Hostname: page-component-cd9895bd7-q99xh Total loading time: 0 Render date: 2024-12-27T06:03:54.095Z Has data issue: false hasContentIssue false
  • English
  • Français

When is a Sale Not a Sale? The Riddle of Athenian Terminology for Real Security Revisited

Published online by Cambridge University Press:  11 February 2009

Edward M. Harris
Edward M. Harris
Affiliation:
Brooklyn College and the Graduate Center, The City University of New York
Get access Rights & Permissions [Opens in a new window]

Extract

In Athens during the late Classical and Hellenistic periods, it was customary for a man who was borrowing a large sum of money to pledge some property as security for the repayment of his loan. To show that this property was legally encumbered, a flat slab of stone, called a horos, was set up, and an inscription, indicating the nature of the lien on the property, was inscribed on the horos. These horoi served to warn third parties that the man who pledged the property as security was not free to sell it or otherwise alienate it until the loan was repaid. The terminology which is used on these horoi to indicate that the property has been pledged as security varies. On a relatively small number of horoi, only seven, the property is described as ‘lying under (an obligation)’ (ὑποκειμένου, -ης, -ων) for a debt, the amount of which may or may not be specified. The texts found on a far greater number of horoi, some 128 in all, use a different type of expression. On these horoi, the property is said to have been ‘sold on condition of release’ (πεπραμένου, -ης, -ων, ἐπὶ λύσει). The terminology used on the horoi to describe this kind of lien presents a striking contrast with that employed by the Attic orators: in their speeches we find the verbs ὑποκεῖσθαι and ὑποτιθέναι when it is a question of pledging security for a loan, but never πέπρασθαι with the addition of the prepositional phrase ἐπὶ λύσει.

Type
Research Article
Information
The Classical Quarterly , Volume 38 , Issue 2 , December 1988 , pp. 351 - 381
Copyright
Copyright © The Classical Association 1988

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 For a description of the physical characteristics of the horoi, see Fine, J. V. A., Horoi: Studies in Mortgage, Real Security and Land Tenure in Ancient Athens (Hesperia Supp. 9 [1951]), pp. 44–6Google Scholar. This work will hereafter be cited by the author's name only.

2 For the purpose of this kind of horos, see Fine, , pp. 42–3Google Scholar; Finley, M. I., Studies in Land and Credit in Ancient Athens, 500–200 B.C. (New Brunswick, 1952), pp. 10–21Google Scholar. Finley's work was reprinted by Rutgers University Press in 1985 with a new introduction by P. Millett which contains the texts of all the horoi published since the appearance of Finley's work. All references in this article will be to the 1985 edition which will be cited by the author's name only. Milieu's introduction to this edition will be cited as ‘Millett in Finley’. When citing the texts of the horoi, I will follow the system devised by Finley and followed by Millett (e.g. horos no. 34, horoi nos. 6, 8, 10).

3 Horoi nos. 1, 2, 2A, 4, 5, 6, 7. The term is restored on horos no. 3A. Since horos no. 3 does not actually contain the actual term ὑποκειμένης, I have excluded it from this category although Finley placed it under the rubric hypotheke. Nor have I included horoi nos. 80A and 81A – on these, see the discussion in Section II. The term is also found on a horos from Amorgos (horos no. 8) and another from Lemnos (horos no. 10).

4 This sum includes all those texts in which the formula is found or reasonably restored. Following the lead of Finley and Millett, I have also included those which omit the phrase ἐπὶ λύσει. There are 92 in Appendix I of Finley, 22 in Appendix III of Finley, and 14 in Millett in Finley. Unlike Millett, I have excluded horoi nos. 80A and 81 A. For discussion of these, see Section II.

5 For the translation, see Finley, p. 31.

6 While it is true that various forms of the verb πέπρασθαι are used to describe transactions involving real security in Dem. 37, they are nowhere in that speech coupled with the phrase ἐπὶ λύσει. For a discussion of the terminology used in this speech, see Section IV.

7 The main proponent of this view was Günter, E., Die Sicherungsübereignung im griechischem Recht (diss. Konigsberg, 1914), pp. 16–25Google Scholar (see p. 17 n. 3 and p. 18 n. I for references to earlier views on the subject).

8 This view was fully presented by Pappulias, D. P., Real Security in Greek and Roman Law (in modern Greek) (Leipzig, 1909), Part IGoogle Scholar. Pappulias' view was thoroughly criticized by Raape, L., Der Verfall des griechischen Pfandes (Halle, 1912), pp. 1–48Google Scholar and by Manigk, A., Phil. Woch. 34 (1914), 205–13Google Scholar.

