heir, or if there be none, the man is free without paying, if at any time he has the means to pay^[1829]. If, however, there are several extranei, they take partes writes unless some other division is prescribed, and thus where there are both heredes and extranei, the heredes will take partes hereditariae, and the extranei, partes writes^{[1830] [1831].}

It may be added that a payment to the heir is retained by him even though he hands over the inheritance under a fideicommissum and even though he entered only under compulsion¹⁶. Ulpian tells us that if the payment was ordered to be to an extraneus, and he became heres to the heres, the payment would be made to him non quasi in extranei

ch. xxi] Condition: Payment of Money: Fulfilment 501 persona, sed quasi in heredis^[1832]. The point appears to be that it might be imputable from the point of view of the lex Falcidia².

We have seen that, on alienation, conditions dando go to the alienee, while services remain with the heres³. Thus, if the condition is rationes reddere, the account is made and the books are produced to the heres, but the money is paid to the alienee¹. The rule applies not only to sale but to all transfers of dominium, e.g. under sale, gift or legacy⁵. And the heir of the acquirer succeeds to the right as heres heredis does⁶. But it must be an alienation of dominium : conferring a usufruct on a third person does not entitle him to receive the pay­ment⁷. On successive alienations the right passes to the last alienee⁸. If the purchase was by a slave, payment may be to the master or to the slave, if the purchase was on account of peculium, and this is not adeemed⁹. A buyer of a part must be paid a proportionate part of the money¹⁶. If the alienation is after payment of a part, the rest must be paid to the alienee^[1833]. We are told that on sale the heres may reserve the payment to himself, and this will have the effect, not merely of a covenant between buyer and seller, but of compelling the man to pay to the heres, in order to satisfy the condition. So in the same case he may nominate some other person to receive the money, with a similar effect¹². It may be presumed that an alienee has the same right, and that a testator may by express words limit and vary the rules expressed in this paragraph. But whether an extraneus not an alienee can nominate a person to receive the payment cannot be confidently stated, though it is suggested by the last words of this lex.

We are told^is, as we should expect, that one dare iussus to a slave (heres or not) may not pay his master except with the slave’s consent, or vice versa, unless the money is versa in rem domini. The texts are general¹¹, but do not expressly refer to gifts of liberty, and though these are probably the commonest case, there are others. And two texts create doubt. In one we are told¹⁵: certe statuliber quin domino dare debeat non est dubium. The use of the word debeat shews, when the adjoining texts are looked at, that there is no concession here. The text, which has been shortened, may be merely emphasising the rule,

but it may refer to the case of one directed to pay to a fellow-slave.

Here as they are in the same hereditas, and the money is res heredi- 502 Condition: Payment of Money: Fulfilment [pt. n taria, payment to a fellow-slave would effect no change in possession, and it may have been thought that for this reason it must be to the dominus. In either case it does not affect the rule. Another text is more serious. We learn¹ that where the condition of liberty is payment to a filiusfamilias heres, it may be done to the father, since he gets the profit of the hereditas, which applies equally to a slave. It is observable that nothing is said of favor libertatis, and the reason would equally apply to other cases in which the rule was as we have seen otherwise. Thus though the text may mean that the rule was relaxed in gifts of liberty, it is more likely that it is an individual view of Ulpian’s.

There is difficulty where the inheritance is disputed: there is only one text and that as it stands is unintelligible^{[1834] [1835]. It seems to begin by assuming that though the will be upset by a judgment, the gift of liberty on paying 10 to the heres may still be good. Part of its incom­prehensibility is swept away if we adopt Kruger’s emendation and read heredis for his early in the text[1836]. On that view the text raises no difficulty as to the date of introduction of the principle that the setting aside the will by judgment in favour of a heres ab intestate is a bar to all claims under it[1837] [1838] [1839] [1840] [1841]. The question it would raise is this: S is to be free on paying 10 to the heres. The heres enters and dies, and there is a dispute as to his succession. It is between one claiming under a will and one claiming on intestacy. The latter wins¼ and the man asks if he can pay the winner. Quintus Mucius says yes, and, further, that whatever be the truth of the matter he cannot pay the one who has been beaten. Labeo thinks that, as he is in no way claiming under this succession, he is free if he pays to the party really entitled. Aristo gives Celsus an opinion to the effect that only the winner is capable of being paid the money: if he is the true heres well and good, if not it is a case of alienation and he is entitled in that way. If the money were paid to the loser, it would be his duty to hand it over, like other acquisitions, to the winner, and when that was done no doubt the man would be free’.}

