1. See Glanvil, , The Treatise on the Laws and Customs of the Realm of England Commonly Called Glanvill, ed. Hall, G.D.C., Tractatus de Legibus et Consuetudinibus Regni Anglie (London, 1955) X. 18Google Scholar; see, e.g., Simpson, A.W.B., A History of the Common Law of Contract (Oxford, 1975) 4Google Scholar.

2. But there are exceptions.

3. For the argument see, e.g., Watson, Alan, Legal Transplants (Edinburgh, 1974) 15Google Scholar.

4. Mancipatio was a formal ceremony needed to transfer certain important kinds of property, and its obligational content was an inherent warranty against the eviction of the transferee from the property. Nexum is obscure, is probably a variant form of mancipatio, and it involved a creditor having real rights over the person of the nexus: see, e.g., Kaser, Max, Das römische Privatrecht I, 2d ed. (Munich, 1971) 165ff.Google Scholar; Watson, Alan, Rome of the XII Tables (Princeton, 1975) 11 ff., 134 ff.Google Scholar; Diósdi, György, Contract in Roman Law (Budapest, 1981) 30 ffGoogle Scholar. It is Diósdi who would add in iure cessio as involving an obligation. It was a fictional law suit to effect the transfer of ownership in which the defendant, the owner, put up no defense to a claim of ownership from the plaintiff, the transferee. None of these three institutions had a major impact on the later development of the law of contract.

5. See, e.g., Thomas, J.A.C., A Textbook of Roman Law (Amsterdam, 1976) 226Google Scholar.

6. Rhetorica ad Herennium, 2.13.19.

7. Also known as the sponsio.

8. See e.g., Kaser, Max, Das altrömische Ius (Göttingen, 1949) 256 ff.Google Scholar; van den Brink, H., Ius Fasque, Opmerkungen over de Dualiteit van het archäisch-romeins Recht (Amsterdam, 1968) 172 ffGoogle Scholar; Behrends, Okko, Der Zwölftafelprozess (Göttingen, 1974) 35–36Google Scholar; and the authors they cite.

9. G. 4.17a.

10. See, e.g., Kaser, Privatrecht I, supra note 4, 168ff.

11. G. 4.19.

12. See e.g., Kaser, Privatrecht I, supra note 4, 170–71.

13. See for the argument, Watson, Alan, Roman Private Law around 200 B.C. (Edinburgh, 1971) 126–27Google Scholar.

14. The condictio furtiva which is exceptional need not concern us here.

15. See e.g., Kaser, Privatrecht I, supra note 4, 492–93.

16. In French law any noncommercial (in the technical sense) transaction above a very small amount can be proved only by a notarial act or a private signed writing except, under art. 1348 of the Code civil, when it is not possible for the creditor to procure writing. ‘Possible’ here refers to moral possibility as well as physical, and in certain close relationships, such as those involving one's mother, mistress or physician, the obtaining of a writing is regarded as morally impossible.

17. Some scholars, for instance Kaser, Ius, supra note 8, 286, suggest that a real action, the legis actio sacramento in rem, was available for mutuum before the introduction of the condictio. There is no evidence for this, and the availability of such an action would make it more difficult to explain the introduction of the condictio. But the suggestion would not adversely affect the idea expressed here that mutuum was given specific protection because the arrangement was among friends and stipulatio was morally inappropriate. At whatever date, a commercial loan would involve interest, a stipulatio would be taken, and there would be no need for specific legal protection of mutuum.

18. Daube, David, ‘Money and Justiciability,’ Zeitschrift der Savigny-Stiftung (rom. Abt.) (1979) 1ffGoogle Scholar.

19. Ibid. 11. See earlier Daube, David, ‘The Self-Understood in Legal History,’ Judicial Review 18 (1973) 120, 129–30Google Scholar.

20. Collatio 10.7.11. The action has often been thought to be something other than an action for deposit or to be an action for what was later called depositum miserabile but see now, e.g., Watson, Roman Private Law, supra note 13, 151; Kaser, Privatrecht I, supra note 4, 160 n.49.

21. See, e.g., Watson, Roman Private Law, supra note 13, 157; Kaser, Privatrecht I, supra note 4, 160.

22. E.g., ibid. 160.

23. See, e.g., Lenel, Otto, Das Edictum Perpetuum, 3d ed. (Leipzig, 1927) 288–89Google Scholar.

24. The literature is enormous, but see, e.g., Litewski, W., ‘Studien zum sogenannten “depositum necessarium,”’ Studia et Documenta Historiae et Iuris 43 (1977) 188ffGoogle Scholar, especially at 194ff., and the works he cites.

