We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 11 October 2013
In a recent volume of this Annual (XLV 126 ff.) I discussed the alphabetic numeral system as employed in Attica; in the present article I examine the use of letters in Attica and elsewhere to identify different items, often similar in character and appearance, with a view of facilitating reference and simplifying inventories. For such a purpose letters have certain advantages over other symbols which might be devised; they are brief and familiar and they occur in a recognized order. They thus approach nearly to the use of ordinal numbers (contrasting sharply with acrophonic numerals, which are invariably used to represent cardinal numbers), though it cannot be said that they constitute a numeral system, any more than we could claim that we in English use an alphabetic numeral system because, e.g., a, b, c, d are used on p. 4 below to distinguish four items which might equally well have been numbered 1, 2, 3 and 4. The letters here under consideration were not, and could not be, made the instruments of arithmetical calculation, and the highest number expressed in this way in any inscription known to me is 106 (Inscr. Délos 1432 Aa ii 21; see below, p. 8).
Various scholars have dealt briefly with the subject, but the accumulation of a mass of new evidence calls for a fresh treatment, especially of the part played by such letters in inventories. No technical name has, I believe, been given to letters so used, and in this article I call them ‘letter-labels’, a title which emphasizes the function they fulfil in the majority of cases where they occur.
1 See Franz, J., Elementa 348 f.Google Scholar, Reinach, S., Traité d'épigraphie grecque 220 f., 472Google Scholar, Woisin, J., De Graecorum notis numeralibus 28 ff.Google Scholar, Meisterhans, —Schwyzer, , Grammatik der attischen Inschriften 10Google Scholar, Larfeld, W., Handbuch II 546 f.Google Scholar, Griechische Epigraphik 3 297 f.
2 IG IX (1) 334, SIG 47, GHI 24.
3 Tod, , BSA XLV 137 f.Google Scholar
4 Seven of these are re-edited in L. Robert, La Collection Froehner I 10 a, b, d, f, h, i, k. Of the five examples which bear the letter eta four have the form 
Add Stähelin, F., AA 1943, 16 ff.Google Scholar, Meletopoulos, I. A., Πολέμων III 33 ff.Google Scholar
5 See the commentary of Hiller von Gaertringen ad loc.; to the bibliography add Schwyzer, DGE 663.Google Scholar
6 Similarly a large B stands at the foot of the right-hand side of another building inscription (103C 329).
7 See, most recently, Dilke, O. A. W., BSA XLV 24.Google Scholar
8 Delorme, J., BCH LXXVII 478 f.Google Scholar
9 W. Peek, Γὲρας ᾿Α. Κεραμοποὺλλου 611.
10 In 1286 we have the letter-names εὶ, зη̄τα and η̄τα and in 1291 κάππα. In the famous list of dedications to Athena Lindia (Lindos II 2) the phrase ἐν τᾶι Ᾱ etc., occurs almost sixty times in book-references, but as the highest number is Δ̅ (for Λ̄ found six times, is an imperfectly engraved or preserved Ā, and Ī in C95 probably stands for Γ̄), we cannot say whether the letters are ‘labels’ or true numerals, i.e. whether Book Χ of the Iliad would have been called Κ or l. In C37 τξ stands for 360.
11 IG I2 91; cf. SEG X 45, ATL II 46 D 1.
12 Cf. SEG X 225.
13 IG I2 92; cf. SEG X 45, ATL II 47 D 2.
14 Cf. SEG X 184–209.
15 See also IG II2 216, 217 (345 B.C.). For the date of II2 120 cf. Schweigert, E., Hesperia VII 288.Google Scholar
16 A. M. Woodward has recently discussed the terms ἀπὸκαυσις and ἀφὲψησις relating to the melting down of golden objects in the time of Lycurgus, (Num Chron 1951, 109 ff.).Google Scholar The Delian inventory for 179 B.C. refers to the ψὴγματα and other objects handed over τοῐς ἀνδρὰσιν τοῐς αὶρεθεισιν ὺπὸ τοῦ δὴμου κατἀ τὸ ψὴφισμα ὢστε συγχωεῦσαι ἐν τεὶ ἐκκλησὶαι ἂ παρὲλαβον χὺματα χρυσᾶ . . . ὰργυρᾶ (Inscr. Délos, 442B 118 ff.; cf. 25 f.).
