CHAPTER IV. THE SLAVE AS MAN. NON-COMMERCIAL RELATIONS.

In this case there was an entrance fee, (which included a bottle of good wine,) and a monthly subscription. Breaches of the statutes were penalised by fine, and in some cases by exclusion from the benefits. The members’ rights were, mainly, to share in periodical banquets, and to the provision, out of the property of the college, of a fixed sum for the expenses of burial, out of which sum a certain proportion was dis­tributed among the members present on the occasion of the funeral. There was a provision that if a slave member was freed he was to give the society a bottle of wine. If the deceased member had left no directions, the funeral was carried out by the officers of the society, but it was open to him to give directions as to the person who was to do it. The forms of the society were ordinarily modelled on those of a town. Thus the members were the populus, the directions were regarded as a will, and the person charged was looked on as a heres, in which character he took any part of the fixed sum which was not needed for the funeral. In the absence of any such claim, it seems to have remained with the society⁸. In the society at Lanuvium there was a rule, (and it may have been general,) that if the dominus would not hand over the slave’s body for burial, and the man had left no tabellae, the rites were gone through without the body ; funus imaginarium fiet*. The statutes of this collegium contain a provision that no creditor, patron or dominus is to have any claim on the funds of the college except as the heres of a member. This no doubt refers to the dis­positions just mentioned, and seems to imply that the dominus could claim nothing, unless so made heir, and that if another person were named, it would go to him.

Of protection to the morality of slaves there is in early law little or no trace. Probably it was not needed. But in the Empire, when it certainly was needed, it was slow to develop. We have already seen⁶ that from the time of Domitian onwards there was legislation limiting the power of the dominus in the direction of protection of personal chastity. But it did not go very far. Not till a.d. 428 was it made penal for lenones to employ their slaves in prostitution, and Justinian confirmed this¹. We have seen that the classical law regarded sale to a leno as a reasonable mode of punishment². Debauching a man’s ancilla was an iniuria to him, and might be furtum³, but the injured woman does not seem to have been considered. The rules already discussed as to the effect of sale with a condition against prostitution⁴ date from classical times, and do actually regard the woman herself, since the restriction could not be remitted, but the protection depends on the initial goodwill of the owner⁵. Rape of an ancilla aliena was made a capital offence by Justinian, but it did not involve forfeiture, as that of a freewoman did⁶. There is no penalty for seduction by the dominus. It is clear that, throughout, the morality of a slave woman was much less regarded than that of a freewoman⁷.

Far more important in law and more fully recorded is the gradual recognition of servile cognation. In no other branch of law is the distinction so marked as here, between the rules of law and the practice of every day life. It is well-known, on the evidence of memorial in­scriptions and lay literature⁸, that slaves lived together in permanent union as man and wife, and were regarded, and regarded themselves, as married, and as sharing all the ordinary family relations.

But the law takes a very different view. In law, slaves were incapable of marriage⁹: any connexion between them, or between slave and free could be no more than contubemium¹⁰, and thus enslave­ment of either party to a marriage ended it¹¹. Accordingly, infidelity between slaves could not be adultery¹², and though a slave could be guilty of adultery with a married freewoman¹³, it was not possible for an ancilla to commit the offence, or for it to be committed with her¹⁴. Nevertheless the names of legal relationship were freely applied to the parties to, and issue of, such connexions: we hear of uxor, pater, filius, frater, and so forth, even in legal texts¹⁵, but Paul warns us that though these names, and the expression “ cognation,” are used, they are without

^I C. Th. 15. 8.2; C. 11. 41. 6.. ² Ante, p. 37.

⁸ 47. 10. 9. 4; 47. 2. 88. 2. It might give rise to actio Aquilia and even servi corrupti, and both would lie in the same case, 48. 5. 6. pr.

⁴ Ante, p. 70. ⁸ Ibid.

⁶ C. 9. 13. le. In earlier law it was dealt with as vis, 48. 5. 30. 9.

’ 47. 10. 15. 15.

8 Wallon, op. cit. 2. 180; Marquardt, Vie Priv4e, 1. 205; Erman, Servus Vicarius, 442 sqq.

9 Ulp. 5. 5. w p. 2. 19. 6 ; C. 9. 9. 23.pr.

^II 23. 2. 45. 6; C. 5. 16. 27. As to Captivi, post, Ch. xm. ¹² C. 9. 9. 23. pr.

19 48. 2. 5; 48. 5. 34; C. 9. 9. 24.

14 48. 5. 6. pr.; C. Th. 9. 7. 1; C. 9. 9. 28. Adultery was essentially interference with a wife’s chastity. Similarly corruption of an ancilla though called stuprum was not punishable as such, P. 2. 26. 16; C. 9. 9. 24.

15 P. 3. 6. 38; D. 32. 41. 2; 33. 7. 12. 7, 12. 33; C. Th. 4. 6. 3 « C. 5. 27. 1; Nov. Marc. 4. 1. significance for the law of succession¹. So Ulpian tells us that the rules of cognatic succession apply to non-servile cognation, nec enim facile ulla servilis videtur esse cognatio^{[311] [312]}. Diocletian says shortly that servus suc­cessores habere non potest, and applies the principle in two cases^[313]. So, even in late law, the title on legitimation makes it clear that an ancilla could not be a concubina for this purpose^{[314] [315] [316]}. This is an enactment of Constantine, who had already made it severely punishable for decuriones to cohabit in any way with ancillae²: it was important that decuriones should have legitimate successors on whom the civic burden should descend. Both enactments were adopted by Justinian. Apart from this, cohabitation with slave women was not in any way punishable³.