25. Cf., e.g., Plautus, Bacchides, 306.

26. See, e.g., Diósdi, Contract, supra note 4, 44–45.

27. See Watson, Alan, Law of Obligations in the Later Roman Republic (Oxford, 1965) 40ffGoogle Scholar.

28. See, e.g., Kaser, Privatrecht I, supra note 4, 546; Jolowicz, Herbert F. and Nicholas, Barry, Historical Introduction to the Study of Roman Law, 3d ed. (Cambridge, 1972) 288ff.Google Scholar, and the works they cite.

29. Mommsen, Theodor, ‘Die römischen Anfänge von Kauf and Miethe,’ Zeitschrift der Savigny-Stiftung 6 (rom. Abt.) (1885) 260ffGoogle Scholar.

30. Scholars who take any one of these approaches also wish to give a central role in the invention to the peregrine praetor. This seems to me to be unnecessary, but the point need not detain us here: see Watson, Alan, Law Making in the Later Roman Republic (Oxford, 1974) 63ffGoogle Scholar.

31. This appears even in Mommsen, ‘Anfänge,’ supra note 29, 260; see also, e.g., Bekker, Ernst Immanuel, Die Aktionen des römischen Privatrechts I (Berlin, 1871) 156ff.Google Scholar; Arangio-Ruiz, Vincenzo, La Compravendita in diritto romano I, 2d ed. (Naples, 1956) 57ffGoogle Scholar. Diósdi objects, asking why it would be necessary to cut up ‘the uniform contract of spot transactions into two separate contracts, to confirm the two promises with a stipulatio, then abandon the stipulationes shortly so that at the beginning of the preclassical age the contract appears as already in its classical shape.’ Diósdi, Contract, supra note 4, 45. By spot transaction, he appears to have mancipatio in mind. There are two flaws in this argument. First, the object of the sale-type transaction would not always be a res mancipi, in which case mancipatio would be inappropriate, Secondly, even in the earliest times, even when the object was a res mancipi, the parties would not always want a spot transaction, but delivery at a later time, and mancipatio would not then be used.

32. Watson, Alan, ‘The Origins of Consensual Sale: a Hypothesis,’ Tijdschrift voor Rechtsgeschiedenis 32 (1964) 245ffGoogle Scholar.

33. In fact, the stipulatio could not be taken from a son or slave with full protection until the introduction of the actio quod iussu. That action appears to be based on an edict of the praetor (Lenel, Edictum, supra note 23, 278) and actions based on an edictal clause which gives the plaintiff a new right of action cannot be safely dated back beyond around 100 B.C.; see Watson, Law Making, supra note 30, 38.

34. Nicholas does not agree, and suggests for the persistence of the stipulations that they imposed strict liability whereas liability on sale would be based only on good faith: Jolowicz and Nicholas, Introduction, supra note 28, 289 n.8 (at p. 290). This does not address the problem, which is not the continued use of stipulatio but the absence of implied warranties in sale. Those who wanted strict liability could still have demanded a stipulatio even if emptio venditio had implied warranties. Again, this approach does not lessen the commercial inconvenience of the lack of implied warranties. Moreover, it must be surprising in a contract of sale which is based on good faith that there is no warranty of title or of quiet possession.

35. Strict textual proof is lacking, but a development from the strict law stipulatio to good faith emptio venditio can have been no other.

36. For this, see, Watson, Alan, The Making of the Civil Law (Cambridge, Mass., 1981) 14ffGoogle Scholar.

37. de re rustica, 2.2.4; 2.3.4; 2.4.5.

38. The impact of the defects in early consensual sale would be less noticeable, of course, where what was sold was a res mancipi and it actually was delivered by mancipatio, which did have an inherent warranty against eviction. Even here, however, there was no warranty against latent defects.

39. For views see, e.g., Jolowicz and Nicholas, Introduction, supra note 28, 294ff. Significantly, one recent writer on ancient hire, H. Kaufman, offers no view on the origins of the consensual contract: Kaufman, H., Die altrömische Miete (Cologne, 1964)Google Scholar.

