¹ The bibliography for the standard view is too extensive to be given here. Chief among the basic works is Meritt, B. D., Wade-Gery, H. T., McGregor, M. F., The Athenian Tribute Quota Lists, iii (Princeton, 1950), pp. 320, 326Google Scholar (hereinafter referred to as ATL); Meiggs, R., Lewis, D. M., Greek Historical Inscriptions to the End of the Fifth Century B.C. (Oxford, 1969), no. 58Google Scholar; Bradeen, D. W., ‘The Kallias Decrees Again’, GRBS 12 (1971), 469–83Google Scholar; Thompson, W. E., ‘Internal Evidence for the Date of the Kallias Decrees’, SO 48 (1973), 24–45CrossRefGoogle Scholar. Among the circumstantial evidence that has been used to support the standard date is, e.g. apparent reference in decree B to work on the Propylaia, which has been thought to support 434/3, because it predates the end of construction. The most notable exceptions to the traditional date have been made by H. B. Mattingly, who proposes 422/1, in a number of articles, the fullest treatment appearing in ‘The Financial Decrees of Kallias (IG i².91/2) PACA 7 (1964), 35–55Google Scholar, and Fornara, C. W., ‘The Date of the Callias Decrees’, GRBS 11 (1970), 185–96Google Scholar, who revives Beloch's date of 418/17. I will show that, although Mattingly and Fornara have highlighted many of the serious difficulties inherent in the date of 434/3, the alternatives proposed are not without problems as well.

² See, e.g., de Ste. Croix, G. E. M., The Origins of the Peloponnesian War (London, 1972), pp. 73–4Google Scholar; Meiggs, R., The Athenian Empire (Oxford, 1972), p. 200Google Scholar; Woodhead, A. G., ‘Before the Storm’, in Mélanges helléniques offerts à Georges Daux (Paris, 1974), p. 382Google Scholar. Cf. most recently, Triebel-Schubert, C., ‘Zur Datierung der Kallias-Dekrete’, Quaderni Catanesi 12 (1984), 355–75Google Scholar, who would place decree A in 435, supposing preparation for war even earlier.

³ With the exception of Pritchett, W. K., ‘Kallias: Fact or Fancy?’, CSCA 4 (1971), 220–5Google Scholar, and ‘The Choiseul Marble’, BCH 101 (1977), 9–10 n. 5Google Scholar, who has not dealt with the date; cf. Triebel-Schubert, art. cit., who places decree A in autumn 435 and decree B in autumn/winter 434/3, but curiously does not address the arguments, both epigraphic and substantive, that have supported the traditional view that the two decrees were passed at the same time.

⁴ Only shortly before: e.g. Meiggs, and Lewis, , op. cit., p. 158Google Scholar. Some years before: Mattingly, art. cit.; Fornara, art. cit. For a different view, see Linders, T., The Treasurers of the Other Gods in Athens and their Functions (Meisenheim am Glan, 1975), p. 53Google Scholar, who, however, has no objection to 434/3.

⁵ Wade-Gery, H. T., ‘The Financial Decrees of Kallias (IG i².91–2)’, JHS 51 (1931), 57–85, at 59CrossRefGoogle Scholar. Exception: Pritchett, art. cit. (1971).

⁶ Bradeen, , art. cit. 469–70Google Scholar, noting that Kirchhoff in his comments on IG i.32 had suggested that the same mason inscribed both decrees; cf. also Pritchett, , art. cit. (1971), 224–5Google Scholar. Wade-Gery, , art. cit. 58Google Scholar, believed that he could distinguish between the hands on the two faces.

⁷ Art. cit. 470.

⁸ Decree A contains the newer form αις (lines 6, 18, 29); in decree B the older form ασι appears (line 21), as Wade-Gery, noticed, art. cit. 58–9Google Scholar.

⁹ A: lines 4, 15, 21, 30; B: lines 5, 11 (partially restored), though ⋯μ π⋯λει is restored in line 13.

¹⁰ Lines 3–4, 5–6, 11–12, 30; Dover, K. J., ‘The Language of Classical Attic Documentary Inscriptions’, Trans. Phil. Soc. (1981), 1–14, at 7–8Google Scholar; cf., however, decree B, line 4, which may possibly be similar to A in this respect.

¹¹ Although he does not rule out the possibility that one might, on closer examination, detect features in the two decrees which distinguish them from other documents.

¹² Dover notes, art. cit. 1, that three individuals could have a part in the final text of a decree inscribed on stone, the proposer, the secretary (γραμματε⋯ς) and the under-secretary (ὑπογραμματε⋯ς).

¹³ West, A. B., ‘The Two Callias Decrees’, AJA 38 (1934), 390–407, at 393CrossRefGoogle Scholar.

¹⁴ Beloch, , Grieehische Geschichte, ii.2 (Strassburg, 1916), p. 346Google Scholar, thought that the term ο⋯ θεο⋯ might include Athena. It is unlikely that Athena on the Akropolis could be included since in decree A payment of money to ο⋯ θεο⋯ is contrasted to Athena's money; certainly it could, however, have included treasures from Athena's shrines in Attica and elsewhere in Athens.

¹⁵ E.g. 1G i³.472, lines 7–9; 1G i³.369, lines 54–5, 77, 94 (restored), 104;IG i³.383passim.

¹⁶ Art. cit. 393.

¹⁷ Cf. in the case of decrees A and B, inconsistency in the use of the aspirate: A, lines 4, 5, 22; B, lines 8, 27. On the aspirate, see Threatte, L., The Grammar of Attic Inscriptions, i (Berlin, 1980), pp. 493–4Google Scholar.

