CHAPTER VII. THE SLAVE AS MAN. COMMERCIAL RELATIONS APART FROM PECULIUM. LIABILITIES.

To alienation of the master’s property his consent was always necessary¹. With that consent, which might be by ratification^[711] [712], or by a general authorisation if wide enough in its terms^[713], the slave could alienate anything^[714].

Similar rules applied where, having authority, he exceeded it¹³. Where A owed B 10 ex fideicommisso and 10 on an independent obligatio natural™, and a generally authorised slave paid 10 expressly towards the whole debt, 5 could be vindicated, as a general authority

160 Loss of Possession by Act of Slave [pt. i

to pay is not held to apply to natural obligations¹.

There is some difficulty as to the loss of possession by the act of the slave. Before discussing the texts it must be pointed out that, from the present point of view, it is immaterial whether the possession was originally acquired by the slave or not: in either case he is now a detentor through whom the possession is realised. Moreover we are told that the rules are the same whether it is a slave, a procurator or a colonus*', the slave’s lack of capacity does not enter into the question, and thus there is no room for the maxim that a slave cannot make his master’s position worse. It is indeed mentioned in this connexion, but only in an enactment of Justinian’s⁴ in which he excels himself in obscurity. It is the fact that possession is on a very different level from other rights, that is at the bottom of the whole difficulty.

If a slave possessed by his master still held the thing, it might be supposed that, however he attempted to exclude the master, the latter would still possess, as the slave’s possession is his. And so the rule is laid down by Africanus for the case where the slave of a pledgee turns his master out of pledged land⁵. But Paul lays down a different rule for moveables. If my slave takes my property I do not possess it till he restores it⁶. In the next text⁷ he cites Labeo and Pomponius in support of the view that, for this purpose, adding it to his peculium is not restoration, unless it was in the peculium before, or the owner assents to its being there. The difference may turn on the fact that, the land being immoveable, no change has occurred in the relation of the dominus thereto, while this would not be the case in regard to the moveables. But the solution is more likely to be found in con­sidering the case as one of a fugitivus. The view that his owner still possessed a fugitive was slowly accepted, and not all those who accepted it agreed that there could be possession through him⁸.

If a slave is deprived of the thing of course the master loses posses­sion. Thus if a slave occupying land is deiectus, the dominus has the

¹ 46. 3. 94. 3. Eisele suggests (Z. S. S. 26. QQsqq.) that where acquisition depended on alienation, as in purchase, no authority was needed. This is inconsistent with 15. 1. 37. 1 and with the rules as to mutuum {ante, p. 159), and has no support in the texts. That in connexion with which Eisele applies this principle does not, on its face, express such a rule and admits of a different explanation. Post, Ch. xv, and App. ni.

² 12. 6. 53. So, it would seem, of an expired authority, whatever the belief of the receiver {Arg. 12. 1. 11. 2).

» 41. 2.25. 1. < C. 7. 32. 12.

5 41. 2. 40. pr. Labeo remarks that, where the heres of a colonus takes possession thinking the colonus was owner, the true owner still possesses, 19. 2. 60. 1.

6 41. 3. 4. 8. Nor even then if I knew the theft, till I know of the return.

’ 41.3.4.9.

⁸ Post, Ch. xii. There are signs of a school controversy in which the Proculians, including Pomponius, took the view that it was impossible. The rule that a holder cannot causani possessionis mutare is sometimes used to explain the rule in the case of land. Ihering, Besitzwille, 347 sqq. He appears to hold that the rule applies to a detentor attempting to convert his holding into possession.

Interdict unde vi, whether he know of it or not¹. Conversely, if his slaves are left in possession of the land, the owner has not been delectus, even though he himself has been expelled, unless indeed they have passed into the possession of the deiector, as would result from their being bound or acting at his orders².

Mere momentary absence with no intention to abandon is of course immaterial³. The same appears to be true of death or insanity of the slave. It is true that he cannot any longer be holding consciously for the owner, but it is clear on the texts that the possession is not lost till a third person has taken the thing or the master has neglected to get actual control of it.

If the possession has passed to an adverse possessor the texts are agreed that possession is lost⁹. Justinian, however, in the enactment just mentioned, in which he appears to lay down the opposite rule for this case also, says that here too there had been dispute. It is some­times said that this is a mere mistake of his¹⁰. But it is at least possible that the dispute was whether the entry of the third party ended the possession, till it was known to the person concerned¹¹. Thus Papinian tells us that knowledge was not required¹², while he says that, of saltus hibemi, which are possessed only animo, the possession is not lost till knowledge, since, till then, the animus exists¹³. It may well be that some lawyers thought the same rule ought to apply where the slave had abandoned possession, for, if possession is retained, it can be only animo.

We shall have shortly to consider how far a slave’s contract can bind his master. But there is a difficult topic, which must first be

¹ 43. 16.. 1. 22. ² h. I. 45, 46. ³ 41. 2. 3. 1.

¹ 41. 2. B. 8, 25. 1, 40. 4. So if he lets the land to another—the possession is still in the master.

⁵ See Windscheid, Lehrb. § 157; Dernburg, Pand. 1. § 183; Girard, Manuel, p. 274.

⁶ 41. 2. 31, 40. 1.

1 41. 2. 3. 8, 44. 2. Proculus may be of this opinion, but he may mean only that the facts he gives do not amount to abandonment, 4. 3. 31.

⁸ C. 7. 32. 12. His enactment is so obscure that a dispute as to its meaning, begun by the Glossators (Haenel, Diss. Domm. 5), still rages. See the reff. in n. 5. The Digest texts are not much guide towards his meaning. Ihering holds that the meaning is merely that he can recover possession by interdict—an extended Unde vi. Grund d. Besitzessch. 114. See also C. 7. 32. 5.

⁹ 41. 2. 40. 1. 44. 2. ¹⁰ Girard, loc. tit. ¹¹ Windscheid, loc. cit.

» 41. 2. 44. 2. ¹⁸ 41. 2. 46..

162 Act of Slave under Master's Contract [pt. i considered. How far is a master bound by the unauthorised acts of his slave in transactions, essentially not the slave’s, but the master’s ? It is obvious that, in a great number of transactions, the actual carrying out of the contract will be left to slaves, and it is of importance to note how far it is material that the performance, or the breach, is not the act of the contracting master himself. The story told by the texts is not consistent at all points, but in general the principle is that a man is not, apart from his own privity or neglect, liable for conduct of his slave in relation to a contract not made by the slave¹. Where a slave, being directed by his master to point out the limits of a property sold points them out wrongly the land sold is that agreed by the master, not that pointed out⁸. A redhibiting buyer is indeed liable for damage done by his familia, but this is by virtue of an express rule of the Aedilician Edict⁸.

The rule protecting the master is laid down in general terms by Ulpian, who says: neque enim esse aequum servi dolum amplius domino nocere quam in quo opera eius esset usus^{[723] [724] [725] [726] [727] [728].}

The principles that a slave has no authority to make his master’s position worse, and that liability through him ought to be limited, are reflected in the texts dealing with his dolus. Dolus is a delict, and we are told that if a slave’s dolus arises in connexion with an affair which gives an actio de peculio, the actio doli is de peculio, but otherwise it is noxal⁸. A master suing can be met by an exceptio doli for what his slave did, but only if the transaction, in which it was done, was that now sued on, and was one in which the slave was employed. If indeed it was a peculiar e negotium then it is immaterial when or in what connexion the dolus was committed⁹, and the same is true if the slave was his master’s actor, i.e. one who had a general authority to act on his behalf¹⁹. We are also told that we may have an exceptio doli for the act of our own slave, (so far is he from binding us,) et de eorum dolo quibus adquiritur¹¹.