Even at law, however, these connexions between slaves were not a mere nullity. So long as all parties were slaves there was of course no great room for recognition, though it went some way; much further in­deed than seems to have been the case in other systems^[317]. In a legacy of fundus cum instrumento or fundus instructus, the slaves who worked it were included unless there was some special indication that the testator did not so intend^[318]. Paul tells us that it must be understood to include the uxores of such slaves^[319], and Ulpian lays down the same rule for wives and children on the ground that the testator cannot be supposed to have intended such a cruel separation^{[320] [321]. It must be noted that all this turns on presumed intent. There was nothing to prevent the legacy of a single slave away from his connexions. Thus, where a business manager employed in town was legated, Paul saw no reason to suppose that the testator meant the legacy to include his wife and children11. And where a certain ancilla was left to a daughter, to be given to her on her marriage, Scaevola was clear that this did not entitle the legatee to claim a child, born to the ancilla before the marriage took place on which the gift was conditional12.}

There were, however, cases which had nothing to do with intent. Thus it can hardly be doubted that the rules we have already stated, according to which the issue of an ancilla do not belong as fruits to the bonae fidei possessor, or to the usufructuary, or, in the case of dotal slaves, to the vir, are largely based on recognition of the claims of nature¹. So, too, it was laid down by Constantine that in indicium familiae ercis- cundae, or communi dividundo, the slaves were to be so distributed that those related as parent and child, or brother and sister, or husband and wife were to be kept together². It is noticeable that in nearly all these cases, the recognition extends to the tie of marriage as well as to that of blood. So, too, in the actio redhibitoria we have seen that if several were sold together who were related as parent or child or brother, they could be redhibited only together’.

The same recognition is brought out in a very different connexion by Venuleius. He tells us that though the lex Pompeia de parriddiis applies on its terms to lawful relationships only, yet, cum natura com­munis est, similiter animadvertetur, in the case of slaves⁴.

When the slave becomes free the question of the importance which the law will attach to these previous relations becomes more important. It should be noted that there are two distinct questions: how far do they restrict the man’s liberty of action ? How far can they create rights ?

Restrictively the recognition was fairly complete. Labeo held, in opposition to Servius, that the rule forbidding in ius vocatio of a father, without leave of the Praetor, applied to fathers who were slaves at the time of.the birth⁵. We are told by Paul that servile relationship was a bar to marriage—the cases mentioned being child, sister, and sister’s child, and, though the parentage were doubtful, the rule applied on the father’s side as well as on the mother’s⁸.

So far as giving rights is concerned, the classical law went no further than, in construing wills, to extend such words as filius to children born in slavery. The earliest case is that of a man who, having no son, but one who was born a slave, instituted him heir, (he having been freed,) and then said, “ If I have no son who reaches full age, let D be free.” Labeo took the strict view, that D was free. Trebatius held, and Javolenus accepted the view, that in such a case the intent being clear, the word filius must be held to denote this son⁷. Scaevola and Tryphoninus lay

¹ Ante, pp. 21 sq.

² C. Th. 2. *25. 1; C. 3. 38. 11. The text speaks of agnatio, but not of course in a technical sense. So also 33. 5. 21.

⁸ 21. 1. 35, 39. Here too the rule is applied to contuberndles: the form of the text suggests compiler's work, though the rule itself would not be out of place in classical law. A slave concubine and her child were not to be seized in bonorum venditio. P. 1.13a. Ig; D. 42.5.38.pr.

4 48. 2. 12. 4. References to cases in which slaves were allowed a de facto power of testation within the familia are not illustrations of the present point: such wills had no legal force. See Marquardt, Vie Privee, 1. 222.

5 2. 4. 4. 3. Severus in the text, but Servius seems more probable. See however, Roby, Introd, to Digest, clviii.

6 23. 2. 14. 2. As to affinitas Paul is less clear. He says that in so doubtful a case it is best to abstain: this must be taken as law in the time of Justinian, 23. 2. 14. 3.

7 28. 8. 11.

ch.iv] Cognatio Servilis: Iniuria to Slaves 79 down a similar rule in a case of which the facts are rather complex, but of which the gist, for our purpose, is that in construing wills filius includes such children: it does not seem to be thought material that there should be no other children¹.

Justinian took a more decided step. He observes that the rules of proximity in bonorum possessio do not apply to servile relationships, but that he, in adjusting the hitherto confused law of patronal relations, provides that if a slave has children by a slave or freewoman, or an ancilla has children by a slave or freeman, and he or she is or becomes free, the children shall succeed to the parents and to each other and to other children of the whole or half blood with themselves“. The enact­ment here referred to is in the Code, more shortly expressed, in the form that children are to exclude the patron whether freed before or after or with the father, or born after his manumission^{[322] [323] [324].}

Later on, while preserving the rule that slavery and marriage are incompatible^{[325] [326] [327], he allows, by a series of Novels, a right of legitimation of children of a freeman by a slave, if he had freed her and them, and obtained for them the ius regeneration's11. Most of these provisions deal with oblatio curiae, and are part of the machinery for keeping the lists of decuriones full.}

The fact that an actio iniuriarum may lie on account of insult to a slave is, again, a recognition of his human character. The matter pre­sents some difficulties: the chief point to note is that though the action is necessarily acquired to the dominus, it is brought sometimes for the insult intended to the dominus, sometimes without reference thereto: it may be either suo nomine or servi nomine“.

The Edict contained a provision that for verberatio contra bonos mores or for subjecting the man to quaestio, without the owner’s con­sent, an action would lie in any case^{[328] [329]. Even a municipal magistrate}

might be liable if the flogging were excessive¹. But any reasonable beating, corrigendi vel emendandi animo, was not contra bonos mores, and so was not within the Edict’. Intention to insult the owner was not needed: it is incorrect to say that it was presumed: it was not required. The action lay servi nomine*. It seems probable, however, that intention to insult the dominus might be alleged in the formula, and proved^{[330] [331] [332] [333] [334], with a view to increased damages.}

There is more difficulty as soon as we pass to less definite forms of iniuria. The Edict continues : Si quid aliud factum esse dicetur causa cognita indicium dabo*, a provision which besides covering all other kinds of insult appears to include the contrivance of verberatio by a third person^[335]. The system of rules of which this text is the origin is not easily to be made out. The texts give indications of conflict of opinion, but the matter may be simplified by striking out two classes of case in which a slave is concerned in an iniuria, but which are governed by principles independent of our present question. These are:

(1) Cases in which an insult is committed to the slave, but is actually expressed to be in contumeliam domini. Here the slave is merely the medium through which the wrong is done: the master’s action is suo nomine, governed by the ordinary law of iniuria.

(2) Cases in which the iniuria does not take the form of an “ insult,” in the ordinary sense, but is a wilful infringement of right^[336]. The wanton disregard of a man’s proprietary and other rights is a form of iniuria too well known to need illustration. Such wanton wrongs might be committed in relation to a slave. But they have no relation to our problem, even where the wrong done was one which could not be done except to a human being.