Whether the money is to be paid to the heres or to an extraneus, it can always be paid out of peculiuml. The statuliber can of course pay it, if he prefers, from other sources’, but not out of moneys entrusted to ch. xxi] Condition: Payment of Money: Prevention 503 him and not forming part of the peculium¹. He may pay it out of subsequent earnings, but may not count towards it money paid to the heres in lieu of services due to him, any more than he could the rent of a farm he hired of the heres^{[1842] [1843] [1844]. Even though he is alienated sine peculio, he can still pay it ex peculio2. But he may not pay it out of the peculium belonging to his new master, for the testator’s intention could not be extended to that, not even though he had been sold cum peculio and the vendor had failed to hand it over[1845] [1846] [1847] [1848]. If he is ordered to pay it ex peculio, and has none, or owes all that is in it to his dominus, he cannot at that time satisfy the condition at all’.}

If a person to whom liberty has been given on such a condition is captured in war, and is redeemed, he may satisfy the condition out of his peculium coram redemptore, provided it is not ex operis or ex re redemptoris*, but he will still be subject to the lien of the redemptor’.

In this case as in others prohibition makes the man free. If the heres refuses the payment or refuses to let it be made to the extraneus, the man is ipso facto free^{[1849] [1850] [1851] [1852] [1853]. If it is to be paid to a coheir and one heir refuses to allow the payment, the man is free8. There are other things besides direct refusal which have this effect. If there is a debt due to the peculium, and the heres refuses to sue for it so as to provide means to fulfil the condition, or money is due from the heres to the peculium, and he will not pay it, the slave is free from the mora. Servius was inclined to limit this to the case where the peculium was left to the slave, but the wider view prevailed, and seems the more logical18 If the heres delays aditio intentionally, the slave is free if he had the money at the right time, even though he has ceased to have it at the time of the aditio u. If, having been dare iussus, he is alienated sine peculio, there is no prohibition, until he actually is prevented from taking the money18.}

These texts create one serious difficulty. It is obvious that if a testator says: “ if S pays 10 to T, let him be free,” there is nothing in these words to give T any right. There is no duty in anyone to pay the money—there is no pact, no juristic relation between the heres (or 504 Condition: Payment of Money: Right to Payment [pt. ii the slave) and the man who is to receive it. The payment if made is a mortis causa capio, but as Gaius and Marcellus tell us it is not a donatio¹. Accordingly several texts tell us that the heres will do wisely to forbid the payment of the money, for thereby he will save it, and the slave will not lose his liberty^{[1854] [1855] [1856]. Some others use argument which involves the same conclusion[1857] [1858]. But these texts do not stand alone and there are puzzling conflicts. Pomponius tells us that if the slave pays it notwithstanding the heir’s prohibition, the receiver holds it only pro possessore and is bound to restore i [1859]t[1860]. On the other hand Paul gives us Julian’s opinion that even in this case he makes the receiver owner[1861]. Both these texts appear to be genuine: they shew a quite intelligible difference of opinion. The view of Pomponius rests on the rule that a slave cannot alienate peculium unless he is authorised to do so8. That of Julian and Javolenus rests on the fact that the pay­ment is authorised by the will under which the heres holds. But other texts go further. Where it is payable to one of coheredes, and another forbids the payment, Ulpian appears to tell us that the coheres will recover (in the actio familiae erdscundae), quod sua intererat prohibitum statuliberum non esse[1862], which is less than the whole amount, since some of it will come to him as heres if it is not paid under the condition. This text may be genuine : there may have been provisions under the will which would have made the payment essential to the carrying out of the testator’s whole intention8. Nevertheless the remark has rather the air of an afterthought, and may be Tribonian’s. One text goes further still: we are told, nominally by Ulpian’, that if the heres forbids the payment, the extraneus to whom it was to have been made, adversus heredem in factum actions agere potest, ut testatori pareatur. It is certain that this is from Tribonian. The expression in factum actions agere is no more than suspicious10, as is the remark that the payment would be testatori parere. But the conclusive fact is that the very same fragment in the immediately preceding sentence lays down the opposite rule, si tamen vult heres nummos salvos facers, potest eum vetare dare: sic enim fiet ut...nummi non peribunt. The fragment is ch. xxi] Manumission coupled with Institution 505 part of an extract many other parts of which have been convicted of interpolation1.}