A few other illustrations may be given of the principle that the slave’s intervention in a transaction, which was not his, does not bind his master. Money lent by a slave can be validly repaid to him¹â€™, even though it was dominica pecunia, provided that in this case the loan was authorised¹⁸, as otherwise there would have been no alienation of the money¹⁴. The same rule is laid down for the case of deposit by the slave, or of sale by him. But here it is observed, on the authority of Sabinus, that this is true only if the other party has no reason to think

¹ 13. 6. 20.

²As to this and the literature, Windscbeid, Lebrb. § 401, n. 5.

⁸ See especially 13. 6. 10.1—11. < In. 4. 5. 3.

⁸ 4. 9. 7. 4; 47. 5. 1. 5. ⁶ Lenel, Ed. Perp. §§ 49,136.

⁷ In societas, owing to its confidential nature (17. 2. 63. pr.; C. 4. 37. 3) a master who was a socius was fully liable for negligence of his slaves, authorised to act, 17. 2. 23.1.

⁸ 4. 3. 9. 4a. ⁹ 44. 4. 4. 17. ¹⁰ 44. 4. 5. 3.

44. 4. 4. 17. This is obscure: the dolus must proceed ex parte a^toris (h. t. 2. 2), and the case must be that of a slave in whom other persons have such rights that they can acquire through him, and who is contracting for them with his master.

12 44. 7. 14. I⁸ 46. 3. 35. 1* Ante, p. 158.

164 Novatio and Discharge by Slave [pt. i the dominus would not assent to such redelivery¹, and doubtless this must be generalised. Payment to a slave is not satisfaction of a condition of payment to the master unless the latter consent’. A similar idea governs the rule that if A owes B a res under a will, and C gives B’s slave the thing, the right under the will is unaffected. There is here, however, another point: alioquin consequent erit ut etiam si tu ipse servo meo earn donaveris, invito me libereris. quod nullo modo recipien­dum est, quando ne solutione quidem invito me facto libereris³. There is no solutio without consent of the person entitled⁴: he may be in mora for refusing a proper tender, but till he has accepted it there is no solutio.

There is not much authority as to acknowledgements and receipts given by slaves apart from peculium. We learn that they can novate, by order or with ratification⁵, but not without any authority: in that case they acquire a new right to the master, without, ipso iure, destroy­ing the old⁶. This is so, whether it was the slave’s or the master’s contract: in the former case one might have thought that as solutio could be made to him, so might novatio. But novatio is not in fact solutio and it re­quires that the obligation should be in some way altered, and this would be to bind the master, and might prejudice him. Conversely we are told by Gaius that if a slave promises, novandi animo, this is a mere nullity: it is as if the stipulation had been a nullo and the old obligation is unaffected⁷. This is said quite generally and seems to exclude even the case of iussum, and the titles in the Digest and Code⁸ dealing with novation do not mention a novatory promise by a slave. The explana­tion is to be found in the character attributed to promises by slaves, shortly to be considered⁹. In a similar way, though he can take an acceptilatio¹⁰, he cannot give one, even iussu domini¹¹.

A slave can give a good receipt for money paid to him”, at least if he lent the money under authority both to lend and to receive”, and we may assume, in view of the texts above cited, that the first implies the second, unless the contrary appears.

Thus, apart from special authority, a slave cannot release or vary in any way an obligation he has acquired to his master: a fortiori, an obligation not acquired through him¹⁴. A gave B’s slave a mandate to ¹16. s. 11.

² 35.1. 44. pr.—3. If a man undertakes to pay to A or the slave of T, the payment may not be to T. So a condition of payment to the heres is not satisfied by paying his master.

⁸ 30. 108. 1. ⁴ Ibid.; C. 8. 42. 19. ⁸ P. 5. 8.

«46.2.16. ’ G. 3. 176, 9; In. 3. 29. 3. ⁸ 46. 3; C. 8. 41.

² Post, p. 165. 1« 46. 4. 11. 1. n 46. 4. 22.

¹² 46. 3. 102. 2. 18 C. 8. 42. 19.

¹⁴ Where A’s slave, B, made a contract of maritime loan with X, and X desired a release from some of the obligations, a release or variation agreed to by C, another slave of A, who was to be with X on the voyage, having certain duties, but no contractual powers, was a mere nullity, 45.1.122.1.

ch. vn] Slave's Contract apart from Peculium 165

pay money which A owed to B. He borrowed it from X. In B’s accounts the slave put it down as received from A. X had not lent the money with any special reference to A. A was not released and B had not acquired an actio mandati against A through the slave. If it had been expressly lent for the purpose of paying A’s debt A would have been released, but would have been liable ex mandate¹. The point is that if the loan was not in pursuance of the mandate it can give no actio mandati. At the time the money was attributed to A’s debt, it was already the property of the creditor, and though the transaction be, as this probably was, within the slave’s general authority, this does not entitle him to give what is essentially a fictitious receipt.

Under what circumstances did a slave’s contract bind his master, apart from peculium ? Obligation was a personal matter. Agency in the modern sense was unknown to the civil law. We know that at civil and praetorian law a slave was pro nullo, but that iure naturali he was a man like another². Hence the rule: serin ex contractu civiliter non obligantur, naturaliter obligant et obligantur[729]. Thus his promise creates a naturalis obligatio, but this obligatio which, as we shall see later, survives manumission, affects only himself, not his master⁴. Here we are concerned only with the master’s liability to action.

Broadly the slave’s contract did not bind the master apart from the peculium unless it came within certain categories for which the Praetor established actions, i.e. unless it was made under iussum, or as magister navis, or as institor, or the master profited. There were, however, some exceptions at least apparent.

(i) In all bonae fidei transactions, of the slave, the master was liable in solidum for his own personal dolus*. The rule in stricti iuris trans­actions is not easily made out, owing to the comparative rarity of references to promises by slaves: there is some obscurity as to the effect of dolus of the promisor, in general. It is sometimes said that a promise by a slave could not have any specifically civil law effects, and was thus in no way different, at least as far as civil law was concerned, from a mere pact⁸. Upon the texts it seems that, in classical law, the only remedy for dolus by the master in such a case was an actio doli, the point being that it was the slave’s contract and another man’s misconduct. In later law an actio utilis was, it seems, given against the master⁷.

166

Actio Quod Iussu

[pt. i

(ii) Ulpian tells us that if a slave quasi tutor egerit, Severus pro­vided that a iudicium utile lay against the dominus', a variant of the actio negotiorum gestorum contraria. The text expresses no limit. But it does not say that the liability was in solidum, and it was probably limited, like the other actions on a slave’s transactions, to the peculium, etc. It may be objected that, if so, it would not have been utilis: it would have been simply an actio protutelae de peculio, on the analogy of negotiorum gestio^{[730] [731]. The explanation seems to be this. The actio protutelae was fictitia, though we do not know the exact form[732] [733] [734]. A slave, though capable of ordinary quasi-contractual relations, could not conceivably be a tutor. Thus the fiction which would suffice for a freeman would not serve for the case of a slave. Hence a double fiction and the designation of the action as utilis*.}

We pass now to the four actions above mentioned.