The question remains: under what circumstances, apart from the Edict as to Verberatio, did an action lie for an insult to a slave, and was it in any way material that there should be intention to insult the dominus'¹. It is clear that, if intention to insult the dominus was present, the action was suo nomine and not servi nomine, the latter action being available if there was no such intention. This is expressly stated in one text⁸, and appears from others, which, comparing the case in which the person insulted is a slave with that in which he is a liber homo bona fide serviens, state that if there was no intention to insult the dominus, he has no action at all in the latter case and none suo nomine in the other¹.

It was not, however, every insult to a slave which gave an actio iniuriarum, servi nomine. It must be something serious; not a mere levis percussio or levis maledictio, but real defamation or serious insult². This restriction is part of the application of the words, causa cognita, in the Edict³. The quality of the slave would affect the question: to a common sort of slave or to one of bad fame, or careless, a greater insult would be needed to cause the Praetor to grant a formula, and perhaps it would be altogether refused (except in case of verberatio, etc.), where the slave was of a very low order. The matter was wholly in the hands of the Praetor⁴.

If intention to insult the dominus were alleged, the words A". A* infamandi causa were inserted in the formula’’. It does not appear that any other legal result necessarily followed. The texts dealing with the matter seem to shew that no action lay suo or servi nomine, for iniuria to a slave, apart from the edict as to verberatio, unless the insult were atrox of a serious kind⁶. There are, however, some remarks to make on this.

(i) The granting of the formula being left by the Edict to the discretion of the Praetor, it is unlikely that iniuria atrox had in this connexion, (if anywhere,) a very precise meaning. On the other hand it is likely that where intention to insult the dominus was alleged in the intentio, and so made a condition of the condemnatio, the formula would be issued more readily than where no such intention was alleged, and damages would be on a higher scale.

(ii) The fact that an insult expressed to be an insult to the master need not be atrox, while one so intended, but not so expressed, must be, to give an action, is not surprising. In the former case the tendency to lessen the respect in which the insulted person is held appears directly from the facts: there can be no difference between different words but one of degree, sufficiently represented in the amount of damages awarded. In the latter case the difference may be one of kind. Contumelious treatment of a trusted steward may well have a defamatory effect on his master, and, if it be shewn to have been done with the intent of insulting him, will give rise to an action suo not servi nomine. But an abusive epithet thrown at a humble slave cannot really affect the respect in which his master is held, no

1 47.10. 15. 45. Si pro libero se gerentem...non caesurus eum si meum scisset, non posse eum quasi miki iniuriam fecerit sic conveniri Mela scribit. See also h. I. 48—if the slave is oonae fidei. possessed, and there was no intent to insult any but the slave, the dominus has an actio servi nomine.

2 47. 10. 15. 44. s h. I. 43. See also C. Th. 12. 1. 39; C. 9. 35. 1, 8.

4 41. 10. 15. 44. The Inst, say minuitur in the case of these inferior slaves. In. 4. 4. 7.

5 Lenel, Ed. Perp. § 194. ⁶ G. 3. 222; In. 4. 4. 3; C. 9. 35. 8. matter what may have been the intention of the speaker. It follow's that the master will have no action suo or servi nomine.

(iii) These considerations explain why the master had an action servi nomine when there was no intention to insult him, and why it was limited to the case of atrox iniuria. There can be no ordinary actio iniuriarum under the general Edict, because there was no inten­tion to insult¹. But under the large words of the special Edict there was plainly a power to give an action in the case, not so much on the general principles of the actio iniuriarum, as on the ground that injury is in fact caused to the plaintiff’s reputation, and justice requires that compensation be given for the harm done. There is no sign that the action was in any practical sense a recognition of the slave as having a reputation to lose: it is the damage to the master that is considered^{[337] [338]. The case is different with verberatio and quaestio: there, at least in the opinion of the later jurists, the feelings of the slave himself are con­sidered[339]. The difference in conception is probably an accidental result of the fact that under the special Edict the action was not given as a matter of course: causa cognita iudicium dabo.}

(iv) The action servi nomine was the last to develop. The Edict does not distinguish it. Gaius shews no knowledge of two types of action resulting from insult to a slave^{[340] [341]. All the texts which expressly mention it are from Ulpian[342] [343]. It has all the marks of a purely juristic creation6.}

Of the slave’s civil position it may almost be said that he had none. In commerce he figures largely, partly on account of the peculium, and partly on account of his employment, as servant or agent. His capacity here is almost purely derivative, and the texts speak of him as un­qualified in nearly every branch of law. They go indeed beyond the mark. General propositions are laid down expressing his nullity and incapacity in ways that are misleading unless certain correctives are borne in mind. We are told that he could have no bona, but the text itself reminds us that he could have peculium''. The liability of slaves on their delicts was recognised at civil law^[344]. But we are told that they could not be bound by contract, (in personam servilem nulla cadit obligatio,)¹ and that they could be neither creditors nor debtors: if expressions contradictory of this are found (and they are common), the legal reference is to the dominus^{[345] [346] [347]. But this ignores the fact that they were capable of natural right and obligation and the true rule is expressed in a text which says ex contractu civiliter non obligantur, sed naturaliter obligant et obligantur3.}

The exclusion of slaves from a number of actus legitimi seems to rest rather on the absence of civitas than on their slavery. Thus a slave could not be witness or libripens in mancipation: such a person must be a ciot’s^{[348] [349]. He could not be or have a tutor[350]. He could make no will, and if he became free a will made in slavery was still void. We are, however, told that slaves have testamenti factio6. This means that they may be instituted, either for their masters, or, with a gift of liberty, on their own account. But this employment of the expression testamenti factio puts the lawyers in some difficulty when they have to explain why a slave cannot witness a will. They put it down to his not having iuris civilis communionem in totum, nec praetoris quidem edicti[351]. This curiously guarded expression is no doubt due to the fact that the writer was face to face with the awkward fact that a slave could be heres. But illogical compromises of this kind are inherent in the Roman conception of slavery.}