9 Millett in Finley, p. xiii delivers some telling criticisms of Fine's theory but does not refute it in detail. Cf. Gernet, L., ‘Horoi’, in Studi in onore di Ugo Enrico Paoli (Florence, 1955), pp. 345–53Google Scholar (‘Il semble vraiment difficile d'accepter une interprétation aussi compliquée, aussi conjecturale, qui postule un concept de la vente singulièrement affiné, et qui utilise d'un bout à l'autre une notion aussi suspecte, pour une époque très ancienne, que celle de fiction juridique.’). In my review of recent opinions I have not treated the views of Germain, L. R. F., ‘Les horoi: exposé de la méthode littérale et des résultats obtenus’, Symposium 1971 (Vienna, 1975), pp. 333–46Google ScholarPubMed and La publicité des sûretés foncières dans la Grèce classique et hellénistique (diss. Paris, 1968)Google Scholar. These are well criticized by Millett in Finley, pp. xix–xxi.

10 The view that in hypotheke the debtor retained both possession and ownership goes back to Hitzig, H. F., Das griechische Pfandrecht (Munich, 1895), p. 1Google Scholar and was endorsed by Beauchet, L., Histoire du droit privée de la République athénienne (Paris, 1897, II, pp. 176–80Google Scholar; Lipsius, J. H., Das attische Recht und Rechtsverfahren (Leipzig, 1905–1915), pp. 690–2Google Scholar; Harrison, A. R. W., The Law of Athens: Law of Property (Oxford, 1968), p. 258Google Scholar.

11 Cf. Harpocration s.v. ἀποτιμηταί ταὶ ἀποτίμημα καὶ ἀποτιμᾶν: εἰώθεσαν δὲ καὶ οἱ τότε, εἰ γυναικὶ γαμουμένῃ προῖκα διδοῖεν οἱ προσήκοντες, αἰτεῖν παρὰ τοῦ ἀνδρὸς ὥσπερ ἐνέχυρόν τι τῆς προικὸς ἅξιον, οἷον οἰκίαν ἤ χωρίον.

12 See Dem. 41.7 with Finley, p. 53 and p. 245 n. 61. Finley's account of dotal apotimema is superior to that of Fine.

13 Horoi nos. 11, 13, 19, 22, 32, 35, 41, 46, 97, 72A, 73A, 95A. This was pointed out by Millett in Finley, p. xiii.

14 The inscription was published by Crosby, M. in Hesperia 10 (1941), 14, no. 1CrossRefGoogle Scholar.

15 The phrase ἐπὶ λύσει is omitted in the text of the inscription, but for the sake of argument I have followed previous scholars who have interpreted the claim as arising from a prasis epi lysei.

16 Crosby, art cit. (n. 14), p. 22.

17 Finley, , ‘Multiple Charges on Real Property in Athenian Law: New Evidence from an Agora Inscription’ in Studi in onore di Vincenzo Arangio-Ruiz (Naples, 1953), p. 491 n. 45Google Scholar.

18 See Ferguson, W. S., ‘The Attic Orgeones’, HThR 37 (1944), 70Google Scholar (for his criticism of Crosby's suggestion, see p. 83).

19 For further criticisms of Fine's interpretation of this inscription, see Finley loc. cit. (n. 17).

20 Fine, p. 93. Fine did not offer any evidence to back up these assertions, both of which are questionable.

21 The verb is used to describe hypothecation in IG ii2 43, line 40 (378/7) and is used by Isocrates (21.2) when describing how a certain Nicias hypothecated his house during the reign of the Thirty. Herodotus (2.136) also uses the verb ὑποτιθέναι when describing the Egyptian practice of pledging the corpse of one's father as security for a loan. On the other hand, the earliest dated horoi which use the language of prasis epi lysei come from the archonship of Praxibulus (315/14): horoi nos. 11, 14, 27, 76, 85. Kirchner, (IG ii 22724)Google Scholar restored the name of the archon in horos no. 13 as Apollodorus (319/18). Dow, S. and Travis, A. H., ‘Demetrios of Phaleron and his Lawgiving’, Hesperia 12 (1943), 162–3CrossRefGoogle Scholar, proposed Nicodorus.

22 Although Fine's own theory is not convincing, his work does contain valuable criticisms of the views of Paoli and Meletopoulos.

23 Manigk, A., ‘Hyperocha’, RE 9 (1916), cols. 292–321Google Scholar.

24 Finley's observation was based on the horoi known to him at the time. The discovery of new horoi has not affected it. See Millett in Finley, p. xiii.