A. Actio Quod lussu.

By the Praetor’s Edict“ the dominus was liable in solidum on an undertaking of his slave of either sex^[735], made iussu eius^[736]. No special form of authorisation was needed: it might be general or special ”, by mandate³ or by ratification^{[737] [738]. There is indeed one text which seems to suggest that ratification was not enough", but it does not say so and another text shews that ratification sufficed[739]. Endorsing the slave’s chirographwm sufficed, and this looks like ratification[740]. The iussum is not a command but an authorisation[741] [742], and the majority of the texts speak of it as an authorisation to the third party, not to the slave1“. There are a few that do not make this presumption, but none expressly contradicts it. It is now almost universally held[743] [744] [745] that a communication to the slave can never suffice to create the liability, unless it involves oh. vn] Quod Iussu. Notice to Third Party 167 an indirect communication to the other party. It may be remarked that this requirement of communication to the third party is nowhere expressly asserted1. Gaius indeed observes that the third party con­tracts with a view to the liability of the dominus*. But it has been pointed out that similar language is used in the case of the actio de peculio where knowledge that apeculium exists is not necessary’. More­over this communication could not have occurred where the actio quod iussu was made possible only by ratification[746] [747] [748] [749] [750] [751]. Such expressions as iussu domini cum servo contractum est* are common and imply that the authorisation is communicated to the third party. But elsewhere the action is given si voluntate domini servus emit*, and this suggests the other view. The fact that communication occurs in most of the texts shews that this is the usual case but not that it is essential. As this additional liability would hardly be undertaken except as an inducement to the other party to contract, it seems obvious to communicate it to him, but not that this should be essential to liability. In some of the texts which speak of iussum to the slave, and are disposed of by Windscheid as implying indirect communication to the third party[752], there is no sign of any such step and their plain sense seems to exclude it.}

It may be pointed out by way of analogy that where a filius familias or slave acted as a nauta, the paterfamilias was, by the Edict, liable in solidum for his receptum, if he received voluntate eius*. Nothing is said of communication to the extraneus. Thus there is nothing im­probable in the idea that quod iussu was subject to the same rule. And the deceptio mentioned in one text was not likely if the iussum had been communicated to the third party’.

The general result is that while the texts leave doubt as to the earlier law, they are clear that, under Justinian, the contractor has the right to actio quod iussu if a iussum exists whether he know it or not^[753]. It would seem to follow that it is not essential that it be known even to the slave “. Whether the iussum might be a general authorisation to any one to contract, or must have reference to a specific person cannot be said from the texts.

168 Quod Iussu. Limits of Authority [pt. i

The liability applies only to contracts of his own slave¹, and not of those acquired after the transaction^{[754] [755] [756] [757] [758], and it is to be supposed that as previous iussum is useless here, so is ratification: the reasoning in the text would certainly cover it. There must be express words of authorisation: mere words of confidence and the like do not suffice[759]. Thus becoming surety for the slave did not suffice: this was acting as a stranger, and if the fideiussio should be unenforceable, quod iussu would not lie[760]. A pupil requires his tutor’s auctoritas. According to Paul the tutor himself can give a iussum, and the action will be given, quod iussit tutor, but only if the transaction was for the benefit of the pupillus6. The reason for the restriction is not clear, in view of the fact that, according to Labeo, the iussum of a curator of a prodigus, furiosus or minor, and even that of a procurator, suffices without this restriction’. In the case of actio institoria, curator and tutor are on the same level7. Paul’s view may be an expression of the idea that the contract must have relation to the affairs of the person to be made liable. It is indeed held by some writers8 that this is the case, but there is little support for this in the texts, this passage not being usually cited in support of it. On the whole it seems probable that it is merely a personal doctrine of Paul.}

The iussum is revocable in all cases up to the time when the contract is actually made⁸. It does not exclude the actio de peculio though quod iussu is always better¹⁸ Like other contractual actions it is available against the heresⁿ, though the iussum itself is revoked by the death of iussor^w.

The iussum must be exactly followed. Thus if a slave, authorised to sell to A, sells to B, the master is not bound¹⁸. But an act in excess of instructions does not wholly vitiate the transaction: it is valid so far as the authority goes. If a slave, authorised to borrow at 6 °/₀, borrowed at a higher rate, the master owed 6⁸/₀¹⁴. If a slave, authorised to sell for 10 sold for 8, the master could vindicate the thing and there was no eaxeptio, without indemnification¹⁵. These texts shew no trace of the dispute which existed on similar points in relation to mandate¹â€™,

though the reasoning which guided those, who there took the view that it was void, would apply equally here: nam qui excessit aliud quidfacere videtur'. Here the dispute is between the principal and the third party, while in mandate it is between the principal and his agent: the matter is not mentioned in the few texts we have dealing with the action ad exemplum institoriae, brought by the third party against the principal’. This, however, scarcely seems material, and the difference, so far as it goes, supports the view that the iussum need not be known to the third party. His state of mind is not material: what matters is that the dominus has declared his willingness to accept a certain obligation’.

The transaction must be by the slave: a dominus borrowing and directing the money to be paid to a slave is liable directly and not quod iussu*. In one text the separate individuality of the slave is very clearly brought out⁵. If he is in partnership, iussu domini, the latter’s liability is quod iussu with no limitation to quod facers potest. He is not the partner, and this defence is not available to anyone else, even heirs or other successors. It is however indifferent’ whether the trans­action be in the master’s affairs or connected with the peculium\

B. Actio Institorla.

This action is given by the Edict⁸ against a dominus or domina who appoints a person of either sex, slave or free’, to manage a business¹⁰. It applies to all transactions of the business, and is in solidum, quasi iussu'¹. The institor may be a servus alienus'³, but, if he is, the liability is not accompanied, as in the case of servus proprius, by acquisition of all the rights also. These vest in the true dominus and the transfer of them, or their results, can be obtained by an actio negotiorum gestorum contraria'³. Thus where A appoints B’s slave, A will be liable to the present action and B will, or may, be liable to the actio de peculio'*. The liability rests on the voluntas of the dominus'³, and thus if a son or slave appoints an institor without actual consent of the paterfamilias, the latter is liable only to an actio institorla de peculio'³. The liability is perpetual and extends to the heres'⁷.

1 17. 1. 5. pr.

’ e.g. 3. 5. SO.pr.·, 14. 3. 5. 8, 16,19..pr.; 17. 1. 10. 5; 19. 1. 18. 25.

³ Cp. 4. 3. 20. pr. where there was no such declaration. Mandry, op. cit. 2. 565 sqq., takes a different view.

* 15. 4. 5. ⁸ 17. 2. 63. 2.

® lb.; 15. 3.5.2; 15.4.1.1,5; 16.1.25, mostly cited Vangerow, Pand. § 240.

7 As to formulation^posi, App. n. ⁸ 14. 3. ⁹ 14. 3. 7. 1, 8.

¹⁰ 14. 3. 1: as to different sorts of ins tit ores, 14. 3. 5,16; P. 2. 8. 2. See also Mayer, Actions Exercitoria et Institorla, 25—32.

n 12. 1. 29; P. 2. 8. 1. ¹³ P. 2. 8. 2.

¹³ 14. 3. 1. For a case in which he is the slave of the other party, see 14. 3.11. 8. 1*14.3.7.1,17.1. i® 14. 1.1. 20; C. 4. 26. 1,6. ¹⁶ 14. 1. 1. 20.

1? 14. 3.15. It is not affected by freeing the slave: if he continues to manage the business, no new appointment is needed, 14. 3.19.1.

A pupillus dominus is liable if he appointed auctoritate tutoris, or if locupletior fact its, the liability in that case having an obvious limit. Apparently on such points the rules are as in the actio quod iussu¹. On the death of the appointer, the heres is liable, and will be liable, if he allow him to continue his management, for future transactions’. As to transactions, vacante hereditate, the heres, even impubes or insane, is liable to any creditor who did not know of the death’, and, according to one text, even if the creditor did know⁴; the reason assigned being propter utilitatem promiscui usus. The fact that actio institoria is available does not bar other actions to which the transaction may give a right, e.g. redhibitoria". But if rightly brought it necessarily excludes the actio tributoria since, while that refers necessarily to res peculiares, this refers to dominica rnerx⁶.

