Prof. G. Pepe for the permission to reproduce them, and to Ms Allison Carrick for her ready co-operation. — I also thank Dr R. A. Coles, who read an earlier draft, and Dott. F. Morelli, who furnished material on prices of wheat. 2 Although I have taken account of the various published corrections, this version is essentially the product of a fresh collation. I have generally not indicated the numerous divergences (mostly minor, except for those stemming from taking the two fragments as separate documents) from the first editions.
‘… for interest on this (capital), each [year from the] present month of Mesore of the current [year 1*3/1*]2 of the fifth indiction, [two] and a half [artabas] of wheat, total 2 1/2 artabas, without fail. The one solidus of gold of the aforesaid capital, being free of all risk, I shall be bound to pay back to you whenever you wish without delay, you having the right of execution against me and against all my belongings. The contract, written in [a single copy], is binding, and in answer to the formal question I gave my assent.’ (2nd hand) ‘I, Aurelius Petrus, son of …, the aforesaid, have received on loan the one solidus of gold, capital, and shall repay (it) together with the interest as aforesaid.’ ‘… son of Aphynchis, wrote on his behalf in his presence because he is illiterate.’ Íp¢r diaf]Òrou. Íp¢r lÒgou diaf]Òrou with PHarr I 86.5 is another, but less likely, alternative. [§niautÚn épÚ toË ˆ]nto!: ¶to! (or m∞na) SB XX 14641. A construction with ¶to! would be unidiomatic. The supplement may be paralleled by POxy XVI 1969.9 kay' ßka!ton §niautÒn. [m∞na épÚ toË eﬁ!iÒ]nto! would have the right length too, and indeed the interest rate is usually stated by the month, but here this would be implausible: 12 x 2.5 = 30 artabas is far too large an amount to represent the annual yield of a solidus; cf. 5 n. We may compare BGU XII 2140, a Hermopolite loan of 432, also of unspecified duration, which calculates the interest yearly. 4 ¶tou! r `g r `]b` . The SB version has ` ¶tou]!`, corrected to ¶touw ` ` ` ` `]w` in BL IX 370 (already alluded to in BL VIII 509). Oxyrhynchite era years follow the pattern x = y, with y = x-31. An era year (*)*7/(*)*6 should correspond to indiction 4 or 14. Here we have indiction 5, which indicates that the last digit of the second figure of the era year must be 2 or 7; see R. S. Bagnall, K. A. Worp, The Chronological Systems of Byzantine Egypt (1978) 75 ff. It is thus unlikely that the traces after the break conceal a 6—and they do not. In fact, it is possible to read b` . Accordingly, the last digit of the first leg of the era year should be 3. Possibilities include years 143/112 and 173/142, cf. Bagnall, Worp, op. cit. 83, 85; the contract was drawn up in the month of Mesore, so it may be dated to 25.vii-23.viii.467 or 497—paleography would not favour a date in 437 (113/82), or in 527 (203/172). 5 [értãba! dÊo ¥]mi!u. On loans of money with interest in kind see H. E. Finckh, Das Zinsrecht der gräko-ägyptischen Papyri (1962) 87-8. The usual rate of interest was 12 % per annum, see CPR VII pp. 162-3 Exkurs 5. If this were the case also here, where the annual interest rate is 2.5 artabas of wheat per solidus, we could posit that 2.5 artabas of wheat had a market value of 2.88 carats, which would suggest a price of 1.15 carat per artaba. But the rate could also have been higher, e.g. 25 %, which would yield a price of 2.4 car./art. Our evidence on fifth-century prices of wheat is too scant to allow comparisons, cf. A. C. Johnson, L. C. West, Byzantine Egypt. Economic Studies (1949) 177, J. R. Rea, POxy LI p. 73 (cf. LI 3628.11 n.), although we may perhaps compare BGU XII 2140, a loan of 8 solidi, the yearly interest being one artaba per solidus. A 12 % rate suggests a price of 2.88 car./art.; with 25 % we get 6 car./art. The latter figure seems too high for the period, so that we may hazard the guess that in Hermopolis in 432 wheat was valued at 2-3 carats per artaba. Naturally, this does not necessarily imply that in our loan the interest was in the range of 25 %. 6-7 ék¤n[d]u`n`[o]n ¯n épÚ p[an]tÚ! k`in`d`Ê`n`o`u`. The phrase is generally thought to be typical of Oxyrhynchite loans, see CPR VII p. 164, but has recently occurred in a loan contract drawn up at Alexandria, POxy LXIII 4395.21-2 (c. 499500) (at the end of line 21 of this text the space allows supplying épÒ). 8 ıpÒ[tan] boulhyª!. See CPR VII Exkurs 4 pp. 161-2; the earliest Oxyrhynchite loan of indefinite duration is PLeidInst 66 of 427. 9-10 parã te §moË. Restore parã (¶k ed. pr.) te §moË in PLeidInst 66.10. The usage parã te §moË ka‹ §k t«n ÍparxÒntvn moi pãntvn is standard in Oxyrhynchite documents of this date. 11 [èploËn g]raf°n. Ed. pr. notes: ‘Either èploËn or di!!Òn suits the initial lacuna’ (PWashUniv I 23.5 n.); similarly, CPR VII 39.8 n. states: ‘Eine Entscheidung zwischen èploË]n` und di!!Ò]n` ist wohl nicht möglich’. But I have found no example of a construction of grammãtion with di!!Òn. 11-12 AÈrÆlio! P°tr[o]! [ c. 6 ] `. Ed. pr. interprets the traces after the break as u, which is likely, but ! may also be considered. An Aurelius Petrus occurs in the Oxyrhynchite PBad VI 168 (V); his patronymic, Patçto!, would fit the space and possibly the trace in line 12, but the name P°tro! is common, and he need not be our man. 14 OÈ]a`l°ri`o`!` is difficult. 2 3