In relation to procedure the incapacity of slaves is strongly accen­tuated^[352]. They could not be in any way concerned in civil proceedings, which must be, from beginning to end, in the name of the master^[353]. As they could neither sue nor be sued, they could not validly stipulate or promise, in the procedural contracts iudido sisti or iudicatum solvi, and so they could not bind a fideiussor by such a promise^{[354] [355]. Judgment against a slave was null and void : it gave rise to no actio iudicati de peculio, since it was not a negotium of the slave. In the same way absolution of a defendant, where the plaintiff was a slave, did not in any way bar his dominus'1. A slave’s pact, ne a se peteretur, was in strictness void, though it might give the dominus, if he were sued,}

84 Incapacity of Slaves in Procedure [pt. i an exceptio doli¹. On the other hand if the pact had been ne peteretur, or ne a domino peteretur, then, whether the original transaction had been the slave’s or not, the pact gave an exceptio pacti conventi^{[356] [357]. The distinction is not unmeaning: whether there had or had not been a pact was a question of fact: whether there had been dolus or not left more to the iudex. They could not interrogare in iure or be inter­rogated to any purpose[358] [359] [360]. As they could not be parties, so they could not sit in judgment. We are told that they could not act as indices, not, it is said, for lack of ability, but because, as in the case of women, monbus receptum est*. Similarly they could not be arbitrators : if a slave were appointed we are told that as a matter of convenience, if he became free before decision, the parties might agree to accept his decision. But this depends on his freedom, and is only a way of avoiding the trouble of a new appointment[361].}

There were other less obvious cases. Slaves could not be custodes ventris against supposititious children⁶, though they might accompany the person responsible. This is an express provision of the Edict: its reason is that such a custos is likely to be required to give evidence, and the evidence of slaves was not readily admitted. They could not opus novum nuntiare, their nuntiatio being a nullity. This seems to be due to the fact that the nuntiatio was a procedural act specially prescribed in the Edict as the first step in a process, aiming at an injunction^[362]. On the other hand, nuntiatio could be made to a slave. The receipt of the notice was no formal act: we are indeed told that it may be made to anyone, provided it be in re presenti operis so that the dominus may hear of it^{[363] [364] [365].}

There are some exceptions to this rule of exclusion, but they are only such as to throw the rule itself into relief, for the exceptional nature of the case is always either obvious, or expressly indicated. Thus though they could not be custodes ventris², yet, if a slave were instituted si nemo natus erit, he was allowed to take some of the formal precautions against supposititious children : the exception being expressly based on his potential freedom¹⁶. For similar reasons, though they could not have procurators in lawsuits, they might have adsertores

(and later procwraiores) in causae liber ales¹. One set of texts raises an apparent difficulty. A slave could offer, and take if it were offered to him, an extra-judicial oath, with the usual obligatory results, subject to some restrictions not here material². There was nothing exceptional in this. But the extra-judicial oath, being purely matter of agreement, could always be refused, and one to whom it was offered had not the right, which existed in case of the judicial oath, to offer it back again : iusiurandum quod ex conventione extra iudicium defertur referri non potest³. Another text says: si servus meus delato vel relato ei iure- iurando iuravit puto dandam mihi actionem vel exceptionem propter

conventionem*. The last words shew that the reference is to an extra­judicial oath: the word relato is explained by the fact that the rule against relatio in such cases means only that if it were offered back, the offeree need not take it. The case supposed in the text is that the slave has offered an oath : the offeree has returned it and the slave has then voluntarily taken it.

As incapable of taking part in procedure slaves could not be formal accusatores in criminal charges⁶. It is no doubt partly on account of this exclusion that Hadrian enacted that complaints by slaves of ill- treatment by their masters were not to be regarded as accusations⁴. But it was in general as open to them as it was to freemen to “ inform,” i.e. to make delationes to the fisc of cases in which property is claimable by the fisc, and also of criminal offences. Both kinds of information are called delatio, though in legal texts the term is more commonly applied to fiscal cases⁶, i.e. to notifications to the fisc of property to which it has a claim (such as bona vacantia), which someone is holding without right. The two classes may indeed overlap, since the right of the fisc may be due to the commission of a crime involving forfeiture. Informers were entitled to a reward, a fact which produced a class of professional delatores, the evil results compelling a number of enact­ments punishing false delations to the fisc, and, in some cases, true ones⁷. Delatio of crime was a form of blackmailing, which called for

¹ Post, Ch. xxvni. A slave could formally begin proceedings for a libellus to the Emperor on murder of his master. The case is exceptional, and moreover, the denouncing slave could claim his liberty, C. 1.19.1. Slaves could not appeal on behalf of absent master, but where possession was held on behalf of an absentee, and was invaded by force, the case being urgent, the judges were to hear even his slaves, C. Th. 4. 22.1 = C. 8. 5. 1; C. Th. 4.22. 4 (396). Slaves could apply for bonorum posscssio for the master, but this could be given without application, 37. 1. 7.

⁸ 12. 2. 20—23. Post, Ch. ix. » 12. 2. 17. pr. < 12. 2. 25.

⁶ This had advantages: one who accuses a will as falsum loses any gift, 34. 9. 5. A slave with a gift of liberty by fc. induced a hcres, on whom his liberty was not charged, to attack the will as falsum. He failed and lost his gift: the fc. was not affected, the slave not being an accusator, 48. 10. 24.

⁶ For cases of its application to criminal charges, see p. 86, n. 2. See however, Mommsen, Strafrecht, 879.

? General prohibitions, P. 5. 13. 1; False delations, 49. 14. 24, C. Th. 10. 10 passim, C. 10.11. 5; True delations, 34. 9. 1, C. Th. 10. 10 passim, C. 10. 11. 5, 7. Rein, Criminalrecht, 824; Mommsen, Strafrecht, 877 sgq.

[pt. I repression as early as A.D. 20but an information, if proved, does not seem to have been punishable in ordinary cases². But even for crimes slaves were forbidden to inform against their domini. It seems that Constantine allowed no exceptions, but ordered the slaves to be in all cases crucified unheard⁸. Several enactments toward the close of the century except maiestas*, and Justinian’s Code omits this prohibition in Constantine’s enactment⁵. And the Digest, laying down the general prohibition as to fiscal causes, and crediting it to Severus⁶, allows slaves to accuse their masters for maiestas, for suppressing wills giving them liberty, for frauds on the annona publica, for coining, regrating, and revenue offences⁷.