The fact that a slave is instituted as well as freed is not a modality, but the treatment of the matter in the same chapter is perhaps justified by the fact that the one gift depends very much on the other, and questions arise as to how far modalities affecting one are to be applied to the other.

The general principle is that a slave so freed and instituted is a necessarius heres, i.e. he is heres without entry, and has no ius absti- nendi'^{[1863] [1864]. We have already seen that in classical times, a gift of the hereditas did not imply a gift of liberty. So strictly were such impli­cations excluded that if a slave was freed, whether instituted or not, and was elsewhere substituted, it was necessary, in the opinion of some jurists, to repeat the gift of freedom, the first gift being bound up with the institution and failing if it failed[1865]. It is essential that the man belong to the testator[1866]. He must be the testator’s at the time of the death, so that if he is freed or sold inter vivos, he is not a necessarius, but enters for himself or his master as the case may be[1867], the gift of liberty in such a case being a mere nullity, just as an institution of your own slave without a gift of liberty is[1868]. If at the time of the death the owner is without testamenti factio, the whole thing is of course void[1869] [1870]. But a slave given to the wife mortis causa is still the husband’s and if instituted with a gift of liberty in the will, he is a necessarius heres*. He must have been the testator’s at the time of the will[1871], though part ownership at that time is enough[1872] [1873]. Where a man gives liberty and hereditas to a servus alienus, and then buys him, both gifts are bad, for the liberty to servus extraneus is a nullity, and the institution cannot stand without itn. On the other hand, if he was the property of the testator at the time when the will was made and at the death, he is a necessarius: the fact that he has been sold and rebought in the interval is not material12. The common form of his institution is Stichus liber et heres esto, but any imperative form suffices : equivalents are »3. liber esto: si liber erit, heres esto, and »3. liber esto, et postea quam liber erit heres esto13.}

506 Manumission with Institution: Error [pt. n

Cases of error in this matter are scantily dealt with. If I institute and free a serous alienus, supposing him mine, his owner takes¹. If a miles institutes his slave, thinking him free, the institution is void as there is no gift of liberty, and no doubt the rule is the same in the case of a paganus*. If it was a serous alienus, it would seem from the compromise laid down in the case of Parthenius, that the institution stands good if there is no substitution. If there is a substitution then, on a reasonable though hardly logical compromise, the substitute takes half in any case’.

Where a slave is freed and instituted ex parte, he is free and necessarius heres before the other heres enters: he is said to derive his liberty from himself and not from his coheres, on whose entry indeed it does not in the least depend⁴. This has noticeable results. Thus, according to Julian, if a slave is a necessarius heres it is not possible to adeem the gift of liberty in a codicil, for just as a legacy to the heres is void, so is the ademption of a gift⁵. The point of the argument is that a gift taken from any beneficiary vests in the heres, and as he is the heres it is in this case a nullity. It may be said that for all purposes but fideicommissa the codicil requires the exist­ence of a will, and the will would fail if effect were given to this codicil⁶.

Necessarii heredes are heredes without their own consent: there is no question either of entry or abstention. In case of insolvency the goods are sold in their name, and the resulting infamia attaches, on the view which prevailed, to the slave personally, though Sabinus was of a different opinion⁷. Such a heres has the beneficium separaiionis, i.e., if he is careful not to deal with the goods of the testator, his own after-acquired property will not be liable to seizure by the creditors. If, therefore, the goods have been sold up once, there is no danger of any further proceedings unless the heres makes some further acquisition ex hereditate⁶.