25 Finley's translation of συνθῆκαι as ‘written contract’ is strictly speaking incorrect. The word refers only to a document which recorded the terms of an agreement and could be used to prove the eixstence of an agreement. For an analysis of the term, see Kussmaul, P., Synthekai: Beiträge zur Geschichte des attischen Obligationenrecht (diss. Basel, 1969), 15–20Google Scholar.

26 E.g. horoi nos. 1, 23, 45, 51, 52, 61, 62, 72, 75, 78.

27 Although he was aware of the problem, Finley (pp. 30–1) did not offer a solution to it. Millett in Finley, p. xiv, tries to explain ‘the puzzling divergence between the horoi and the Orators on the phraseology of security obligations’ with a theory about the use, or rather misuse, of terminology for hypothecation. He begins by citing Finley's observation (p. 8) about the lack of precision in Athenian legal terminology, then goes on to claim that the ‘Orators were not primarily concerned with the technical accuracy of their language; instead they were intent on giving a plausible (though possibly misleading) argument’. Finley's observation is sound, but it is irrelevant to Millett's point here. Finley was talking about lack of precision in Athenian legal terminology which did not possess an elaborate set of subtle distinctions. Millett wishes to make a point about lack of accuracy, a very different matter (one can be imprecise without being inaccurate).

Millett continues by asserting that the desire of the orators to produce a persuasive argument ‘would naturally lead them to avoid the ambiguous terminology of prasis epi lysei with its implications of sale rather than security in favour of the more straightforward vocabulary of hypothecation’. This is implausible – if the orators ‘were intent on giving a plausible and persuasive (though possibly misleading) argument’, one would expect them to have opted for ‘the ambiguous terminology of prasis epi lysei’ with its greater potential for obfuscation in preference to ‘the more straightforward vocabulary of hypothecation’. Millett next turns to consider the horoi and states that there is ‘no reason to expect the horos inscriptions to reflect with any great precision the nature of the transactions that lay behind them’. Drawing attention to the fact that ‘the prime function of the horoi was to warn third parties that the property was somehow encumbered’, Millett came to the rather startling conclusion that it ‘mattered little whether the stone said hypotheke, prasis epi lysei, or even apotimema’. This is not plausible either. As Finley (p. 19) correctly states, ‘Whatever information the stones did give had to be accurate and up-to-date if the function of public notice were to be properly served.’

28 ‘The Alienability of Land in Ancient Greece’, Eirene 7 (1968), 25–32Google Scholar. Cf. ‘Homer and Mycenae: Property and Tenure’, Historia 6 (1957), 133–59Google Scholar where it is maintained that in the Homeric poems the property regime is ‘one of private ownership’ and that there was ‘free, untrammelled right to dispose of all movable wealth’. By pointing out that land was regularly transferred by inheritance, Finley implies that this also held true for immovable wealth.

29 The following summary of Roman practices in regard to real security derives from Schulz, F., Classical Roman Law (Oxford, 1951), pp. 406–10Google Scholar. This work will hereafter be cited by the author's name only.

30 This is an example of ‘semantic marking’. For a discussion of this phenomenon, see Lyons, J., Semantics 1 (Cambridge, 1977), pp. 307–8Google Scholar.

31 In this case it is clear that the debtor Timotheus remained in possession. See Fine, pp. 67–9.

32 The use of these two different terms cannot be explained on the assumption that one arose from a security obligation whereas the other did not. See Finley, , art. cit. (n. 17), p. 474 n. 4Google Scholar.

33 E.g. Hitzig, , op. cit. (n. 10), pp. 1–3Google Scholar (prasis epi lysei is equated with fiducia cum creditore); Szanto, E., ‘Hypothek und Scheinkauf im griechischem Recht’, WS 9 (1897), p. 279Google Scholar; Fine, p. 61 n. 3.

34 For a recent critique of attempts to explain Greek laws in terms of concepts drawn from Roman law, see Cohen, D., Theft in Athenian Law ( = Münchener Beiträge zur Papyrusforschung und Rechtsgeschichte 74; Munich, 1983), pp. 5–7Google Scholar.

35 For mancipatio and cessio in iure, see Schulz, , pp. 344–9Google Scholar. Formal procedures for conveyance also allowed Roman law to distinguish between two forms of loan, the mutuum and the commodalum (Schulz, 508)Google Scholar. Such a distinction did not exist in Athenian law; see Wolff, H. J., ‘Die Grundlagen des griechischen Vertragsrecht’, ZSS 74 (1957), 49–50Google Scholar.