The capacity of slaves as witnesses requires fuller treatment. As a rule their evidence was not admissible in civil cases⁸. But the exclusion of such evidence, besides being a sort of self-denying ordinance, must have led to miscarriages of justice. Accordingly, convenience suggested a number of exceptions. Of these the most important is that they might give evidence in matters in which they were concerned—de suo facto—in the absence of other modes of proof, e.g. in case of transactions with them without witnesses⁹. We have no limitative enumeration of the cases in which their evidence was admitted¹⁰. Justinian adverts to a distinction drawn by earlier leges in the case of hereditas, according as the question is of the hereditas itself or of res in it, and provides that, whatever the form of the action, slaves shall be put to question only as to res corporates, and only those slaves who had charge of the thing, but in that case even if they had freedom by the will. Probably the older law allowed no “ examination ” of slaves given freedom unless the will was disputed, and then allowed it freely¹¹. A text in the Digest may be read as saying that slaves may be tortured in any res pecuniaria if the truth cannot otherwise be reached¹², but it probably means rather that it is not to be done in any res pecwniaria if the truth can otherwise be reached. If understood in the former sense, it would render meaning­less the texts which speak of torture of slaves as admissible in certain

¹ Con. 8. 7. 2, 3.

² Delation of crime, 34. 9. 1; 37. 14. 1, 5; C. Th. 9. 5. 1; 9. 7. 2; 9. 16. 1; 10. 10. 1, 2. Hein, loc. cit.; Mommsen, op. cit. 493 sqq. C. 10. 11. 4, notissimum est cos solos exsecrabiles nuntiatores esse qui fisco deferunt.

⁸ C. Th. 9. 5.1. Bruns, op. cit. i. 249 for a fragment of the original of this lex. See also C. 10. 11. 8. 2.

< C. Th. 9. 6. 2, 3; C. 9. 1. 20; 10. 11. 6.

⁸ C. 9. 8. 3; C. Th. 10. 10. 17. ⁶ 49. 14. 2. 6.

⁷ P. 5. 13. 3; 48. 4. 7. 2; 48. 10. 7; 48. 12. 1; 5. 1. 53; C. 10. 11. 7, 8. 2. The rule as to suppression of wills dates from M. Aurelius. The rules are somewhat similar to those as to evidence of slaves against their masters.

⁸ Nov. 90. 6. » P. 5. 16. 1, 2; D. 22. 5. 7 ; C. 9. 41. 15.

10 Cases as to ownership of them (C. 3. 32. 10; 9. 41. 12), tutela, disputed hereditas (P. 5. 15. 6; 5.16.2; D.34. 9.5.15; 48.10.24; C. 9.41.13),eAs early as Augustus it was enacted that torture was not to be resorted to except in serious crime¹⁸, and Hadrian provided that those slaves were first to be tortured who were most likely to be informed on the matter¹¹. A third person’s slaves could be tortured without his offering them, but only singly, and only when security or promise had been given for their value, with a double penalty if it were per calumniam accusatoris^{[374] [375]. The value could be recovered by an action praescriptis verbis though the agreement were informal1’. A slave manumitted to avoid the torture could still be tortured[376] [377]. On the other hand slaves ex dome accusatoris were not to be too readily accepted for torture[378]. Slaves were not to be killed under torture, ut salvi sint innocentiae aut suppliciow. It is clear that the officer in charge of the tormenta had a very wide discretion. But the torture was to be in reason and this was for the judge to determine1. It seems indeed that the question whether a man should be tortured at all was always in the discretion of the court, and not of a party[379] [380].}

It is frequently laid down that a slave is not to be examined for^[381] or against^[382] his dominus, or one jointly owned for or against either master^[383]. As to evidence against domini this is a very ancient rule. Tacitus, speaking of A.D. 16, alludes to it as based on vetus senatus-consultum^[384]. According to Dio Cassius^[385], Julius Caesar solemnly confirmed the rule. Cicero in several passages^{[386] [387] refers to it, basing it not on the doubtful­ness of the evidence, but on the reason that it exposes the master to an ignominy worse than death. Augustus and Tiberius evaded the rule (in maiestas), by ordering the slave to be sold to an actor publicus*. Tiberius even disregarded it altogether[388] [389].}

The exclusion of evidence on behalf of the master seems a much later notion. From the language of Tacitus it does not seem to have existed in a.d. 20^u. A text from Papinian quotes Hadrian as holding such evidence admissible^[390]. On the other hand Paul speaks of the evidence as excluded^[391], and an enactment of A.D. 240 speaks of this as an old established rule^[392]. It is plainly the settled rule of the Corpus luris.

The rule applied even though the master offered them or an out­sider was willing to pay their price^[393]. Ownership shewn as a fact, whatever its origin, barred the quaestio¹⁶. Nor could those who had formerly belonged to him be heard¹⁷. Bonae fidei possession equally barred the evidence¹⁸. It was not merely excluded: it was capitally punished, at least if volunteered¹⁹, and it may be added that evidence without torture was equally inadmissible²⁰. The exclusion applied also

ch. rv] Evidence of Slaves for or against Dominus 89 to slaves owned by father, child, or ward, except, in the last case, in the actio tutelae¹.

On the other hand an ownership created after proceedings were begun was no bar, nor was apparent ownership under a transaction which was absolutely void². The slave of a corporation could be heard against its members: they did not own him^s. And servi hereditarii are not slaves of the claimants of the hereditas, at any rate in an action concerning it, involving an allegation that the will was forged⁴. The uncertainty of ownership is mentioned, but this might better have led to exclusion.