The necessarius is not necessarily the institutus in first instance : he may be a substitute, or substituted to a substitute, or even a pupillary substitute, in which case he is necessarius heres to thepupillus*. This rule is accepted utilitatis causaℱ, at least as to pupillary substitution. The point is that the testator is making him necessarius heres to someone

¹ 46. 1. 33. »29.1.13.3. » 28. 5. 41, 42; In. 2. IS. 4.

⁴ 29. 2. 58; 40. 7. 2. 3. Paul points out that where the gift to him is made to be dependent on entry of another heres (cum mihi quis heres erit 8 liber et heres esto) this independence does not exist, 29. 2. 58.

⁶ 28. 5. 6. 4. ⁶ In. 2. 25. 2.

⁷ G. 2. 152—4; In. 2.14. pr._t 1. C. Th. 2. 19. 3 (332) observes that they are instituted quite as much to get the infamia as to get the hereditas.

⁸ 42. 6.1. 18; G. 2. 155; In. 2. 19. 1. A minor slave so instituted and meddling with the property, can get restitutio in integrum, 4. 4.« 7. 5. This title gives details as to the separatio.

» 28. 6.10. 1, 36.pr,\ 40. 7. 2. 4.

¹⁰ 40. 7. 2. 4, 36. Plus and Severus may have dealt with the matter, 4. 4. 7. 10.

to whom he certainly did not belong at the time when the will was made. On the same notion of utility depend also the rules that if a man institutes a young slave as necessarius, and substitutes another to him, this second slave, even a postumus, will be necessarius heres to the first¹, and also that a slave made a pupillary substitute is in effect a statuliber; a rule laid down by Celsus, and justified by Papinian, on the ground that the rule has the effect that if the heir sells him he is sold cum sua causa, while any other rule would have enabled the filius, or rather his tutor, to upset the father’s intentions². Strictly he cannot be a statu­liber as he acquires his liberty from himself⁸. Where a man made his impubes son his heres, and gave his slave liberty and then made the slave pupillary substitute without a fresh gift of liberty it was doubtful whether this could make him a necessarius, as the liberty and the insti­tution were in different grades. Justinian of course provides that the gifts are valid and make him a necessarius heres*. If the substitution of a slave takes effect, and the slave becomes free, his liberty being irrevocable, he remains free even though the heres is restitutus in integrum*. In like manner it seems from some obscure provisions that as a slave is instituted in order that he shall bear any resulting infamia, he is an inf amis, and thus brothers or sisters can bring the querela against him. The will may thus be upset, but the slave retains his liberty’.

A serous proprius instituted with liberty is thus always a necessarius heres, but it is only in case of insolvency that the most important point arises. The lea: Aelia Sentia allows institution of necessarii heredes even in fraud of creditors, partly no doubt on account of the extreme dislike of intestacy, but more in order that the infamia attaching to insolvency shall fall on the slave and not on the memory of the dead man. But as one is enough for this purpose, only one is allowed, and thus if two are named, only the first is free⁷. Where A was instituted and two slaves with direct liberty were given a fideicommisswm of the hereditas, the testator proved insolvent. The heres refused the inheritance and was compelled to enter on the principle of the sc. Pegasianum. He handed over the whole hereditas, but only the first of the two slaves was entitled⁸. So if a slave was instituted, and another substituted to whom the testator owed fideicommissary liberty, Neratius held that if the testator was insolvent, the second was heres, since his manumission

I 28. 6. 10. 1. » 40. 7. 2. 4, 36. ⁸ 40. 7. 2. 3.

â€? C. 6. 27. 4.. » 4. 4. 7. 10; cp. In. 3. 11. 6.

⁶ If the goods were not sold he ought not to be infamis: if they were, the brother gains nothing. The case is no doubt that of a testator who has mistaken his own financial position. Justinian abolished the rule, C. Th. 2. 19. 3; C. 3. 28. 7.

⁷ Ulp. 1.14.