36 Schulz, pp. 406–7. Pignus was later protected by the actio Serviana.

37 For the principle of the cash sale, see Pringsheim, F., The Greek Law of Sale (Weimar, 1950), pp. 179–219Google Scholar.

38 Awareness of the distinction can be seen in various texts (e.g. [Dem.] 7.26). See Pringsheim, , op. cit. (n. 37), pp. 9–13Google Scholar; Kaser, M., ‘Der altgriechische Eigentumsschutz’, ZSS 64 (1944), 136–7Google Scholar.

39 Finley, pp. 113, 296–7.

40 My approach is similar to that of Cohen, , op. cit. (n. 34), pp. 6–7Google Scholar. Cohen, however, fails to appreciate the difference between ordinary language and legal language and in places appears to assume that words in common usage are used with the same precision and consistency as they are in statutes which often begin with legal definitions of the terms they employ. For instance, his attempt to draw a clear-cut distinction between the verbs κλέπτειν and ἀποστερεῖν breaks down in the face of passages which refer to the embezzlement of public funds, as Cohen (pp. 30–3) himself recognizes. For a shrewd assessment of the differences between ordinary language and legal language, see Austin, J. L., ‘A Plea for Excuses’, in Philosophical Papers (Oxford, 1961), pp. 132–7Google Scholar.

41 E.g. Dem. 37.4; [Dem.] 56.3 and the passages listed in n. 43. The same verb is used in the middle with the meaning ‘to borrow from’: [Dem.] 56.3, 6.

42 [Dem.] 49.12; 56.1; Dem. 35.9.

43 Dem. 32.12, 14; 36.6, 18; 37.50; [Dem.] 49.53. For the term ἐπιδάνειζειν which appears to have had the same meaning in certain cases, see Finley, p. 297.

44 For the active, see Dem. 28.17–18; [Dem.] 49.12, 52; Isoc. 21.2; Lys. 19.25. For the middle, see Dem. 28.18; [Dem.] 49.51 (the verb is mistranslated by A. T. Murray in his Loeb edition of the speech); 50.55; IG ii2 43, line 40. Sometimes the simplex takes the place of the compound: [Dem.] 53.10, 12, 13; Is. 5.21; horos no. 102, line 12. In IG ii2 43, lines 40–2 we find the compound followed by the simplex in the next line. This is an example of a syntactic phenomenon which has been noted by several scholars. See Watkins, C., ‘An Indo-European Construction in Greek and Latin’, HSCP 71 (1967), 115–19Google Scholar.

45 Dem. 33.10; [Dem.] 49.2, 52, 53; 56.3.

46 [Dem.] 49.11; horoi nos. 1, 2, 4, 5, 6, 7.

47 Dem. 21.11; 33.6; 45.70. Cf. Pollux 3.85.

48 Dem. 27.26; 37.7; [Dem.] 35.25. Note also the verb ἐνεχυράζειν at Ar, . Nub. 241Google Scholar; Ec. 567 with Finley, pp. 222–3 n. 6.

49 For the genitive denoting ownership as opposed to mere possession, see Kaser, loc. tit. (n. 38). Note also that Nicobulus calls the property that was pledged by Pantaenetus to Evergus and himself ‘ours’ (Dem. 37.7, 9, 29).

50 Note also that the noun ἔγκτησις which is formed from the same roots means ‘the right to own land’. For a study of this term, see Pečirka, J., The Formula for the Grant of Enktesis in Attic Inscriptions (Prague, 1966)Google Scholar.

51 Finley, p. 263. If the author of the decree had meant to refer to the acquisition of property by means of foreclosure, he would surely have used one of the verbs normally used for that procedure.

52 For the name of the debtor, see Dem. 27.27.

53 For the traditional analysis of the problem posed by this passage, see Davies, J. K., Athenian Propertied Families, 600–300 B.C. (Oxford, 1971), pp. 129–30Google Scholar.

54 Finley, p. 116 (‘Yet Demosthenes … does not think of re-calculating the value of the slaves in market-terms.’). Finley curiously ignores the additional loan of 500 drachmai made on the security of the slaves (Dem. 27.27). That Moerades was able to contract this additional loan on the slaves from Aphobus clearly indicates that foreclosure had not taken place before the death of Demosthenes' father. One cannot, therefore, argue that Demosthenes regards Moerades' slaves as part of his father's property because he had obtained them through foreclosure.