The liability is only on those transactions connected with the business to which the man was appointed⁷. This rule plainly leads to a number of distinctions. Thus one appointed to buy cannot so bind his master by selling, and vice versa⁶. But a loan, for the purpose, to one appointed to buy, was enough, and if the creditor knew that the loan was for the purpose of the business, he need not see that the money is so spent⁹. A loan of oil to one appointed to deal in oil is good¹⁰, and, generally, if a transaction is within the scope of the employ­ment a pledge or security in connexion with it is good and imposes on the master the liabilities of a pledgee¹¹. Where A was appointed to two distinct functions, to trade in oil, and to borrow money, and X lent him money in view of the first business, but it was not received for that purpose, X sued on the assumption that it had been so received, but failed as being unable to prove this point¹â€™. Novation of the obligation destroys the actio institoria, the obligatio being no longer that contemplated by the appointment¹³.

The liability may be limited in various ways. Thus a number of institores may be required to act together¹⁴, or dealing with a par­ticular person may be prohibited by notice to that person¹â€™, or they may be required to contract only with security¹⁶. But the exception based on such prohibitions may be met by a replicatio doli, if the defendant do not offer what might have been recovered by the actio de peculio et in rem verso¹''. Any other conditions imposed on the power

^I 14. 3. 5. 18, 6, 9, 10, 17. 2. 2 14. 3. 17. 2.

«14.3.5.17,17.2. � h. t. 17. 3. « h. l.pr.

⁶ 14. 3. 11. 7. Post, App. n.

⁷ 14. 8. 5. 9—11; G. 4. 71; In. 4. 7. 2, not confined to operations in any one place, 14. 3.18.

⁸ 14. 3. 5. 12. ⁸ 14. 1. 7. 2; 14. 3. 5. 13. ¹⁸ 14. 3. 5. 14.

^II If arrha was taken and not returned by an institor to sell, the master was liable, h. t. 5. 15, 16. An institor appointed to lend does not necessarily bind his master by becoming surety, but if instead of lending money to A he promises it to A’s creditor, this is good, h. t.19. 3.

¹⁹ 14. 3.13. pr. As to the point of litis consumptio raised by the text, see post, App. n.

¹⁸ 14. 3. 13. 1. 1« h. t. 11. 5. 1« lb.\ h. t. 17. 1.

¹⁸ h. t. 11. 5. W h. t. 17. 4. of contracting must be observed: just as they might be barred from contracting with a person or class, so their contracts might be limited to dealings with a person or class. If these restrictions are repeatedly changed, in such a way that contractors are deceived, they do not pro­tect¹. In like manner the liability may be limited, or barred, by notice over the shop door^{[761] [762] [763]}. This must be plain and in a conspicuous place, and couched in a language locally known² But if it is duly set up it is immaterial that a contracting person did not see it^{[764] [765] [766]}.

The liability of the institor does not concern us. Of the master’s right against a third party it is enough to say that in late law the principal acquires rights of action against the other party to the con­tract, though the institor be not his own slave, or even not a slave at all, provided there is no other way of recovering“. But he always has an action of mandate, or negotiorum gestorum, against the institor for cession of his actions and against his master if he was a slave. In this case it may be only de peculio if the slave offered his own services. If he should be the slave of the other party, the dominus is not directly liable, since the contract is made with his own slave. But he can be sued de peculio, as on the mandate given to his slave, or de in rem verso, for the price which he owes to his own slave“.

Lenel^[767] holds that the action for the case where the institor is a slave is properly called utilis: the primary action being that for the contract of a liber homo. He accounts for the fact that it is not so called in the Digest on the ground that it was the commonest case, and he shews a text of Julian, in which the word utilis does survive^[768]. This case is however exceptional on other grounds: the institor is the slave of the other party. Lenel sees in this not the original-cause of the epithet utilis, but the cause of its retention in the Digest. The point is not very material in substantive law, but the fact that the dominus is acquiring by his contract with his own slave, a right against a third person, is, as Lenel himself notes, a reason for hesitating as to whether the action was the normal actio institoria. He observes however that Ulpian tells us that it is a sale⁹, and thus would satisfy the Edict, which gave the action on actual legal transactions alone. But he does not note that this question was in dispute. Paul, and even Ulpian himself in the case of a son, say definitely that such a transaction was not a sale¹⁰. They are writing long after Julian. It is thus easy to see why

172 Actio Institoria. Knowledge of Third Party [pt. i

he calls the action in this case utilis. In the next text^[DCCLXIX] another excep­tional case is considered, and there too Julian is cited as holding that, though on the facts the actio institoria was excluded by consumption, an actio utilis lay. Here too the institor was a slave. This is hardly a likely form, if the action lost had also been an actio utilis, and the explanation which Lenel offers for the other text, (this one he does not note,) namely that Julian’s language has been freely altered, seems hardly sufficient. On general principles it seems unlikely that the action which was primary in importance, and in all probability first in time², would be called utilis. Nor does the fact, probable in itself, that the actio was fictitia require that it should be called an actio utilis.

So far the rules of the action are fairly simple, but there is one point which has been the subject of much controversy. It is the question whether, and if so, how far, the fact of the appointment, and the pertinence of the contract to the business, must be known to the other contracting party³. It is clear that if the fact of the appointment and the relevance of the contract are known, the action lies in the absence of special restrictions, proper steps to secure the publication of which must have been taken⁴. But no text anywhere hints that it is an essential of liability that the third party know of the appointment, and when it is remembered that the rules relate to continuous commercial enterprises, it seems far more probable that there was no such require­ment, but that the setting up of a man in trade, and so inviting people to deal with him, imposed on the principal the Edictal liability. This is confirmed by the words of Ulpian upon the actio exercitoria which is governed by the same principles: igitur praepositio certam legem dat contrahentibus*. It is the appointment, not notice, which creates the situation contemplated by the Edict. Moreover, unless the praepositio bound, without express notice, it is difficult to see how Ulpian should have thought it worth while to say: Conditio autem praepositionis servanda est: quid enim si certa lege vel interventu cuius­dam personae vel sub pignore voluit cum eo contrahi vel ad certam rem ? Aequissimum erit id servari in quo praepositus est^[DCCLXX]. On the whole the better view seems to be that the agency need not be communicated.

But the attention of commentators has been mostly turned to the other part of the question: was it necessary to the liability that the

third party should have known that the contract related to the business to which the institor was appointed ? The dominant view is that this too was necessary, that it was not enough that it had to do with the business, but the parties must have also contemplated this. Karlowa^[DCCLXXI] supports this view, partly on the ground that the words, si eius rei gratia cui praepositus fuerit contractum est², must grammatically mean “ with a view to,” and not merely “ within the scope of.” He adds that any other view would make the principal liable if the institor contracted only on his own account. The intent of the institor to act for the business is thus necessary, and this could have no meaning unless it were communicated. All this is of doubtful force when it is a question of a piece of positive legislation. Mandry³, taking the contrary view, denies that eius rei gratia, nomine, causa, need bear the meaning for which Karlowa contends, but rests his case mainly on the texts. Those that have played a part in the controversy are set forth by Schlossmann⁴. They are not conclusive either way. He observes of the texts', that one⁸ has no relation to the action, that the force of another depends on taking lea: to mean a condition of which notice is given, which it does not imply⁷, that of another the force depends on understanding permisit to mean “ expressly authorised,” which it need not mean⁸, that in another⁹ there was in fact no existing authority, and that in the others the transaction is of an ambiguous nature¹⁰. Of the texts cited in reply^[DCCLXXII] one shews that there was no communication of the agency, and that this of itself is plainly not regarded as fatal to the action¹².