⁸ 28. 5. 84. 1, port, p. 509. An insolvent instituted “the two Apollonii.’’ If one died the survivor took: iTboth lived the gift was void, for only one could take, as the testator was insolvent, 28. 5. 43, 44. would not be in fraudem creditorum¹. This is carried still further. If the substitute to the slave was a free man or one entitled to freedom, he must be asked first, for it is a fraud on creditors to allow the slave to be free if there is a free man willing to accept the inheritance’. A curious case is given in which there may be two heredes in such a case. A slave is instituted and the testator then says: T heres esto si S heres fuerit. The testator is insolvent. S is heres necessarius. T can now take, S is heres still, because semel heres semper heres, and of course T’s claim does not in any way prejudice the creditors’.

It may be noted that one who is barred from liberty by any enact­ment other than the lex Aelia Sentia cannot be a heres necessarius. The provision of this lex frees him from the restrictions created by the lex itself, but not from any other^[1874]. Also, the institution of a heres necessarius frees from the creditors none of the property except himself. Where, not knowing that the estate is insolvent, he pays certain legacies, these are recoverable by utilis actio under the edict for revocation of acts done in fraud of creditors^{[1875] [1876].}

On the other hand he is not a necessarius unless he actually gets his liberty by the will. Thus where a slave is freed under conditions, and before these are satisfied, is given liberty by the Praetor for de­tecting his master’s murder, he is not a heres necessarius, but on satisfying the condition he can take the inheritance if he wishes^[1877].

It is enough that he belong to the testator. A slave is given liberty by fideicommissum under a condition. The heres institutes him and dies before the condition is satisfied. He becomes heres necessarius to this testator. But if the condition on the other gift occurs, he will cease to be necessarius, not, we are told, that he will cease to be heres, sed ut ius in eo mutetur successionis^[1878]. A person to whom fideicommissary liberty is due is a quasi statuliber^{[1879] [1880] [1881], and the heres cannot make his position worse. Thus his position as a necessarius heres must depend on the non-arrival of the condition[1882]. Where a slave, S, is instituted and freed, si meus erit cum morior, the words are not mere surplusage, though the gift of liberty would fail in any case if S were alienated vivo testatore, since it requires ownership at the time when the will operates.}

ch. xxi] Necessarius Heres: Exceptional Cages 509

But apart from them he would acquire the hereditas to the alienee¹. They operate as a sort of condition. The text goes on to consider what will happen if he has been freed inter vivos. He cannot be a necessarius, but he can take the hereditas, since he satisfies the terms of the gift: he is mens, not servus, but libertus². If he was freed si mens erit and instituted pure, he can, if alienated, take iussu domini. Here too the text points out that words which so far as their primary purpose goes lay down only what the law enacts may nevertheless incidentally change the effect of the gift³.

It is in general essential that institution and liberty be in eodem gradu, and, a fortiori, that both be direct gifts. But there are relaxa­tions of which the limits are not clear: perhaps it is useless to seek for a principle. The relief is greatest in the case of a miles. A soldier institutes X and gives S liberty and a fdeicommissum hereditatis. X dies without making aditio. Ulpian tells us that Severus and Caracalla construed this as a direct gift to S⁴. Maecianus considers whether this applies to pagani, and decides, or is made by the compilers to decide, that it applies only if the testator did not know of the death⁶. Where the heres does not die but refuses, the risk of infamia makes the need of relief more urgent. Accordingly Gaius holds that the same relief is given here : he treats it as a direct gift to S, ex sententia legis (Aeliae Sentiae), i.e. of the clauses as to fraud of creditors and necessarii heredes. He remarks⁶ that, on the facts, the estate being insolvent, and S not a necessarius heres, X cannot be made to enter, and if he does enter S cannot be free or take a transfer. But in another case in which the facts are the same so far as the present point is concerned, Scaevola says⁷ that a senatusconsultum of Hadrian’s time provides that S can compel X to enter, whether the gift of liberty is direct or only fideicommissary⁸. In the actual case there are two su,ch slaves, of whom only one can take, but that does not seem material. The solution of Gaius evades the difficulty by a forced construction: that of Scaevola involves a new definition of necessarius heres. Another text goes further. Even where the fdeicommissum is conditional, the slave, freed pure, compels the heres to enter, says Marcian, and if the condition fails, his freedom will stand good⁹.