55 If we calculate the market value of the slaves in the furniture shop at about 65–70 mnai, the annual income of 12 mnai would represent an annual return on investment of about 18% which is comparable to that received from the slaves in the cutlery shop (30 mnai a year on an investment of about 160–5 mnai). I do not think that the Athenians actually calculated such tilings with any precision. Nevertheless, we would expect the price of slaves to be proportional to the value of goods they were, able to produce. For the prices of slaves varying in relation to the value of their skills, see Meiggs, R. and Lewis, D. M., A Selection of Greek Historical Inscriptions to the End of the Fifth Century (Oxford, 1969), p. 247Google Scholar.

56 For securities worth twice as much as the amount of the loan, see [Dem.] 34.6; 35.18; Dem. 37.4, 31.

57 For Demosthenes' victory, see Dem. 30.2. To judge from Demosthenes' reply to Aphobus' defence (Dem. 28), it does not appear that Aphobus disputed this part of Demosthenes' calculation of the amount owed to him.

58 ἔχει θέμενος is a perfect periphrastic and means ‘has received as security’, ἔχει is only an auxiliary verb and has no independent meaning. For the construction, see Aerts, W. J., Periphrastica (Amsterdam, 1965), pp. 128–60Google Scholar. Paoli, U. E., ‘Ipoteca e ἀποτίμημα nel diritto attico’, in Studi di diritto attico (Florence, 1930), pp. 153–4Google Scholar, did not recognize the construction and misconstrued the force of ἔχει which he took to imply that the creditor had possession. Finley, p. 205 made the same mistake and asserted incorrectly that the inscription indicated ‘that there had been default …’.

59 For a decisive refutation of the view that the legal practices of the Greek city-states were roughly similar, see Finley, M. I., ‘The Problem of the Unity of Greek Law’, in Atti del Io Congresso Internazionale delta Società del Diritto (Florence, 1966), pp. 129–44Google Scholar ( = The Use and Abuse of History (New York, 1975), pp. 134–52Google Scholar).

60 For the accessory character of real security, see Schulz, pp. 420–2.

61 For the custom of calling the agreement between the creditor and the debtor who remained in possession of the security a lease, see Section IV. Note also the use of the word μίσθωμα on horos no. 102, line 15, where we would expect to find the word τόκος.

62 I do not understand how Finley, , art. cit. (n. 17), p. 481Google Scholar, can state that this inscription ‘offers the first unequivocal instance of collateral security in Athens’. To provide an ‘unequivocal instance of collateral security’ we would have to come upon a case where the creditor foreclosed on a defaulting debtor and was required either by law or by contract to sell the security and to return to the debtor the difference between the sale price and the amount of the loan. But all the inscription tells us is that the poletai confiscated the house, paid off the debts owed by Theosebes, and then sold the house. It deals with confiscation, not foreclosure, and is thus not relevant to the issue of the existence of collateral security.

63 The creditor who delivered the speech Against Apaturius was also of the same opinion ([Dem.] 33.8). For other creditors who regarded themselves as the owners of the security see Dem. 37 with the discussion in Section IV.

64 The law recorded on IG ii2 43 considers the creditors the owners of the security for the sake of completeness; it is trying to cover all possible ways of acquiring ownership.

65 My interpretation of this speech owes much to the fine analysis of Meyer-Laurin, H., Gesetz und Billigkeit im attischen Prozess (Weimar, 1965), pp. 12–15Google Scholar.

66 Of course, Socrates and Plato viewed the problem somewhat differently. See Robinson, R., Plato's Earlier Dialectic 2 (Oxford, 1953), pp. 49–53Google Scholar.

67 I do not consider it anachronistic to speak of ‘businessmen’ and ‘entrepreneurs’ in Classical Athens. For the presence of ‘economic activities of a capitalist kind’ in the Ancient World, see the sensible comments of Goody, J., The Logic of Writing and the Organization of Society (Cambridge, 1986), pp. 177–84CrossRefGoogle Scholar. For Athens in particular, see Thompson, W., ‘The Athenian Entrepreneur’, AC 51 (1982), 53–85Google Scholar.