It was not only in relation to evidence on behalf of the dominus that the rules underwent change: it is clear that in many other points the rules of later law are the result of an evolution, the tendency being always in the direction of exclusion. Thus Paul allows torture of a slave, collusively purchased, the purchase being rescinded and the price returned⁵. The Digest appears to limit this to the case where the acquisition is after the case has begun⁶. So Paul says that a slave manumitted to avoid torture can still be tortured⁷. The Digest in an extract from a work of Ulpian lays down the same rule, attributing it to Pius, and adding, dummodo in caput domini non torqueatur*. If a slave under torture did incidentally reveal something against his master, it was laid down by Trajan that this was evidence⁹, and Hadrian speaks, obscurely, in the same sense¹⁰. Elsewhere, however, Hadrian and Caracalla are credited with the contrary view¹¹, and we are told that the opinion of Trajan was departed from in many constitutions¹². Severus and Caracalla say that such evidence is to be received only when there is no other proof¹³. Paul declares that it is not to be listened to at all¹⁴. In a.d. 240 this is declared to have been long settled¹⁵, and, the enactment of Severus and Caracalla having been inserted in the Digest, in a somewhat altered form¹⁶, this must be taken as the accepted view: the safety of owners is not to be in the hands of their slaves. What is demonstrated in these cases is highly probable in some others. Thus it is likely that the extensions from owner to bonae fidei possessors¹⁷, and to slaves of near relatives and wards, are late¹⁸: the original rule having applied only to actual owners.

Ih. t. 10. 2, even castrensis peculii, C. 9. 41. 2. ² 48. 18. 1. 14. 15.

⁸ 1. 8. 1; 48. 18. 1. 7.

⁴ C. 9. 41. 10. D. 48. 18. 2 lays it down more generally.

⁵ P. 5. 16. 7. ⁶ 48. 18. 1. 14. 7 p. 5. 16. 9.

⁸ 48. 18. 1. 13. ⁹ 48. 18. 1. 19. ¹⁹ h. t. 1. 22.

IIh. t. 1. 5. ¹² h. t. 1. 19. is C. 9. 41. 1. 1.

1⁴ 48. 18. 18. 5; P. 5. 16. 4. « C. 9. 41. 6.

16 48. 18. 1. 16. Cf. C. 9. 41.1. Wallon, op. cit. 3. App. 12 for some temporary cases.

17 48. 18. 1. 8.

18 Ante·, p. 88. So the rule that servus damnati can be tortured, in caput eius, may have been law always, but the assigned reason, quia desierunt eius esse (48. 18. 1.12), squares ill with what has been said as to past ownership (p. 88) and suggests that the rule of exclusion was late.

There were some crimes to which the rule did not apply¹. Cicero speaks of corruption of Vestal Virgins, and coniuratio² as exceptions. It is, however, remarked by Mommsen that these republican exceptions are political³, and he thinks legal exceptions do not begin till Severus. It seems likely, however, that the exception, shortly to be mentioned, for the case of murder of a master was earlier. However this may be, Severus allowed the evidence in adultery, maiestas and fraud on the revenue⁴. These exceptions are constant (except for a short time under the Emperor Tacitus, who abolished them all⁶) and are repeatedly re­affirmed⁶. Other exceptions are mentioned. Several texts mention regrating, i.e. creating an artifical scarcity in food supplies⁷. Hermo- genianus mentions coinage offences⁸. Constantine allowed the evidence where a woman cohabited with her slave⁹, and also laid it down that a slave might be tortured, to discover if his dominus had prompted him to run away to a third person in order to involve him in the liability for receiving fugitivi™. The evidence was not admitted in ordinary crimes of violence". Thus the texts of the Digest allowing the slave of common owners to be tortured in the case of murder of one of them, where the other is suspected¹², are the result of the Sc. Silanianum, and the complementary legislation¹³.

Paul¹⁴ tells us that if a slave, who has run away, says, on discovery, in the presence of trustworthy people, that he had previously run away from his master, this is evidence available in the actio redhibitoria. Elsewhere¹⁶ he tells that in absence of proof of earlier flight, servi re­sponsioni credendum est: in se enim interrogari non pro domino aut in dominum videtur. This text appears in the Digest with quaestioni instead of responsioni. The reason is bad and Paul is the only authority for the rule. In the Sententiae he expresses a rule that a slave’s evidence in such a matter is admissible; the change of word in the Digest means little. But the other text, which may be the original statement, need mean no more than that the evidence of trustworthy people as to what the slave had been heard to say on such a matter, out of court and not under pressure, was admissible.

In relation to offences under the Lex lulia de adulteriis elaborate provisions are laid down. Slaves could be examined against their

^I 1. 12. 1. 8. 2 Pro Milone, 22; Part. Orat. 34. 118.

⁸ op. cit. 414. ⁴ C. 9. 41. 1. ⁵ Flav. Vop., Tacitus 9. 4.

⁸ C. Th. 9. 6. 2; C. 5. 17. 8. 6; 9. 8. 6; D. 5. 1. 53; 48. 4. 7. 2; 48. 18. 10. 1. Some of the texts deal with delation and accusation, but if this was allowed evidence was.

⁷ 5. 1. 53 ; 48. 12. 1. AU dealing with accusation. Cp. 48. 2. 13.

⁸ 5. 1. 53. ⁸ C. Th. 9. 9. 1; C. 9. 11. 1.. w c. 6. 1. 4. 4.

^II Milo’s manumissions are a precaution not so much against law, as against an uncon· troUable administration.

12 29. 5. 6. 2; 48. 18. 17. 2. Hadrian. „

13 Post, p. 95. Thus when owner is kiUed, servi hereditarii may he tortured though heres is a suus, and the evidence implicates him. 29. 5. 6.1.

14 21. 1. 68. 2. 15 P. 2. 17. 12; D. 22. 3. 7.

ch. iv] Evidence of Slaves in a charge of Adultery 91 owners, whether the accuser were a relative or not¹. It might be a slave of the accused or of the husband or wife of the accused^{[394] [395]. The point seems to be not only that slaves may here be tortured against their master, but that this is the regular mode of procedure and that there need be no preliminary evidence, or any special reason to think this slave knows something about the matter. If a slave, liable to torture in such a case, is freed to avoid the torture, the manumission is null, a rule of Paul, somewhat stronger than that laid down by him in other cases[396]. The accuser and the accused must both be present[397] [398]. After torture the slaves vest in the State, if and so far as the accused had any interest in them, in order that they may not fear to tell the truth’. Even if they deny, they still become public property, that they may not profit by any lie[399]. So also do slaves of the accuser, but not slaves of extranei, since in their case the reason does not exist’. If the accused is acquitted he or she can recover from the accuser, apart from calumnia, the estimated single value of the damage[400] [401]. If he is condemned, the surviving slaves publicantur6.}

