The liability of the mandatarius

(a) Dohts or dolus and culpa?

Mandatum was a gratuitous transaction. This must have been an important feature in determining the standard of liability imposed upon the mandatary.

"[DJolum et culpam mandatum, commodatum, venditum, pignori acceptum, locatum, item dotis datio, tutelae, negotia gesta [recipiunt]: in his quidem et diligentiam."¹⁰²

How can we reconcile this apparent conflict of opinions? Interpolation hypotheses have provided a popular and convenient answer. As a result, for some time, the view prevailed that in classical law the mandatary was Hable for dolus only. Later on, his liability came to be gradually extended, and references to culpa were interpolated into the classical texts in order to bring them into line with the more modern position.¹⁰³ Other authors have attributed the transition from dolus to culpa liability to late classical law.¹⁰⁴ More recently, an attempt has even been made to show that, already in high classical law, HabiHty was for culpa.¹⁰⁵

® Harder, Festgabe von Lubtow.

® Cf. supra, pp. 198 sq.

^B) Cf. supra, pp. 209 sq.

^Q Coll. X, II, 3.

“ Ulp. D. 50, 17, 23.

^B Cf. e.g. Arangio-Ruiz, op. cit., note 7, pp. 188 sqq.; Schulz, CRL. p. 556; Jors/Kunkel/Wenger, § 139, 2 a; William Μ. Gordon, "The Liability of the Mandatary", in: Synteleia Vincmzo Amngio-Ruiz. vol. 1 (1964), pp. 202 sqq.; Wieslaw Litewski, "La responsabilite du mandataire", (1983-84) 12 Index 106 sqq., 119 sqq.

Cf. e.g. Buckland/Stein, p. 516; Hoffmann, Fahrlässigkeit, pp. 21 sqq.; cf. also, as far as Ulpian was concerned, Tafaro, Regula, pp. 276 sqq. ¹⁰⁵ Hannu Tapani Klami, Tetieor Mandati (1976).

(b) Terminological problems

To me, there appears to be a certain element of truth in both the traditional and the two other points of view. For here, as elsewhere, Roman law was not static but developed in time; and this development is bound to have implied a broadening, not a restriction of liability. On the other hand, it is no longer permissible today summarily to suspect and brush aside all references to culpa in our classical sources relating to mandatum. Any analysis of the position in classical law must, however, take account of two further factors, which have been emphasized by Geoffrey MacCormack and Alan Watson respectively. The first of these relates to terminology.¹⁰⁶ Culpa, in classical Roman law, did not have a precise, rigidly defined meaning. It could cover a broad range of situations. More particularly, it could be used in (at least) two different ways: it provided (roughly) a Roman equivalent to the modern concept of negligence, i.e. the failure to exercise the care that a bonus paterfamilias would have exercised. But "culpa" could also refer to fault or blameworthiness in general, i.e. including dolus.¹⁰⁷ In the first sense, it was complementary to dolus, in the second it comprised it. If employed in the second sense, the term "culpa" could be used where the circumstances in fact disclosed a case of dolus.

(c) Altruistic and not so altruistic mandatarii

The second point is this. Mandatum, as we have seen, covered a very wide and heterogeneous range of situations. There was the truly altruistic person who took care of his friend's affairs. There was the professional lawyer, doctor or surveyor. There was the procurator in receipt of a salarium. Was it really reasonable to treat all these mandatarii alike, as far as their liability was concerned? Also, we must remember that the mandatarius did not necessarily act solely in the interest of the mandator. Where the mandatum is either mea et tua or tua et aliena gratia, the mandatarius' activity can hardly be described as altruistic, even if he is not remunerated for his services. The contract is (also) in his interest, and thus the dolus liability might not seem adequate. Even in the case of depositum, the standard of liability

¹¹¹⁶ Geoffrey MacCormack, "The Liability of the Mandatary", (1972) 18 Labeo 156 sqq.