In one case the slave has a gift of liberty, and a fdeicommissum of the hereditas. He compels the heres to enter. Then the slave, now free, dies before he has in any way delayed to take over the hereditas, leaving T his heres. T refuses to take the hereditas. Marcellus observes that the senatusconsultum (Trebellianum) deals only with the manumissus

1 Ante, p. 137. ² 28. 5. 52. 1. ² Cp. 35. 1. 47.

⁴ 29. 1.13. 4. ⁶ h. t. 14. The language is that of a legislator.

⁶ 36. 1. 65. 15. ⁷ 28. 5. 84. 1. Or Tribonian as to tideicommissum.

8 JW,p. 523. 9 36. !. &2.pr.

510 Necessarius Heres: Exceptional Cases [pt. n and not with his heres, but concludes on the whole that the heres cannot refuse what the manumissus would have been bound to take. He adds that if the slave had died without successor before the estate was handed over, the creditors would have had the right to seize the goods as if there had been restitutio hereditatis¹.

Substitutions gave rise to some rather complex questions. It is hardly possible to deal with them systematically, for they represent a series of “ hard cases,” in which favor libertatis and the desire to save a will, and to secure a successor to an insolvent, led to distorted views of principle.

A father substitutes to his impubes son the slave S, with liberty. The impubes sells him to T. T, having already made a will, makes another in which S is made free and heres. This will upsets his first, since it is validly made and there may be a heres under it. But so long as S can be heres to the impubes, he cannot be liber and heres under the will of T. If the impubes matures, S will be heres necessarius to T. If the impubes dies under age, he will be heres necessarius to the impubes, though of course there is nothing in that to prevent his being heres voluntarius to T^s. The object is, as the texts say, to save the necessarius to the father’s will, and the principle applied is that the slave is a kind of statuliber, and is thus alienated cum sua causa, i.e. subject to his becoming necessarius heres of the impubes, though at the time of the death of the latter the slave is in other ownership³.

In a very long, very obscure, and in some parts, corrupt text, a will is considered which ran: T heres esto; S Maevio do lego; S heres esto; si S heres non erit, S liber heresque esto. It is impossible to be sure of the meaning of the words, which have already been considered from another point of view⁴. The first point is: under what circumstances can a man be substituted to himself? It is held that there is no substitution here: there is one institution with a gift of freedom, the whole dependent on the failure of T. The legacy to Maevius is void⁶. It is an attempt to interpret hopelessly obscure words⁸.

In the cases in which the institution or the liberty, or both, are subjected to modalities of various kinds, there is a strongly marked tendency to such a construction as will preserve the status, if it may be so called, of the necessarius heres and to secure that he shall not get the liberty without the hereditas.

If a slave is instituted pure, and freed ex die, the institution is valid,

¹ 36. 1. 46. pr. The text is obscure and seems to contain a truncated discussion of the possible effect oi delay on the part of the heres.

² 28. 5. 55 J 28. 6. 48. 2. 8 Cp. 40. 7. 2. 3. < Ante, p. 449.

8 28. 6. 48. 1. 6 Cp. 40. 4. 10. 1; 40. 5. 50. ch. xxi] Manumission with Institution: Modalities 511

being deferred until the day named. When that day comes, if there has been no alienation, he will become free and heres necessarius, and if he has been alienated or freed, he can at once take the hereditas for his master or himself¹. The difficulty thus avoided by reading the dies into the institution is that if this is not done, the institution must necessarily fail, since at the time the will is opened he cannot take the hereditas, as he cannot be free. In the next following text the matter is carried still further. If the slave himself is not alienated, but the usufruct of him is, he is still the property of the testator. But he cannot be free, during the usufruct, at least in classical law, and accordingly the institution is postponed to the expiration of the usufruct, when he will be necessarius².

Where a servus proprius is instituted pure, and given liberty con­ditionally, the same difficulty is evaded in the same way; the institution is deferred till he is entitled to freedom, when he becomes heres neces­sarius². If while the condition is still pending, the testator sells him, the effect is to destroy the gift of liberty, and he can therefore enter at the command of his new master⁴. But if the testator alienates him after the condition has failed, he cannot enter at the command of the buyer, quia eo tempore ad eum pervenisset quo iam exstincta institutio inutilis fuerat². All this is an artificial construction. In order to save the institution the condition on the liberty is read into it, and as it is read in for one purpose on the assumption that the testator meant it to be there, there is nothing to be done, but to read it in for all purposes.