The general proposition that slaves were liable for crime needs no proof^{[402] [403]. The master’s right of punishment (which did not necessarily exclude the right of the public authority) was lost, as to serious crime, early in the Empire11. They must be tried where they had offended[404], and thus the dominus, (who could defend by himself or a procurator[405],) must defend there, and could not have the case removed to his own province[406] [407]. The master’s refusal to defend did not amount to a conviction, or to dereliction. He remained owner; the slave might be defended by anyone, and would in any case be tried, and if innocent acquitted16. Slaves might be tortured on suspicion, and there was an actio ad exhibendum for their production for this purpose16. They might}

not, however, be tortured till the accuser has signed the charge, and given the usual undertakings¹. One to whom fideicommissary liberty was due was not to be tortured till the confession of someone else had raised suspicion against him^{[408] [409]. Servi hereditarii left to a heres or extraneus might be tortured on suspicion of having made away with property, and need not be delivered till after this was done[410] [411] [412]. So a slave might be tortured on suspicion of adultery with the wife, she being tried first to avoid praeiudicium*.}

In capital charges whoever was defending must give security indicia sisti, otherwise the slave would be kept in chains³. The rules of pro­cedure and general principles are the same as when the accused is a free man^[413], but it must be remembered that at no time was there a general criminal law. There was a mass of criminal laws, and principle is not easy to find.

It should be noticed that the rule that slaves cannot take part in judicial proceedings is applied even where they are the accused. We have seen that the master, or indeed anyone, may defend them, and that the defender is the real party is shewn by the fact that it is thought necessary to say that, after trial, it is the slave, not the defender, who is condemned^[414]. If no one defends, the court will not sentence at once, but will try the issue^[415], and in such a case the slave is allowed, ex necessitate, to plead his own cause—ut ex vinculis causam dicat'¹. In like manner slaves could not appeal though others could for them. Modestinus says that, if no one will, ips'i servo auxilium sibi implorare non denegabimus^w. The meaning of this is not very clear: in any case it seems probable that in earlier law the slave who could get no one to appeal was helpless. The concession, whatever it amounts to, may be due to Justinian.

The conditions of liability are not always the same. Some crimes could be committed only by slaves. Thus none but slaves could incur the penalties falling on fugitivi". It was capital for a slave knowingly to offer himself for military service¹. Slaves might be capitally punished for bringing claims at law against the Fisc, in certain cases^{[416] [417]. Slaves or liberti were punishable for aspiring to the decurionate[418]. Slaves were capitally punished for cohabiting with their mistresses[419] [420]. In some cases delation was punishable in a slave where it would not have been so in a free man[421]. Conversely there were crimes for which a slave could not be tried, owing to the punishment or to the definition. Here the hetero­geneous nature of the criminal law is brought into strong relief. Venuleius tells us that slaves can be accused under any law except those imposing money penalties, or punishments, like relegation not applicable to slaves, such as the lex lulia de vi privata, which fixes only money penalties[422], or the lex Cornelia iniuriarum, for the same reason. But in this last case he says: durior ei poena extra ordinem immanebit. He also tells us that the lex Pompeia de parricidiis does not, on its terms, apply to slaves since it speaks of relatives, but that, as natura communis est, it is extended to them[423]. On the other hand we are told by Callistratus that terminum motum, for which the old law imposed a fine, was capital in a slave unless the master paid the multa, a rule akin to that applied in delict, and one which might have been expected to be generalised[424] [425] [426]. For sepulchri violatio a freeman incurred a fine: a slave was punished, extra ordinem1“.}

In relation to punishment there were numerous differences. In theft and similar cases the criminal liability was alternative with a noxal action¹¹. There was prescription in adultery but not if the accused was a slave^[427]. The punishment might be different in the case of slaves, and in most cases was more severe^[428]. And though they had obtained freedom in the interval, they were to be punished as slaves^[429]. Vincula perpetua though always unlawful seem to have been occasionally imposed on slaves^[430]. A sturdy vagrant was given to anyone who denounced him, a right of action being reserved to his dominus¹. Furtum ceased to be capital in slaves when the Edict made it a private delict in freemen². Though condemnation as a servus poenae ended ownership, temporary punishment did not, and the peculium of any slave condemned was restored to his dominus³.

A slave being bound to obey, the command of the dominus, or of his tutor or curator, might be a defence in matters quae non habent atrocitatem fadnoris vel sceleris*. Where a slave wrote a gift of liberty to himself, at the order of his dominus, who did not subscribe it, but acknowledged it in letters, he was not free but was not liable under the lex Cornelia de falsis³. But command was no defence for murder, robbery, piracy, or any violent crime unless committed in the course of a bona fide claim of right⁶. It did not excuse for ocdsio, under the lex Cornelia de sicariis, though reasonable defence of the master would. Apparently it did not excuse for iniuria⁷ or for furtum³. In some cases it reduced the penalty. Thus, for a slave who committed gross violence, death was the penalty, under the lex Julia de vi, but if it were by his master’s orders he was condemned in metallum³. So, for demolishing sepulchres the penalty was metallum, but, if it were done iussu domini, the penalty was relegatio¹⁰. We are told elsewhere that this punishment was not applicable to slaves¹¹. Mommsen suggests¹² as the reason that their place of residence was not at their discretion. The reason is hardly conclusive, and we have here an exception. But these present enactments are somewhat haphazard: it is not clear that they express any real principle or policy.

The killing of masters by their slaves was the subject of special legislation. There was a tradition of an ancient usage for all the slaves in the house to be killed, if one had killed the master: Nero, in a.d. 56, obtained a senatus consult confirming this in general terms. The rule errs by excess and defect: it is needlessly cruel, and it requires prior proof that one of the slaves has actually killed. It does not appear in the later law¹³.

¹ C. Th. 14. 18. 1; C. 11. 26. 1. For light offences, flagellis verberati, 48. 2. 6; plotting against life of dominus, burnt alive, 48. 19. 28. 11; atrox iniuria, condemned in metallum; ordinary cases, scourged and returned for temporary chains, P. 5. 4. 22; 47. 10. 45; similar rule for abactores, P. 5. 18. 1. As to return to dominus, Mommsen, op. cit. 898. If dominus would not receive them, sold if possible, if not, perpetual opus publicum, 48. 19. IQ. pr., post, Ch. xvn. For coining, killed, but no right of fisc arose: no forfeiture unless dominus knew. This rule was general, P. 5. 12. 12; C. 9. 12. 4.