It may be observed that the arguments, in favour of the view that the agency must be communicated, seem to confuse two things ; intent to contract in view of the agency and intent to contract in relation to the business to which the agent was appointed. Thus Karlowa¹³ infers from eius rei gratia that the contract must have been made with the institor, as such. But the res is the negotiatio, not the praepositio, and even on his own narrow interpretation of the words, they can mean no more than that it was with a view to that trade and they need mean no more than that the matter must be connected with the business. Thus the text lends no support to Karlowa’s thesis. It should also be noted that the two principal texts¹⁴ relied on by the supporters of this view go no further. They both speak of dealing with express reference to the negotiatio: they say nothing of the praepositio. The right con­clusion seems to be that it was necessary to shew that the transaction was with reference to the business to which he was in fact praepositus. In most cases this needed no proof—res ipsa loquitur. But some transac­tions were ambiguous: a loan of money to a shopkeeper may not be meant for any purpose connected with the shop. For the lender to be entitled to the actio institoria he must be able to shew that it was. This he may do by shewing that it has been applied to shop purposes or that its application thereto was expressly contemplated¹.

C. Actio Exercitoria.

On nearly all points of principle which concern us, this action is on the same footing as that we have been discussing. It is a praetorian remedy modelled on the actio institoria, and therefore later, though it is described as even more necessary². The general principle is that the person who is receiving the profits of a ship, (whether the owner or not,) called the exercitor*, is liable in solidum on the contracts of the person placed in command of the ship, (who is called the magister navis,)* if the ship was to serve a commercial purpose and the contract was within those purposes for which he was appointed⁵. The purposes covered money lent for the purposes of the ship, even though not so used, if the creditor took care to see that it was reasonably necessary, and pro­portionate to the needs’. Authority is the limit of liability. Voluntas of the exercitor must be shewn, not merely sciential. Thus if the ship carried goods of an unauthorised class, or was otherwise used for an unauthorised purpose, or was let, without authority, the action was not available⁸. If the borrower of money did not say it was for the ship, and meant fraud ab initio, there was no remedy against the exercitor*.

A magister must be in command of the whole ship¹⁰. If, however, there are several with undivided functions, the contract of any binds the exercitor: if they are of divided functions, e.g. one to buy and one to sell, each binds only within his scope¹¹. Their power may be so limited that all must act together¹². A contract by one of the sailors does not give rise to this action: they are not authorised to contract¹⁸. The liability covered, however, ex utilitate navigantium, the contracts of a deputy appointed by the magister, even though the exercitor had

¹ 14. 3.17. 3. So substantially Schlossmann, loc. cit. For similar case, post, p. 183. The Edict as restored by Lenel says nothing of notice (Ed. Perp. § 102), but elsewhere L. argues in favour of the existence of this requirement. See post, App. I.

² 14.1. l.pr.; C. 4. 25. 4. But the relative dates of introduction of the aedilician actions are very uncertain. Mayer, op. cit. 18—25.

s 14. 1. 1. 15. < 14. 1. 1. 1. ₄

6 14.1.1. 3, 7. Magister might be male or female, slave or free, proprius or alienus, even an impubes, D. 14. 1. 1. 4.

8 14.1.1. 7, 9. Or a loan to pay a debt incurred for such a purpose, h. 1.11.

7 14. 1. 1. 20, 6. pr. - 8 14. 1. 1. 12. ⁹ 14. 1. 1. 9—10.

io 14. 1. 1. 1. ¹¹ h. I. 13. 12 h. I. 14.

¹⁸ h. I. 2. In delict the rule was different. Ib.\ 4. 9. 7. 3; ante, p. 122.

ch. vn] Actio Exercitoria. Complex Cases 175 forbidden this, or any, deputy. In this point this action differs from the actio institoria^[DCCLXXIII], but the rule shews how little agency in the modern sense had to do with the matter.

The action is perpetua, is available to and against the heres, and is not lost by death or alienation of the slave². The case of my slave who is your magister gave rise to questions as in the actio institoria. I have an action against you if he contracts with me. But the exerdtor has no direct action against one who contracts with his magister, who is not his slave. We saw that in the institoria this was allowed only as a last resort³: here it exists only, extra ordinem, at the discretion of the praeses*. His remedy is to claim cession from his magister, by action ex conducto, or ex mandato, according as the man was paid or not, and in the case of a servus alienus this will be limited to the peculium unless the master was privy to the appointment⁶.

The exerdtor himself may be man or woman, pater or filius, slave or free“. If he be a slave or filius familias the paterfamilias is liable in solidum, if the exerdtio is voluntate dus⁷. There is mention of a difference between this, and the rule in institoria, due to the greater importance of the present case. But in fact the text⁸ lays down the same rule for both, i.e. that if it is voluntate, the liability is in solidum, but if only sciente domino, it is either tributoria or de peculio et in rem verso⁹. If such an exerdtor is alienated or dies, the liability continues as in the case of a magister¹⁰, and is not subject to an annual limit, as de peculio is^[DCCLXXIV], but this rule applies only where it is not in fact itself an actio de peculio, as we have seen it may be¹².

A further complication arises if my slave is exerdtor and I contract with his magister. I can have no actio exercitoria, but if the magister is free I can sue him¹³, and, if he is a servus alienus, his owner. In like manner if a filius familias appoints a servus peculiaris, or a slave a vicarius, as exerdtor, the paterfamilias is liable only de peculio unless he approve, in which case he is liable in solidum whether the contract is with exerdtor or magister, the filius who appointed being also liable¹⁴. The liability on contracts of the exerdtor also in such a case is insisted on, though the Edict speaks only of the magister. What this action on the contract of the exerdtor would be is not clear. It is not stated as an equitable extension of the exercitoria: it seems more probable that it

was an ordinary actio quod iussu, and that the text supports the view that knowledge of the authority was not necessary in that action¹.

D. Actio de in rem verso.

This, as we know it, is not strictly an independent action. It is always found combined with the limitation to the peculium, and is thus a clause by way of taxatio inserted in the condemnatio of the action, whatever it may be. It expresses the rule that, on a slave’s transaction, a master is liable, even beyond the peculium, to the extent to which he has profited. But as the liability has its own rules it can be con­veniently considered by itself.

The general principle is that a dominus is liable on the contract of a servus so far as the proceeds have been applied to his purposes², irrespec­tively of consent or even knowledge’. The action is not subject to an annual limit, on the death of the slave, and is available against the heres* of the dominus. It is regarded as the owner’s personal liability, and it is considered in the action before the question of peculium⁶.

The main question is; what is versio ? We are told that a versum is what is handed to the master or spent on purposes necessary or useful to him or ratified by him⁶, or disposed of at his orders however wastefully’, or, generally, used in such a way as would give a procurator a right of action⁸. The texts give us many illustrations’. To spend the money in a normal way on the master’s property is a versio, but not useless and unauthorised ornamentation of his house¹⁰. Money paid to a creditor of the dominus is a versum¹¹, even, it seems, where the creditor is the slave himself, since a debt due from the master to the peculium is, in the developed law, a burden on the peculium¹³. An acquisition may be in part versum, and so subject the dominus to this liability only in part”. Thus, if unnecessary slaves are bought as necessary, they are

¹ Of several exercitores, each is liable in solidum (14. 1. 1. 25; h. t. 2, 3), whether one is magister (A. t. 4. pr. 1) or they have appointed another, slave or free, 14. 1. 1. 25, 4. 2, 6. 1 (and thus if one of them contracts as customer with the magister, he has actio exercitoria against the others, 14. 1. 5. pr., perhaps utilis, arg. 14. 3. 11. 8, 12). But if they are actually working the ship together each is liable only 737-0 rata. Where each is liable in solidum, there is adjustment by pro socio.

» 15. 1.1. 3, or ancilla. 15. 3. Lpr., 7. 4; C. 4. 25.1, 2; In. 4. 7. 4; P. 2. 9.1.

« 15. 3. 5. 1; C. 4. 26. 3; Greg. Wis. 9. 1. < 15. 2. 1. 10; C. 4. 26. 7.

6 15. 3. l.jir.; In. 4. 7. 4. No liability for interest, apart from promise.

« 15. 3. 5. 2, 7. 1. 7 h. t. 3. 6. 8 h. t. 3. 2.