40. Actually, locatio conductio is so obviously a residual category—every bilateral transaction involving a money prestation that is not sale is hire—that one need not start with the assumption of the priority of sale. From the very fact of the residual nature of hire one can deduce the priority of sale. Unless, that is, one were to argue (as I think no one would), that originally sale transactions were within the sphere of locatio conductio and that emptio venditio was carved out of the all-embracing contract.

Perhaps it should be expressly stated that the three gratuitous contracts just mentioned do not correspond to the three major types of locatio.

41. Cato, de agri cultura 149.

42. Rhetorica ad Herennium, 2.13.19: see, e.g., Watson, Alan, Contract of Mandate in Roman Law (Oxford, 1961) 22Google Scholar.

43. See, e.g., Visky, Kàroly, Geistige Arbeit und die Artes Liberales in den Quellen des römischen Rechts (Budapest, 1977) 146ffGoogle Scholar.

44. For the dating, see Watson, Law Making, supra note 30, 31ff, especially 38.

45. See, e.g., Lenel, Edictum, supra note 23, 254ff., who, however, thinks there was such an action; and Kaser, Privatrecht I, supra note 4, 537, who apparently tends to think there was not.

46. See Watson, Obligations, supra note 27, 182ff.

47. See, e.g., D.13.7.9 pr; 13.6.16.1

48. Though the actio quod iussu is not evidenced for the Republic. See Watson, Obligations, supra note 27, 187–88.

49. A further reason for the introduction of the new contractual action was that it could allow more of a role for reliance on good faith even though the praetorian action did not have a condemnation clause framed ex fide bona. In favor of this explanation is the fact that fiducia—the older form of real security (and not contractual in terms of the definition given at the beginning of this paper)—was erected by using mancipatio with a special clause relating to trust and faith: see Watson, Obligations, supra note 27, 172ff. Indeed, it is possible that the existence of fiducia was influential by way of analogy for the creation of pignus. Fiducia had two limitations: its dependence on mancipatio meant that only res mancipi could be so pledged (unless the cumbrous in iure cessio were used) and that only citizens (or those with commercium) could be creditors or debtors. The praetor might thus have introduced the very different contract of pignus, also because of the difficulties involved in framing stipulation that would adequately cover the debtor's rights.

50. Cicero, de officiis, 3.58.

51. See, e.g., Thomas, Textbook, supra note 5, 267ff.

52. See, e.g., Watson, Obligations, supra note 27, 21ff.

53. G. 3.154a.

54. G. 3.154b.

55. See above all Watson, Alan, ‘Consensual societas between Romans and the Introduction of formulae,’ Revue Internationale des Droits de l'Antiquité 9 (1962) 431ffGoogle Scholar.

56. G. 3.149.

57. D. 17.2.29pr., I: see for the argument, Watson, Alan, ‘The Notion of Equivalence of Contractual Obligation and Classical Roman Partnership,’ Law Quarterly Review 97 (1981) 275ffGoogle Scholar.

58. Laesio enormis is post-classical, whether it is to be attributed to Diocletian or Justinian: C.4.44.2; 4.44.8.

59. G. 3.159; D.17.1.12.16. That damages were doubled for breach in depositum miserabile is not a problem. That could still be subjected to special regulation.

60. D.16.3.12, 13, 14.

61. D.45.1.122; 45.1.126.2; 45.1.140pr.

62. There is something illogical about accepting a written document as evidence of a stipulatio. It can show the intention of the parties, but scarcely that they went through the formalities.

63. See, e.g., MacDowell, Douglas M., The Law in Classical Athens (Ithaca, 1978) 233Google Scholar.

64. G. 3.134.

65. See, e.g., Crawford, Michael H., Roman Republican Coinage (Cambridge, 1976) 35ffGoogle Scholar.

66. The state of development of barter before the time of Justinian is very obscure, much disputed, and need not be gone into here. For literature, see e.g., Thomas, Textbook, supra note 5, 312–13; and Kaser, Privatrecht I, supra note 4, 381.

67. G. 3.141; J.3.23.1; D.19.4.1 pr.

68. For the argument see Daube, David, ‘Three Questions from Homer in D.18.1.1.1,’ Cambridge Law Journal 10 (1949) 213ffCrossRefGoogle Scholar.

69. A relatively satisfactory outcome, I believe, from the Sabinian viewpoint would be that barter is sale, and both parties have the obligation of sellers.

70. Daube, ‘Money,’ supra note 18, 8.

71. Ibid. 9.

72. See the texts collected in D.19.5.

73. D.19.5.5.pr.

74. For these see Watson, Obligations, supra note 27, 257.