68 There have been several discussions of this speech. Among the more recent are Harrison, , op. cit. (note 10), pp. 274–9Google Scholar; Finley, pp. 32–5; Fine, pp. 146–50. Carey, C. and Reid, R. A., Demosthenes: Selected Private Speeches (Cambridge, 1985), pp. 105–59Google Scholar provide a very detailed and generally helpful commentary on the speech, but make no original contribution to the discussion of problems involving real security. On most issues they tend to follow Finley. Finley's analysis of the speech is flawed by his refusal to accept the implications of Nicobulus' use of the language of sale. Finley, p. 224 n. 11, tries to justify his interpretation of the use of the language of sale in contexts dealing with hypothecation by appealing to a passage from Pollux, , Onomasticon 8.142Google Scholar ( = Hyperides fr. 193 Blass): Ὑπερείδης δὲ ἐν τῷ πρὸς Χάρητα ἔϕη ἀποδόμενος ἀντὶ τοῦ ὑποθείς. Finley claims that in this entry Pollux ‘had noted the use of ἀποδόδωμι without the appended ἐπὶ λύσει, and indicated that in that passage it meant “hypothecate”, not “sell”.’ That is not what Pollux indicates at all. The lexicographer merely indicates that in the passage in question Hyperides used the word ἀποδόμενος instead of (ἀντὶ) the word ὑποθείς. All this means is that in this passage the word ἀποδόμενος is a virtual synonym of ὑποθείς. Nowhere in the entry does Pollux deny that the verb means ‘sell’ or that it carries the implications associated with ownership. (I owe this observation to Professor M. Dilts.)

69 Carey, and Reid, , op. cit. (n. 68), pp. 114–17Google Scholar, fail to realize that the Athenians, like the Romans, did not share our modern notion of legal personality and thus had nothing akin to the modern juristic concept of the corporation. (On this topic, see Finley, p. 275 n. 5, with the literature cited there.) As a result, their discussion of the relationship between Evergus an d Nicobulus is confused and misleading.

70 The position of Telemachus (5) can be explained in either of two ways. The first, that of Fine, p. 147, is that Telemachus had been the owner in the sense that he had lent money to Pantaenetus and had accepted the workshop and slaves as security. He then ‘sold’ them to Mnesicles and his associates in the sense that Mnesicles and his associates had paid Telemachus the amount owed by Pantaenetus and thereby became Pantaenetus' creditors and the owners (at least in their own eyes) of the security. The other view, maintained by Finley, p. 32 and Harrison, , op. cit. (n. 10), p. 275Google Scholar, is that Telemachus was the actual owner of the workshop and slaves and sold them to Pantaenetus. ‘Presumably to pay for his purchase, he borrowed 10,500 drachmas, (…). The mill and the slaves were to serve as security for the loan and this was accomplished by having Telemachus, the original owner, “sell “ directly to Mnesikles, the principal creditor’ (Finley, , p. 32)Google Scholar.

71 For the absence of a public land registry in Classical and Hellenistic Athens, see Finley, pp. 13–15.

72 For the procedure, see Kaser, .art. cit. (n. 38), pp. 159–78Google Scholar. For a convenient summary of the evidence in English, see Wyse, W., The Speeches of Isaeus (Cambridge, 1904), pp. 435–7Google Scholar.

73 Harpocration s.v. βεβαιώσεως mentions a δίκη βεβαιώσεως, but there is no contemporary evidence for the action. Even if it did not exist, the buyer whose claim was not supported by the seller could still have proceeded against him with a δίκη βλάβης for failing to testify on his behalf ([Dem.] 49.19–20).

74 Pace Finley, , pp. 228–9 n. 33Google Scholar.

75 Carey, and Reid, , op. cit. (n. 68), p. 126Google Scholar, misunderstand the nature of Mnesicles' declaration when they state ‘Mnesicles confirmed that when he “sold” it the property did not stand as security for these other debts’. The text reads Μνησικλέους βεβαιοῦντος ἡμῖν, Mnesicles supported the claims of Evergus and Nicobulus. The text does not indicate that Mnesicles said anything about the claims of the other creditors.

76 The words διαλῦσαι (12) and συνεχώρησα (13) used by Nicobulus to describe the proposals of the other creditors and his response to them make it clear that they were offering a ὁμολογία or a release. For other examples of these verbs employed in connection with either of these two types of agreements, see Dem. 21.122, 216; 38.24; [Dem.] 41.14; 42.10–14; Lysias 4.1. For the ὁμολογία, see Kussmaul, , op. cit. (n. 25), pp. 30–7Google Scholar, esp. p. 34: ‘Die ὁμολογίαι, (…) sind wahrscheinlich pacta, welche ein bestehendes Rechtsverhältnis abändern oder einen Streit daruber beenden, nicht contractus, welche eine neue Obligation begründen.’ For releases, see Dem. 36.10. 24–5.