915. 3. pass.; In. 4. 7. 4; P. 2. 9.

10 15. 3. 3. 2, 4. In the last case the creditor may take the things away so far as is possible without damage. Money used about the household, perfumes used by the slave in a funeral in which the dominus was interested, these are versa; h. t. 3. 1, 3, 7. 3, In. 4. 7. 4. If your slave sells me an inheritance belonging to you and you take it away after I have paid a creditor, I can recover the amount as a versum, 15. 3. 7. 4.

h. t. 3. 1, 10. pr. In. 4. 7. 4, even a supposed creditor, if it is recoverable by condictio indebiti h. t. 3. 1.

» e.g., h. t. 1.1. 18 h. t. 10. 4; In. 4. 7. 4. versi for value, but not for price¹. A let a farm to his slave, and gave him oxen. These being unfit, he told him to sell them and buy others. The slave sold and bought, but did not pay, having wasted the price received. The new oxen being in the possession of the dominus, the vendor had the actio de in rem verso for the difference between the value of the new oxen and the price paid for the old’. A slave owing his master money borrows and hands the money to his master: this is a versum so far as it exceeds the debt. So far as it does not exceed the debt it is not a verswm whatever else it may be, even though borrowed at the master’s advice^[775].

The money would usually be received under express contract^{[776] [777] [778] [779] [780], but this is not essential: negotii gestio is enough[781], and even condictio furtiva lies for what a slave has stolen, so far as it is versum*· The fact that there is another remedy is no bar: money is lent to the slave of a pupMus by the slave of one who is absent reipublicae causa, the tutor signs and makes himself personally responsible. Nevertheless if the money has been devoted to res pupillares, this action lies[782].}

It is essential that there actually have been a versio. The slave’s statement that he is going to apply the thing to his master’s purposes does not make the latter liable: the creditor should see that it is so applied, or rather, not applied to anything else⁸. You gave silver to my slave, and he was to make you a cup, not necessarily out of that silver. He made a cup out of my silver, gave it to you and died. Clearly I could vindicate the cup. Nor was there any versio. So far as appears the silver you gave had been devoted to no purpose of mine. Mela was of a different opinion, because, it seems, of the right to vindicate the cup. But this could give only an actio de peculio on the slave’s contract’.

It is essential that the property remain versum. All that this means is that payment to the master may cease to be versum, if it be handed back to the slave’s peculium¹*. It has no relation to the actual preser­vation of the thing: though that be lost by accident, it is still versum^u. A slave borrows money to buy clothes: the price being paid, the lender has de in rem verso though the clothes perish. If the price is not paid, and the money is lost, and the clothes are in use in the family, both creditors have the action, as also if both money and clothes have perished¹. This is Ulpian’s view, and in accord with principle: for the time being, both were versa, and the destruction of one or both makes no difference. But, in the next text, Gaius³ says the dominos is not liable to both; the first person who sues gets the benefit, on some obscure principle of fairness. This application of the rule, " first come, first served,” is isolated: it disturbs the principle that the loss should lie where it falls, and that destruction of the versum is immaterial’.

The rule that the versio ceases if the thing return to the peculium is illustrated in several texts. If the master hand back the money to the slave it is no versum, even though the slave lose it, and do not pay the creditor, and even though the master knew this to be the likely result, though here there would be actio doli*. If the dominus pay the versum to a creditor it is still a versum, unless it were to a creditor of the peculium⁶. A slave who had borrowed on his master’s account lent the money to X, also on his master’s account: it was still versum. But the dominus can free himself by ceding his claim against X, since the nomen against X is the form the versum now has⁸. If the versio has once ceased by any form of merger in the slave’s counter debt, it does not revive, if that debt is paid⁷. Conversely it must be noted that some forms of versio are in their nature indestructible⁸.

It is usually said that versio is enrichment⁹, but this needs some limitation or explanation. An addition to the peculium is an enrichment of the master, but it is not a versio¹⁶. The law regards the peculium as distinct from the master’s property: that only is a versio which increases the latter fund¹¹. The expression, locwpletior foetus, is commonly used to express the condition of liability resulting from enrichment¹². It is not used in this case in the formal statement of the obligation¹³, though it is incidentally¹⁴. Moreover the case differs from ordinary cases of lia­bility resulting from enrichment, in that destruction or loss by accident does not destroy the right to recover“, as it does in other cases¹⁸. When

^I15. 3. 3.10. 2 h. t. 4.

⁸ It must be remembered that the whole theory starts from the single word versum. It is probable that the words from Gaius are misapplied by the compilers.

ⁱ 15. 3.10. 6. The payment to the slave must have been to repay him: a casual gift even of the same amount would not destroy the versio, h. I. 7.

« 15. 3.1. 2.

⁶ h. t. 3. 5. If the master has a versum through a particular slave, who is or becomes indebted to him, the versum is reduced by the amount of the debt, though there be in the peculium enough to meet it, h. t.10. 7, 8.

’ h. I. 9..

⁸e.g. money paid to creditor, or in perishables which are consumed, h. t. 3.1, 3.6—10,17. pr. etc.

⁹ Windscheid, Lehrb. § 483; Demburg, Pand. 2. § 14; Mandry, op. cit. 2. 467 sgq., etc. w 14. 3. 17.4; 15. 3. 2, 5. 3, 6.

^IIh. t. 3. 5, 11. Some texts confuse this distinction, but raise no real difficulty. See h. t. 1. 1, 19.

12 3. 5. 33. pr.; 12. 6. 14; 50. 17. 206. ^w 15. 3.1. pr. “ e.g. 14. 3. 17. 4.

15 Ante, p. 177. 16 3. 5. 36. pr.; 5. 3. 36. 4, 40. pr.; 11. 5. 4. 1. it is remembered that the principles of this action are developed by the jurists from the scanty words of the Edict¹, and are governed by those words, it will not seem strange that its rules should not exactly square with those of the iure civili remedy for causeless enrichment.

So far the matter seems fairly plain. We have now to consider some controversial points.

We have seen that if a slave has expended money in a way which would give an extraneus an action on negotia gesta, this is a versio*. This may be read as expressing a limit: it is very widely held that it does, and that the principle governing the action is that it lies, then, and then only, when a free person would have an actio on the negotium gestum*. But this idea seems to have been struck out to explain one or two awkward texts, which can, however, be far better explained without this doctrine, which raises more difficulties than it settles, so that on the whole the modified theory of enrichment, which also has many supporters^{[783] [784] [785] [786] [787] [788], is to be preferred. But strictly it is not possible to fit in the texts with the theory appropriate to any other remedy or claim. Versio is a conception by itself: in the hands of the jurists, it seems to have meant embodiment in the patrimonium as opposed to the peculium. The gestio theory fails in conciliating the texts : the enrichment theory nearly succeeds. The chief texts are the following. A slave makes a present to his master, out of the peculium. This is not a versio. So says Ulpian’, and this text is taken as an authority for the gestio theory. But the context shews that the reason why it is not a versio is that, in the writer’s opinion, the dominus is not enriched. It seems to mean that the thing is still a res peculiaris, and that the transaction is on the same footing as the case where the master sells a res peculiaris and keeps the price. This is a dolo malo removal from peculium, and so leaves its amount unaltered as against creditors’. In the immediately following text a slave borrows money from an extraneus and pays it to the master donandi animo, not intending him to be a debtor to the peculium. This is a versio. This is irreconcilable with the gestio theory, and also with the text just cited, if it is explained in terms of that theory. But the texts adjoin and are from the same pen. The point is that the present transaction is wholly independent of the peculium’’.}

A slave borrows money to procure his freedom. He pays it to his master and is freed. There is a versio as to any excess in the loan over his value. This is clear apart from the gestio theory, but cannot be reconciled with it¹. A slave pays a master’s debt with money he has borrowed, he himself being indebted to the master at the time. There is a versio of the difference. If he was not indebted to the master, and the latter reimburses him, the versio ceases, but if the master’s payment to him were not by way of reimbursement, but independent gift, the versio is not affected². The text is a long one and discusses the reasons: it speaks of nothing but enrichment.