¹¹⁾⁷ For a detailed analysis, see Geoffrey MacCormack, "Culpa", (1972) 38 SDHRTi sqq.;

cf. also idem, (1972) 18 Labeo 156 sqq.; Klami, op. cit., note 105, pp. 2 sqq.

^ltM MacCormack, (1972) 18 Labeo 157. Contra: Litewski, (1983-84) 12 Index 107 and passim. changed if, for once, the contract was not in the interest of the deposi­tor, but of the depositarius.^[2162]

Whilst, therefore, liability for dolus seems to have been the historical starting point,'^[2163] it is likely that by the time ot classical law the question was approached in a very flexible manner.^[2164] After all, what the judge had to assess was "quidquid ob earn rem N^m N^m A° A° dare facere oportet ex bona fide"; and the yardstick of "bona fides" called for a comprehensive analysis of all factors determining the circumstances of each case.^[2165] Thus, on a more general level, it had to be considered that breach of contract in the case of mandatum was a matter of particular concern:

"Mandati vioiatio violat duas res sanctissimas, amicitiam ct fidem.

On the other hand, condemnation involved infamia, a sanction not to be imposed too lightly. More specifically, however, the position of the parties had to be taken into consideration: whether the mandatum was in the interest of mandator, mandatarius and/or third party, and whether the services of the mandatarius were in fact, if not in law, remunerated.

(d) Mandatum: between suretyship and procuratio

If we want to gauge how all this worked out in practice,^[2167] we can take, at the one end of the scale, the liability of a person who stood surety at the request of the (main) debtor. According to Ulpian, he is liable, for instance, if, when sued for the amount owed, he fails to raise certain exceptions, even though he is aware of their availability.^{[2168] [2169] The same applies if, having paid the debt, he omits to inform the debtor, who consequently pays again.llft In both these instances, the fideiussor is probably not deliberately trying to prejudice the debtor; he simply does not bother to safeguard the position of the latter.[2170] This type of behaviour does not accord with the precepts of good faith; in fact, it comes very close to dolus. It is, in the words of Ulpian, "dolo proximum";[2171] or more fully: "dissoluta enim neglegentia prope dolum est."[2172] Thus we can see that dolus was the normal standard of liability for this prototype of the altruistic mandatarius, but that there was a tendency to extend the scope of liability to cases of extreme recklessness which were not very far from dolus.12}

At the other end of the scale we find the procurator. Here liability seems to have gone beyond what could conveniently be expressed in terms of dolus or dolo proximum: "A procuratore dolum et omnem culpam...

(e) Spondet diligentiam et industriam negotio gerendo parent

Arguably, then, Justinian's compilers interfered much less with the classical texts than has often been supposed; on the whole, they seem to have been happy to preserve the flexible interpretation of the classical Roman lawyers.^[2177] Such an attitude, of course, made it difficult in subsequent centuries to reconcile the texts on the basis of a unitary formula. Thus, there have always been attempts to show that liability was (or ought to have been) for dolus (and culpa lata).^[2178] By and large, however, when it came to the question of generalizing a standard of liability, Ulp. D. 50, 17, 23, the (not so) "ill-directed tirades of young Cicero"^{[2179] [2180] and the fragments dealing with procuratio carried the day. Even where mandatum was (or is) still regarded as gratuitous (as under the BGB), the scope of liability was usually widened to comprise culpa. [2181]-13° This has often been justified, dogmatically, on the basis of an implied promise, on the part of the mandatarius, to exercise all the care and skill required by the business that he had undertaken to carry out: "Spondet diligentiam et industriam negotio gerendo parem";[2182] or, in the words of Donellus:}

"Quippe negotium aliquod gerendum suscipimus, quia geri quidquam sine industria et facultate gerendi non potest: ideo et hac ipsa industria ac facilitate et profiteri et spondere intellegitur."^[2183]

6.