77 None of the recent discussions of the speech analyzes the difference between the two proposals offered by the other creditors. Finley, p. 34, misses the point of the revised proposal. According to him, the other creditors wanted Nicobulus and Evergus to become πρατῆρες because ‘Should the creditors (i.e. the other creditors) be compelled to accept the property, the question of title would become important.’ But Nicobulus supplies a very plausible motive for the new proposal: ἑώρων γὰρ ἡμᾶς οἱ' ἐσυκοϕαντούμεθ' ὑπὸ τούτου. They were worried that the same ruse that Pantaenetus had used against Nicobulus and Evergus would be used against them. This sentence does not refer to Pantaenetus' case against Evergus nor does it imply ‘that the case against Evergus was at least under way at the time of the transaction with the claimants’ (Carey, and Reid, , op. cit. (n. 68), p. 129Google Scholar). The verb ἐσυκοϕαντούμεθ' is plural, referring to both Nicobulus and Evergus. It plainly refers to Pantaenetus' deceitful manoeuvres, especially his concealing the presence of the other creditors to Nicobulus and Evergus and his bringing them forward when it was advantageous for him to do so. The verb can be used of any attempt to extort concessions from another person and need not imply legal proceedings (e.g. Lys. 26.24).

78 I imagine that it was Nicobulus alone who became πρατήρ, that is, the seller who would be required to warrant the sale, and was the only one who gained the release from Pantaenetus. It appears that when there were several sellers, one could be designated as responsible to warrant the sale; this seems to have been the case with Mnesicles (5). Such a hypothesis would explain why Evergus was not able to fend off Pantaenetus' prosecution with a παραγραϕή (8–9) in the way in which Nicobulus attempted to.

79 For the actio de dolo malo, see Schulz, , pp. 605–7Google Scholar.

80 For the procedure used in hypotheca, see Schulz, , pp. 423–5Google Scholar.

81 Pace Finley, ch. 8. Cf. the criticism made by Pringsheim, , Gnomon 25 (1953), 224Google Scholar. Even P. Millett, who takes a primitivist view of the Athenian economy, admits that ‘The easy availability of credit was essential to the smooth functioning of Athenian society; loan transactions of one type or another are a pervasive feature at all levels of Athenian life.’ (‘Maritime Loans and the Structure of Credit in Fourth-Century Athens’ in Garnsey, P., Hopkins, K., and Whittaker, C. R., edd., Trade in the Ancient Economy (Berkeley, 1983), p. 42Google Scholar).

82 There is some evidence for this assumption: see IG ii2 1183, lines 28–9.

83 Finley, p. 204 n. 11, denied that the expression indicates ownership and translated it ‘to have and to have power’. He attempted to justify his translation by arguing that ‘κρατεῖν does not necessarily mean “to own “’, and cited several texts in support of his assertion. In two of these passages, [Dem.] 35.25 and 49.11, the verb is used to describe the rights of creditors to a security, but nothing in the surrounding context would enable us to decide whether these rights included ownership or not. At Is. 8.2 the speaker uses the words ἔχουσι…καὶ κρατοῦσι when describing how his opponents were enjoying ownership of Ciron's estate. He is not disputing the fact that they own it, but readily admits it; his charge is that they had obtained ownership through force and that they, therefore, were not entitled to retain ownership. In the passage from Herodotus (2.136) an old Egyptian custom which originated during the reign of King Asychis is described. At this time, we are told, a son was allowed to pledge the corpse of his father as security for a loan. It was also laid down that the creditor ἁπάσης κρατέειν τῆς τοῦ λαμβάνοντος θήκης. Once again nothing in the surrounding context allows us to determine whether or not the rights expressed by the verb κρατέειν included ownership or not. Finley makes no mention of the use of the verb to denote ownership in the law of inheritance preserved at [Dem.] 43.51 (the authenticity of the wording of this part of the statute is guaranteed by the quotation at Is. 7.20) where it occurs as a synonym for κυρίους εἶναι and μοῖραν λαγχάνειν. Nor does he cite Ath. Pol. 56.2 where the phrase ἔχειν καὶ κρατεῖν obviously means ‘to own’. It is found in a description of the oath that the archon swore upon entering office and is clearly a promise not to deprive citizens of the property they own. This promise is equivalent to the assurance found in the heliastic oath not to carry out a redistribution of land or a cancellation of debts (Dem. 24.149).

84 If we adopt such an explanation to account for the language of the horoi, we can view the expression ὑποκειμένης ἐπὶ λύσει (horoi 80A and 81A – both inscriptions are on the same stone) as an idiosyncratic combination of elements from both of the standard formulas for real security. For those who maintain the traditional position that there were two forms of security, this unusual expression poses insoluble difficulties. One might suggest that the creditor who set up the horos made a mistake, but that is unlikely. If these were two forms of security, each of which was referred to by a different expression and bore different legal implications, such an error would have been quite serious, and serious mistakes do not go unnoticed. Yet on this horos the expression not only remained unconnected, but was actually repeated!