If the man is freed pure and instituted conditionally, there is also reason for reading the condition into both gifts. Unless it is satisfied when the will operates, the man will not be a heres necessarius. Both gifts therefore await the condition, and if it occurs he will be liber and heres necessarius. But what if the condition does not occur ? Here, favore libertatis, logic is disregarded and he gets his liberty. Ulpian states this generally, but Julian is more guarded—habetur ac si libertas sine hereditate data fuerit: unless there is another heir, the gift must fail⁶. If a slave is freed pure and instituted under a condition, and to have a legacy if he is not heres, Marcian cites Pius as saying that the legacy is subject to the same condition⁷. This is puzzling, but Marcian’s source is Papinian, whose text⁸ shews that Pius meant the condition of liberty, not the other.

1 28. 5. 9. 17—19. This is very like dies in an institutio.

* 28. 5. 9. 20. Quite apart from this point a condition on one gift might be read into

another as a matter of construction. Where a jideicommissuni of liberty was given to a servus alienus and he was also substituted, the condition of liberty was read into uie substitution, 31. 83. s 28. 5. 3. 1.

* h. t. 38. 2. So if he were a common slave, h. t. 7 (Mommsen). ⁵ A. t. 38. 3.

8 40. 4.14; 28. 5.21,22. It might be effective as bjideicommissum on the heres ab intestate, but the texts treat it as a direct gift.

1 28. 7. 18. pr. 8 35. 1. 77._pr.

If the slave is freed ex die and instituted conditionally, it follows from what has been said that if the condition is satisfied before the day, the institution will take effect on the day, being also subject to dies. If the condition is fulfilled only after the day, we are expressly told that he is free and heres only from the day when the condition arrives¹.

Of the possible case of liberty under one condition and institution under another we hear little. We can, however, infer from one text that though logic requires, on the principles we have stated, that each gift should be subject to both conditions, the view of Julian was accepted, that if the condition on the institution fails he may still get his liberty, being regarded as an ordinary statuliber, the gift of hereditas being ignored². This is a simple case of favor libertatis*.

One case is rather puzzling. S is instituted pure and given freedom if he pays 10 before a certain date. In a codicil there is an unconditional gift of liberty. He will not be free or heir before the date, unless he pays the 10, but if at that date he has not given the 10 he will be free by the codicil⁴. The principle appears to be this: as the testator has given liberty and inheritance by the will, S cannot have the latter without the former, so far as the will is concerned, and therefore the condition is read into the institution. The codicil cannot alter that: an institution cannot be varied by codicil. But the institution being thus conditional, the gift of liberty must also, even in the codicil, have the condition read into it: the codicil is treated as if it were in the will’. Thus he cannot get his liberty without satisfying the condition. But, when it is no longer possible to give effect to the institution, Julian allows favor libertatis to have play, and the gift of liberty has its effect as if there had been no condition on it.

Where liberty is given to S directly but ex die, and there is a fideicommissum hereditatis in his favour, as he is not ipso facto free by the entry he has no locus standi to make the heres enter and hand over the hereditas, nor is there any evidence of the testator’s wishing him to be necessarius heres*.

It should be added that if the inheritance is conditional, e.g. on the payment of money, and no time is fixed for satisfaction of the condition, the creditors may apply to have one fixed, and if he does not pay the money within this time, they may proceed as if he was not instituted. But this affects only his right to any bona, not probably a very serious matter in such a case. It in no way affects his right to be free when­ever he satisfies the condition. Strictly he would still be heres: the rule is a purely praetorian one, affecting nothing but the bona⁷.

1 28. 5. 9. 18. ² 40. 7. 2. 3.

³ 28. 5. 21. pr. gives another case of favourable construction. A testator says: quisquis mihi heres erit, 8 liber et heres esto. Strictly this is a condition, and S ought to take nothing until there is another heres. It is allowed effect as if unconditional. ⁴ 28. 5. 38. 1.

6 Machelard, Regie Catonienne, 40. ⁶ 36. 1. 32. 1, 57.1. ¹ 42. 5. 4.