One text raises a difficulty. A father owes money: the son promises it and is sued. This is a versio, so that if the son does not pay the father can still be sued, “ unless the son in taking over the obligation intended a gift to the father.” This is exactly in point, for it makes the right depend on the son’s having a claim as negotiorum gestor. But the words are in a nisi clause, of suspicious form’, and it must be remembered that in Justinian’s time a son’s finances constituted for practical purposes a distinct estate. If he gave donandi animo, it was as if a stranger had done so.

The fact that the jurists do repeatedly refer to the principle of gestio is explained when we note that, as Windscheid observes⁴, the text which states it’ most fully is considering what amounts to the necessary en­richment. From this point of view the use of the conception of gestio is clear. A man has the actio de in rem verso when there would have been an actio mandati or on negotia gesta, if it had been done by one acting with the intent needed for those actions: whether in the actual case it was done with that animus is immaterial. There, as here, the action lies, though the benefit conferred is destroyed by accident. There, as here, the action does not lie if the expenditure is of a useless nature, with whatever intent it was made’.

Mandry⁷ distinguishes between “ direct ” versio, where the thing was never in peculio, and “ indirect ” versio, where the thing, having been in peculio, is transferred in some way to the patrimonium. He observes that in several cases of such transfer there is no versio³, and holds that here, (though not in direct versio,) there is no claim unless the transfer would have given rise to an action on negotium gestum, or the like.

¹ 15. 3. 2, 3. pr. (considered by Von Tuhr, op. tit. 78 sqqA. Your slave lets to me a vicarius: I make him institor, and in that capacity he sens to you. There is a versio. But there has been no gestio on your behalf, 14. 3.11. 8,12. If I give notice to you not to give credit to a certain slave, my institor, and you do so, your emtio institoria against me is barred. But if I have received the thing and do not return it there is a replicatio doli, i.e. I am liable for the versio, though there is no real gestio, 14. 3.17. 4; cp. 3. 5. 7. 3. ² 15. 3. 10. 7.

⁸ 15. 3. 10. 2. nisi si donare patri filius voluit dum se dbligat. In this and the adjoining texts some cases of versio are discussed which could not arise in the case of a slave. See Mandry, op. tit. 2. 502.

⁴ loc. tit. ⁶ 15. 3. 3. 2. ⁶ 3. 5. 9 etc. Ante, pp. 176, 7.

7 op. cit. 2. 522 sqq. ⁸ 15. 3. 5. 3, 3, 7.pr. etc.

His reasoning seems to be that the right results from the act of the slave—the slave himself would have no claim except in such a case— and the relation of the master to the third party must be governed by the same principle as that with his slave, on pain of “ inner contra­diction.” Apart from this rather doubtful principle, the author finds support in those texts which speak of gestio as a basis^[789]. But these apply equally to direct versio, and there the author admits that they do not set a limit. He relies also on a text’ which says that if the master adeem the peculium or sell it or part of it and keep the price, there is no actio de in rem verso. This text is one of those which must be considered later⁸, in connexion with difficult questions as to the effect, on the liability de peculio, of ademption and sale of the peculium. The text following it in the Digest⁴ rests the exclusion on the ground that there has been no enrichment. From this and some other texts, it seems likely that all these facts are viewed as not affecting the liability de peculio at all: the things are still regarded as in the peculium, and the dominus is in no way enriched. Thus the text which Mandry cites shews merely that it is difficult to frame a case of indirect versio in which there was no debt to the peculium.

We have seen that there must be a negotium and a versio. What is the connexion between the two ? It is sometimes said on the authority of the gestio texts, that the versio must be an act of the slave’s, and it is clear that in the majority of cases it was so, for direct versio by the third party, under a contract with the slave, is substantially the same thing. But there is nothing in the form of the edict or the formula so far as we know them’, requiring or stating any such limitation. And there is one text⁶ which gives the action where the master himself applies the thing, and the circumstances shew that it was impossible for the actual acquirer of it, (in the case, a son,) to have been privy⁷. Not a few writers require however a great deal more than this. They hold that there must have been, between the original negotium and the ultimate versio, what may be called a causal nexus. There seems a close connexion between this and what has been called above the gestio theory. But in fact it is held by some who reject that theory, and re­jected by some who accept it. Karlowa⁸, who adopts the gestio theory, thinks there was a difference of opinion on the present point.

If the versio is direct there is no difficulty^[790]. It is only where the thing has been for a time in the peculium that the question arises. In relation to this the idea that the creditor must have contemplated the versio, ab initio, has little, a priori, to recommend it. The claim is a remedy for unfair enrichment and the intent of the third party seems rather immaterial. The Roman law was, perhaps, not liberal in remedies in cases of this kind², but here the remedy does exist and there is no obvious reason why it should be so limited.

On the texts however the question is not without difficulty. The majority of them are opposed to the requirement, though it is no­where expressly denied. In one text Paul quotes from Neratius³ (in a passage which can hardly be interpolated) the case of a son who bought a toga. The son died, and his father applied the toga, thinking it was his own, to the purposes of the funeral. The text adds that if the circumstances were such that the pater was under a duty to buy a toga for the son, the versio dates from the purchase; if not, from the funeral. It is clear that this was in the beginning a “ peculiar ” transaction, and the intent of the creditor was not material. In another group of texts Paul and Ulpian⁴, citing and limiting the views of Mela and Pomponius, discuss the case of a son who, having borrowed money, applies it to the dos of his daughter or sister. This is a versio in rem patris, so far as the father was going to give a dos, provided the application was with a view to carrying out a negotium of the father but not otherwise. Nothing is said of the intent with which the money was lent: the point of the text is that it might equally well on such facts be a negotium of the son s. The text then lays down the same rule for the case of a slave. The form of the addition is against any causal connexion, but the remark may be compilers’ work. In many texts, Ulpian in discussing the nature of a versio, uses language which seems to exclude the materiality of the creditor’s intent’. It may be added that the Institutes, which explain the action at some length, say nothing of any such requirement⁸. Less direct evidence is afforded against the need of causal connexion in the texts which make the versio destructible, by the fact that the slave becomes indebted to the dominus⁷. Such a rule makes the intent of the creditor a very unsafe protection.

But there are texts the other way. In his Sentences, Paul definitely⁸ subjects the right to bring the action to the condition that the money was given for the purpose of the versio. In the Digest¹ he gives it in a case in which he speaks of the contract as made with this object, as if that were a material factor. These might pass as mere expressions of Paul’s preference for subjective tests’, but there are texts independent of Paul. Ulpian³, in a case of loan of money, says that there is an actio de in rem verso if the money was lent for the purpose. In the immediately preceding text⁴ he seems to lay down a similar rule in a case of acquisition of goods. But all that he is there discussing is the question whether, if it is not applied to the master’s purposes, the fact that it was given for that purpose suffices to give the action, and he decides that it does not. In another text⁵ Africanus seems, though not very clearly, to require it in a case of loan of money. In the next text’ Neratius discusses a case in which goods have been bought expressly for the dominus, and A has become surety for the price. He holds that A has no actio de in rem verso, though he pay the price. The actual decision does not here concern us: the point for us is that the intent of A is clearly regarded as material. If now we examine the texts which really treat the intent of the third party as material’, we shall see that they are, as it seems without exception, cases in which the claimant of the action has paid money. This circumstance seems significant and enough to explain them. What is needed in this action is, as Neratius says’, identity of what was received with what was versum. A payment of money was in itself a colourless thing. It was no easy matter to follow and prove the application of the actual coins, and accordingly some lawyers lay down the rule, (and none deny it,) that if money is lent for the purpose of a versio, and the versio follows, the identity of the money received with that versa is assumed. This view is confirmed by the fact that in the case⁸ where the question is whether the lender of money to buy goods, and the supplier of the goods, have both in certain events the actio de in rem verso, the text emphasises the need of privity in the case of the lender, but does not mention it in the case of the vendor.