86 For written documents in loan transactions, see [Dem.] 33.15; 34.6; 35.1–13; 56.6; horoi nos. 1, 2, 2A, 6, 11, 13, 17, 27, 32, 39, 65.

86 For widespread literacy in Classical Athens, see Harvey, F. D., ‘Literacy in the Athenian Democracy’, REG 79 (1966), 585–635CrossRefGoogle Scholar.

87 For exchange of services see Fuks, A., ‘Kolonos misthios: Labour Exchange in Classical Athens’, Eranos 49 (1951), 171–3Google Scholar.

88 For the Homeric poems as evidence for social conditions in the eighth century B.C., see Morris, I. M., ‘The Use and Abuse of Homer’, CA 5 (1986), 81–136Google Scholar.

89 Scholars are divided over the question whether coinage was introduced in the first or second half of the seventh century, but there is a general consensus that there was no coinage before 700 B.C. For the earlier dating, see Weidauer, L., Probleme der frühen Elektronprägung (Freiburg, 1975), pp. 72–109Google Scholar; Cahn, H., SNR 56 (1977), 279–87Google Scholar; Kagan, D., AJA 86 (1982), 343–60CrossRefGoogle Scholar. For the later dating, see Cook, R. M., JHS 66 (1946), 90–1Google Scholar; Robinson, E. S. G., JHS 71 (1951), 156–64CrossRefGoogle Scholar; Price, M. J., NC 136 (1976), 274–5Google Scholar; Kroll, J. and Waggoner, N., ANA 88 (1984), 325–33Google Scholar.

90 For the date of the introduction of the Greek alphabet, see Jeffery, L. H., The Local Scripts of Archaic Greece (Oxford, 1961), pp. 12–21Google Scholar.

91 Finley, M. I., The World of Odysseus 2 (London, 1978), pp. 74–107Google Scholar.

92 E.g. Il. 1.490; 2.93; 7.382, 414; 11.139; Od. 2.10, 37, 150; 8.109; 16.361; 24.420. See also Finley, , op. cit. (n. 91), p. 78Google Scholar.

93 For the notion of debt, see Hes, . Op. 404, 647Google Scholar; Hom, . Od. 8.353Google Scholar. For seizure of body of the debtor resulting from default, see Ath. Pol. 2.2; 6.1.

94 Gernet, L., Droit et institutions en Grèce antique (Paris, 1968), pp. 15–16Google Scholar.

95 For the role of the polis in the creation of coinage, see Kraay, C. M., ‘Hoards, Small Change, and the Origin of Coinage’, JHS 84 (1964), 76–91CrossRefGoogle Scholar; Wallace, R. W., ‘The Origin of Electrum Coinage’, AJA 91 (1987), 385–97CrossRefGoogle Scholar.

96 The legal status of contracts in Classical Athens is complicated by the fact that the Athenians did not distingish between actions ex delicto and actions ex contractu. The analysis of Pringsheim, , op. cit. (n. 37), pp. 13–85Google Scholar, is vitiated by his assumption that the Athenians did make such a distinction. On the topic of contracts in Athenian law one can consult Wolff, op. cit. (n. 35), but more remains to be done.

97 For the role of the demarch in seizing the property of defaulting debtors, see Ar, . Nub. 37Google Scholar.

98 For the agora as a sacred area, see de Ste Croix, G. E. M., The Origins of the Peloponnesian War (London, 1972), pp. 271–2, 397–8, 399Google Scholar.

99 The volume of commodity exchange was significant enough to merit treatment by no less a person than the philosopher Aristotle; see Meikle, S., ‘Aristotle and the Political Economy of the Polis’, JHS 98 (1978), 57–73Google Scholar.

100 For a general discussion of the effects of literacy on economic life, see Goody, , op. cit. (n. 67), pp. 45–86Google Scholar.

101 My position is similar to that of Finley, , op. cit. (n. 28), pp. 31–2Google Scholar except in regard to the rise of chattel slavery where I think he is mistaken. Despite my numerous disagreements with the views of the late M. I. Finley, I must confess that I have learned a great deal from his writings. Of all the works on real security in ancient Athens, I have found his to be the most useful.

I would like to thank the journal's anonymous referee for judicious criticisms of this article which caused me to rethink several of my arguments. Gratitude is also due to my wife, Victoria F. Harris, who read an earlier draft of the introductory section and strongly recommended that I revise it.