17 For Attic inventories see Larfeld, W., Handbuch II 870 ff.Google Scholar
18 Cf IG II2 1533, Michel 836–8.
19 IG II2 1378.13 f., 1392.16 ff., 1396.2 ff., 1397–99, 1407.21.
20 IG II2 1421.50 f., 55 f., 1425.94 f., 97 f., 1429.28. 31.
21 L. 12 [ὰσὴ]μου ἀργυπὶου τοῦ εὶς τὰ στρατιωτικὰ ὲξαιρεθὲντὶ[ος]
22 Cf. 1472.43, 45, 1475.9 ff., 1480.7 ff.
23 Cf. 1477.24 ff, 1485.8 ff., 1486.5 ff., 1491.13, 15.
24 IG II2 333, 1493.5; cf. Plut. Vit. X or. 852B.
25 The best preserved item runs [τὲ]ταρτος ρ`υμὸς δὲκα, ο[ὶς βῆτα] καὶ τὲτταρα ιῶτα πα[ρασεσὴμανται] σταθμὸν ΗΗΗΗ (ll. 184–6).
26 If we may connect this list with Plutarch's statement (Vit. X or. 852B) that Lycurgus παρεσκεὺασε . . . . κὸσμον χρυσοῦν εὶς ὲκατὸν κανηφὸρους
27 A further opisthographic fragment of II2 1496 was found in the Agora in 1938 and published by Schweigert, E. in Hesperia IX 328 ff.Google Scholar, no. 37. The text on the reverse is very mutilated and satisfactory restoration is impossible. The words ὲπιγὲγραπται σταθμὸν can be restored with tolerable certainty in ll. 2, 4, 6, 8, 10, and στὲφανος ὲφ᾿ὼι in ll. 3, 5, 7, 9; ῶι is followed in l. 5 by [τὸ] ΛΦ in l. 7 by τὸ and in l. 9 by τὸ Λ.
28 IG II2 1633–53; cf. Coupry, J., BCH LXII 236 ff.Google Scholar
29 BCH LVIII 96 ff.
30 In his first publication Meritt commented that ‘the writing indicates a date in the first half of the second century B.C.’ (Hesperia III 53). I am not sure that κ]αὶ ὲπτὰ is rightly restored in l. 16. The φιὰλαι χρυσαὶ λεὶαι ὲπτὰ of IG XI 161B 27 ff., Inscr. Délos 296B 33 ff., 298A 136 ff., 442B 78 ff., 1430b 7 ff., 1443A i 105 ff., 1450A 74 ff. suggest that the αι may belong to φιὰλ]αιor χρυσ]αΙ or λεὶ]αι though in the Agora text the weights of the several φιάλαι are not separately recorded.
31 These will shortly be published in the Inscriptions de Délos by Coupry, J. (Actes du deuxième Congrèh International d'Épigraphie 253).Google Scholar
32 The records of the Delian archons (IG XI 105–134), extending from 284 to a little later than 170 B.C., though they include a considerable element of inventory, contain no examples of letter-labels. For the administration of the hieropoioi see the bibliography in IG XI (2), p. 18, and add Laidlaw, W. A., History of Delos 147, 152 ff.Google Scholar
33 IG XI 135–289, Inscr. Délos 290–498 (ed. F. Durrbach). Since the texts in these two works run consecutively, I cite them by number only.