² G. 3.189. A slave who dug up a public way might be fustigated by anyone: a freeman would be fined, 43. 10. 2. Another case, 47. 9. 4. 1.

³ Schol. Bas. (Heimbach) 60. 52. 12; C. 9. 49. 1.

⁴43. 24. 11. 7; 50. 17. 157. pr. Factum vi aut clam, 43. 24. 11. 7, undertaking tacit fideicommissum, 35, 2. 13.

⁵ C. 9. 23. Q,post, Ch. xxv. Effect of subscriptio, 48. 10, 1. 8,14,15. 3, 22. 9.

³ 44. 7. 20. 7 47. 10.17. 8. 8 25. 2. 21. 1.

⁹ C. Th. 9. 10. 4; C. 9. 12. 8, >0 C. 9. 19. 2 (390).

¹¹ Ante, p. 93. But 40. 9. 2 also assumes its possibility. ¹⁸ Mommsen, qp. cit. 968. 1» Tac. Ann. 13. 32,14. 42.

A Sc- Silanianum, apparently of the time of Augustus, confirmed by a Sc. Claudianum and a Sc. Pisonianum, and again by an Oratio M. Aurelii¹, provided for the torture of slaves if there was reason to think the master had been killed by them. After the truth had been discovered by torture the guilty slave might be executed^{[431] [432] [433]. The slaves who might thus be tortured were those under the same roof or hard by—all who were near enough to help the master and failed to do so3; not, for instance, slaves who were in a remote part of the property, or on another estate[434]. If it occurred on a journey, those with him, or who had fled, might be tortured, but if none was with him the Sec. did not apply[435]. Those partly his might be tortured unless at the time protecting another owner[436]. Slaves freed by the will might be tortured, but with caution[437]. Trajan added even inter vivos liberti with ius anuli aureis.}

The power extended to slaves of children not in potestas, to slaves castrensis peculii, and, by. a Sc. of Nero, to those of wife or husband^[438]. It applied also on the death of a child, actual or adopted, living with the paterfamilias, whether in potestas or not (though the latter case was doubted by Marcellus), even if the paterfamilias were at the moment cum hostibus or even dead, if his hereditas were not yet entered on^{[439] [440]. But it did not apply to slaves of'the mother where a child was killed11, nor of socer where vir or uxor was killed[441]. Where a son, instituted by his father, was killed before entry, a slave legated or freed by the father’s will might be tortured, the gift failing by the torture[442]. The difficulty is that he is not and never would be the heir’s. Scaevoia decides that the Sc. applies, probably because the slave is the property of the hereditas, which represents the deceased father[443]. If it were a disinherited son, Paul holds that the slaves of the father could not be tortured till it was seen if the hereditas was entered on: if not, they could be tortured, for they would be his; if it was, they were alieni1'.}

The basis of the liability was that they did not render help, armis, manu, clamors et obiectu corporis¹. The torture was not punishment: it was a preliminary to the supplicium which awaited the guilty person. Not doing his best to save the dominus sufficed to justify torture: more than this would of course be needed to conviction of the murder. Though it were clear who killed, the quaestio must continue, to discover any prompters^{[444] [445]. The lex Cornelia gave a money reward for revealing the guilty slave[446]. Though the heir was accused the slaves might still be tortured[447] [448]. The Sc. applied only to open killing, not to poisoning and secret killing, which the slaves could not have prevented[449]: it must be certain that he was killed violently”. If the owner killed himself, only those might be tortured who were present, able to prevent, and failing to do so; in that case they were liable not merely to torture, but to punishment[450].}

Fear of personal harm was no defence, if they took no steps in pro­tection : they must prefer, says Hadrian, their master’s safety to their own⁷. But as failure to help was the ground of liability there were several excuses. Thus, unless circumstances shewed them to be doli capaces⁶, child slaves might not be tortured, though they might be threatened^[451]. Nor could those be tortured who did their best though they failed to save^{[452] [453]. If the master lived some time and did not complain of the slaves, or if, as Commodus ruled, he expressly absolved them, they were not to be tortured11. If the husband killed the wife in adultery, there was no torture, and if either killed the other, slaves were not to be tortured without proof that they heard the cries and did not respond[454]. It should be added that even the master’s dying accusation was not proof entitling the authorities to proceed at once to supplicium without further evidence[455].}

These provisions are merely ancillary to the main provisions of the Sc. Silanianum, the object of which was to secure that the death should be avenged, by preventing beneficiaries of the estate from taking it, and therefore freed slaves from getting freedom so that they could not be tortured, till steps had been taken to bring the slayer to justice. It provided that the will should not be opened till the quaestio had been held (i.e. all necessary enquiry made), with a penalty of forfeiture to the Fisc, and a further fine¹. The will was not to be opened, no aditio was to be made, or bonorum possessio demanded, till the quaestio*, the time for claims of bonorum possessio being prolonged accordingly, except in case of poisoning, where as there would be no quaestio there need be no delay⁸. There was an actio popularis (half the penalty going to the informer) against any who opened the will before the quaestio had been held⁴. If some slaves ran away and the will was opened and they were freed by it they could still be tortured®.

1 P. 3. 5. 1; C. 6. 35. 3. Other enquiry may be needed besides torture of slaves, 29. 5. 1. 25. By Sc. Taurianum penalties not enforceable after five years, save in parricide, when they are perpetual, 29. 5. 13.

2 29. 5. 3. 18,29; P. 3. 5. 1.

⁸ 29. 5. 21. pr. It appears that in later law similar delays might be ordered where other offences were supposed to have been committed by slaves. Dareste, N. B. H. 18. 583.

⁴ 29. 5. 25. 2.

⁸ h. t. 3. 17, h. t. 25.1. Justinian provided for a doubt left by this legislation as to the date at which, in such cases, the liberty took effect, C. 6. 35. 11. Many details of these matters are omitted, and seepost, Ch. xxv.