The question of the relation of this action to the actio de peculio is one of some difficulty. As described to us, it is not so much an in­dependent action as a clause in the formula of the actio de peculio*, and the question arises whether it had an independent existence; whether

¹ 14· 6. 17....

³ Ab where he says that if a slave borrows, ut creditors suo solveret, this is no versio though the dominus is released from an actio de peculio. The objective fact that it is not applied to any patrimonial purpose is enough to exclude de in rem verso, 15. 3.11.

8 15. 3. 3. 10. * 15. 3. 3. 9.

« A. e. 17. pr. 6 A. 1.18.

7 P. 2. 9.1; D. 14. 6.17; 15. 3. 3.10; A. t. Yl.pr.\ A. 1.18. Doubtful: 12.1.12; 14.3.17.4. See also 15. 3. 7. 4.

8 15. 3. 3.10. » e.g., In. 4. 7. 4b.

184 Relation between de in rem verso and de Peculio [pt. i

there was such an action which contained in its formula no reference to a peculium, and, in any case, whether it could be brought if there were no peculium. Von Tuhr¹ holds, as an outcome of his special theory as to the basis of our action, that there could be no de in rem verso if there were no peculium. He considers its purpose to be to provide for the case where the liability of the dominus to the slave is to release him from an obligation, not to pay money. This duty does not admit of exact estimation and so cannot be treated in the ordinary way as an addition to the peculium. As there can be no natural obligation to the slave, unless there is a peculium, it follows that there can be no actio de in rem verso. We shall shortly consider his general theory: here it is enough to say that he has to treat the texts with some violence in order to support this minor part of it’. He explains’ the perpetuity of the action, notwithstanding the ending of “peculiar” liability, apparently by the principle that the liability to the creditor is the primary liability and that subsists: the liability to the slave was little more than a facultas solvendi, and that is ended. But it cannot be both an obligation and a facultas solvendi, and the rule is in fact in conflict with his general theory⁴. On the whole evidence it seems likely that this action could be brought independently. It is clear that it lay when there was nothing in the peculium, for even the actio de peculio did’. It is also clear that it could be brought when de peculio no longer existed, because either the peculium had been adeemed without dolus, or the slave had ceased to be the defendant’s and the year had passed*. There are of course many texts which give it without mention of peculium, but there is none which unequivocally gives it where there has never been a peculium. But all that this shews is that an extraneus would not ordinarily trust a slave who had neither a peculium nor authority from his master. It may also be remarked that the use of the formula referring to the peculium, as well as to the versio, no more shews that an actual peculium was necessary than it shews that de peculio would not lie unless there was also a versio. It must not be forgotten that de in rem verso appears as the primary liability.

The peculium, as described in the Digest, includes not only

¹ op. cit. 238sgg.

³ He cites Baron as holding the same view. See Bekker, Z. S. S. 4.101.

⁸ od. cit. 236.

⁴ Karlowa thinks there was an independent de in rem verso, introduced later than de peculio, arguing from the introductory words oi D. 15.1 and 15. 3 (R. R. G. 2. 1154). Mandry takes the same view, de in rem verso having a separate basis in enrichment (op. cit. 2. 456). Lenel dealing with the question shews that the Edict gave only the one formula (Ed. Perp. § 104). See also Windscheid, op. cit. § 483.

⁸ 15. 1. 30. pr. But this is not to say that it lay when there was no peculium.

• 15. 3.1.1, 2,14, as to which last, post, Ch. xvi. ch. vn] Purpose of the Actio de in rem verso 185 corporeal things, but also debts due to the peculium from the master. As the subject of an actio de in rem verso is also usually the subject of such a claim from the master, and is thus already covered by the actio de peculio', the question arises: what purpose is served by the actio de in rem verso 1 The point is raised in the title, and it is answered by reference to certain circumstances under which it gives a remedy where there is no actio de peculio. Thus it is said that our action is available, though that de peculio is extinct, owing to ademption of the peculium sine dolo, or death or alienation of the slave, and expiration of the annus utilis*. But it cannot be supposed that these exceptional cases were the cause of introduction of the action, and indeed the texts shew clearly that this was not so. It is contemplated that, in the normal case, the actions are brought together—the question of versio being first considered’, and it is clear that the actio de in rem verso is regarded as giving the plaintiff more than he could have recovered by de peculio alone—proficere ei cuius pecunia in rem versa est debet, ut ipse uberiorem actionem habeat*.

The elements of a solution may be found in the answer to certain historical questions. The natural obligation between slave and master is of later introduction than the actio de peculio*, and the actio de peculio did not at first cover anything but the corporeal things in the peculium. At that stage the actio de in rem verso would have the obvious advan­tage of giving the particular creditor a better claim’. When the peculium is extended to cover debts to it, this utility is lost, and⁷ the subsidiary advantage of perpetuity alone remains. This view is confirmed by the fact that the classical jurists see little use in this action, and, in explaining it, fall back on these subsidiary cases. Von Tuhr, however, while he notes these changes’, is not satisfied with this explanation. He holds that when debts were included in the peculium, the actio de in rem verso changed its basis. Instead of resting on enrichment, it came to rest on a liability of the master to the slave, of a kind which could not be added to the peculium, because it could not be exactly assessed in money*. The case he has in mind is that in which the master’s obligation is, not to pay money, but to release the slave from some obligation he has undertaken. This may be done by other means than payment, and at less expense. It cannot be added to the peculium, and thus becomes the special subject of the actio de in rem verso, available only to the creditor whose property has been versum. The application of this theory to the texts, in which it is nowhere

¹ e.g. 15. 3. 19 in fin. ^a Ib.\ 15. 3.1.1, 2. See post, p. 227.

⁸ In. 4. 7. 4. â€? 15. 8.1. 2.

⁸ Pernice, Labeo, 1.152 sag. Post, Ch. xxix. * 15. 8.1. 2.

⁷ For expression and citation of contrary views, Mandry, op. eit. 2. 81, 82.

⁸ op. eit. 259 sqq. * op. eit. 82.

186 Actio de in rem verso. Condictio generalis [pt. i, ch. vn

indicated, and with a number of which it is irreconcilable, involves a great number of emendations and insertions.

The foregoing pages are an attempt to explain the rules of the actio de in rem verso, as set forth in the Digest. But even if they be regarded as doing this, it must be admitted that they do not account for all the language of the texts. Thus, to take a single instance, though we have not accepted the gestio theory, it is clear that the language of many texts is coloured by it. It is easy to account for this. The task of the lawyers was to define the meaning of the expression versio in rem of the Edict. To this end the existing institutions of the civil law, while they gave no sure guide, provided many analogies. These different analogies have coloured the language of the lawyers. The title shews indeed that there were differences of opinion as to the actual rules. How far these differences went, in particular, how far specific views can be associated with individual jurists, is a question too speculative to be here considered. Attempts to answer it have not been lacking, the writer in some cases going into very exact detail^[791].

A text in the Institutes’ tells us that what could be recovered by any of these four actions could also be recovered by direct condiction. This proposition, which has no equivalent in Gaius, has a little support from two texts in the Digest’, one at least of which has a prima fade look of genuineness⁴. As the substantive rights of the parties are not affected, the topic is of small importance to us, though it is of great interest in connexion with the general theory of condictio. The text has been the starting-point of a great mass of controversy’. Here it is enough to express a doubt as to the classical character of the rule, notwithstanding the reference in one of the texts to Julian⁸.