34 The cost is recorded in 161A 117 ff. (279 B.C.) and 199C 66 ff. (274 B.C.); in the earlier case the stele cost 25 dr., transport 1½ dr., engraving 100 dr., lead 5 dr., wood 1 dr., erection 2½ dr., total 135 dr.; in the latter the stele cost 24 dr., but the charge for engraving rose to 126⅔ dr., the number of letters being roughly calculated as 38,000. The engraver receives a drachma per 300 letters.
35 See IG XI (2), p. 37, note on 154A 61 ff., and BCH XV 155 f., XLV 179 ff.
36 Very similar is 164A 40 ff. (276 B.C.).
37 So also in 205Ab 20 ff., 230B, 250, 283.6 ff., 287B 77 ff.
38 In 284.9 εῐκοσι ε̆να should be restored in place of εῐκοσι εis (cf. 154A 62, 199B 29). In 287B 77 the heading runs ου̅τοι (where we should expect οἵδε) ὲν τῶι τοῡ ῾Απὸλλωνος πρὸς τῶι τοὶχολ (sic) εῐσω· στὲφανοῐ χρυσοι ΔΔΙ
39 For acrophonic numerals in Delos see BSA XVIII 115 f., XXXVII 250 f.
40 1403 Bb i 23 ff. (soon after 166 B.C.)
41 1408A ii 38 (?), 1441A i 64; cf. Roussel, Délos colonie athénienne 128.
42 1416.B i 1, ii 69, 93, 116 (156 B.C.); in B ii 116 νεοῦ is an error for θεοῦ
43 1417B ii 78 (155 B.C.).
44 1432Aa i 1; cf. 1450A 2 (139 B.C.).
45 Cf. Tod, , BSA XXXVII 250 f.Google Scholar
46 Cf. Tod, , BSA XLV 138.Google Scholar
47 A few trivial misprints affecting letter-labels may be noted. In 1429A ii 29 τὸ Δ should be τὸ Γ, and in l. 31 τὸ Γ should be τὸ Ε; in 1449Ba i 28 τὸ l should be τὸ Η, in l. 142 I suspect that οῦ has dropped out before τὸ ΑΑ, in l. 63 τὰ ΓΓΓ should be τὰ ΔΔΔ, and in l. 72 τὰ ΓΓΓΓ should be τὰ ΓΓΓ; in 1450A 130 ΒΒΒ should be ΒΒΒΒ.
48 The phrase is τὰ Ι[Ι] κὰππα οὺ παρεδὸθη The absence of the object ΚΚ is also recorded in 1429A ii 47 [τὰ ΚΚ οὺκ η̄ν] 1432Ab i 72 [τὰ ΚΚ] οὺκ η̄ν
49 Except in 1449Ba i 42, where the editors give ὲφ᾿ τὸ ΑΑ without comment; the stone probably has, or should have, ὲφ᾿ οῡ τὰ ΑΑ (see above, n. 47), though τῦ ΑΑ etc. occurs in IG II2 1443.73 ff.
50 I have noted no examples of ὲφ᾿ ῶι in place of the very common ὲφ᾿ οῡ and only three of ὲφ᾿ ῆι (ὲφ᾿ εΙ in 1402.5, ὲφ᾿ η̄ twice in 1450A 120), as compared with seventy-three of ὲφ᾿ η̄ς
51 Of ὲφ᾿ οΙς I know only one example, 1450A 138.
52 In 1450A 151 the stone has οῡ τὸ Γ which the editors correct to (ω̇̄ν) τὸ Γ unparalleled elsewhere. We have seen (p. 6) that the inventories of the period of Delian independence always use ου̇̄ (never preceded by ἐφ') in a similar position. Both there and in the examples of ου̇̄ quoted above, ου̇̄ bears a local sense, ‘where’. This is shown by (a) the use of the word relating to φιάλας in 1450A 151, and (b) the use of ̇ινα in some Athenian inscriptions of the same nature (e.g. IG II2 1429.27 ff., 1443.14 ff).