¹⁸ Since the demos cannot have known the precise amount that some of those funds would yield, nor did they know how much was owed; and the dekate, ‘whenever it is farmed out’ (line 7), has an indefinite temporal sense.

¹⁹ Art. cit. 59 n. 7.

²⁰ Meritt, B. D., ‘Note on the Decree of Kallias’, AJP 55 (1934), 263–74, at 273Google Scholar.

²¹ Meritt, art. cit. This is puzzling a propos of Wade-Gery's comment in his 1931 article that ‘Prof. Meritt has very kindly confirmed for m e that λλ in καλλιας is certain’ (art. cit. 59).

²² The lower half of a n epsilon which was clear to Wade-Gery was also read clearly by both Ross inIG i.32 an d Hiller in IG i².92, subsequently dotted by Meritt, and seen clearly by Pritchett and Daux and also myself (see below).

²³ Those for whom a visit to the basement of the Louvre is not imminent may consult with profit Pritchett's photographs, art. cit. (1971), plates 2 an d 3.

²⁴ Art. cit. (1971 and 1977); in the latter, by proposing new readings for line 3 which might necessitate a shorter prescript.

²⁵ Pritchett, , art. cit. (1971), 223Google Scholar. Two of the letters, a certain epsilon and a dotted iota, are not included in the spaces where the names have been restored.

²⁶ It is a pleasure to thank the Director of Greek an d Roma n Antiquities at the Louvre, M. Pasquier, for permission to examine the stone, an d his efficient staff for their generous assistance. My readings are based primarily on autopsy and less so on a squeeze, which has served more to confirm (both certainties and uncertainties) than to provide further information.

²⁷ This would be compatible with tau and upsilon, as well as iota.

²⁸ The surface is badly worn; the ‘sigma’ seems to be mere discoloration; cf. Pritchett-Daux, : ‘Wewere not positive that there was any sure trace of a sigma’, art. cit. (1971), 222Google Scholar. The bottom of a vertical stroke in the centre (3–4 mm) may be discernible; the line of weathering runs down from this ‘stroke’ to the next line; cf. IG i.32, which records a vertical stroke read by Ross.

²⁹ See Pritchett, , art. cit. (1971), 223 n. 9Google Scholar. I am judging on the basis of the normal, central placement of the vertical bar of an upsilon, not the sole exception in line 27. The stroke is also too far left for a tau; neither is a pi probable, since the right vertical stroke should be visible, nor a kappa, as Hiller suggested, since the bottom tail should be visible. The stroke would be compatible with an iota, since the iotas often appear in the left of the stoichos, e.g. line 19, space 12, 16, 41, line 20, space 16, 21, etc. Rho may be possible.

³⁰ Hiller printed a certain omicron; Ross, the earliest known examiner of the stone, read a horizontal stroke in the lower part of the stoichos, which would be compatible with an epsilon, delta, and zeta. Pritchett reads no letter here; he notes a ‘patina’ that is ‘roughly circular', art. cit. (1971), 222.

³¹ Pritchett and Daux saw no traces of original letters in these two spaces and doubt that the existence of ‘any trace of original letter remains’, art. cit. (1971), 222; cf. Hiller's readings: ΛAΣΕΙ. Ross read nothing in these spaces.

³² As did all editors previous to Wade-Gery.

³³ Hiller seems to have read a full vertical stroke, an d Ross, the lower part of one.

³⁴ Art. cit. (1971), 221.

³⁵ As well as those of the earliest reports which conflict with Wade-Gery an d Meritt. These should not be ignored, since the condition of ancient stones has often deteriorated at a rate at least equalling the improvement of epigraphic techniques.

³⁶ Since the beginning of the inscription is not preserved, and proper names can of course appear in the text of a decree; cf. Pritchett, , art. cit. (1971), 223 n. 9Google Scholar.

³⁷ If indeed the first preserved line of decree B is partof a prescript of which the preceding lost line is the beginning, nevertheless Wade-Gery's observation that the length of the prescript of both A and B is the same carries little weight, since that of A (91 spaces) is very close to the average length of prescripts of its approximate date: cf. Pritchett, , art. cit. (1971), 224Google Scholar.

³⁸ I.e. the inscriptions concerning the Athena Nike temple and priestess, IG i³.35 and 36.

³⁹ Although I must necessarily consider decree B where it is relevant to the date of decree A, it is beyond the scope of this article to consider fully its date, clearly an important matter as well. For some brief comments see n. 84.

⁴⁰ On the connection between this formula and Greater Panathenaic years, see most recently, Develin, R., ‘From Panathenaia to Panathenaia’, ZPE 57 (1984), 133–8Google Scholar; cf. Gauthier's, P. comments on Develin's arguments in REG 100 (1987), 317–18Google Scholar.

⁴¹ See above, p. 96.

⁴² Wade-Gery, , art. cit. 77Google Scholar, attractively supposed that the opening of the Parthenon for use could explain the beginning of these accounts in 434/3.

⁴³ Wade-Gery, , art. cit. 69Google Scholar; Meritt, B. D., ‘Thucydides and the Decrees of Kallias’, Hesperia Suppl. 19 (1982), 112–21, at 112CrossRefGoogle Scholar.

⁴⁴ Art. cit. 69.

⁴⁵ I am referring solely to the ⋯π⋯δοσις to ‘the Gods’ and not to the 3000 talents of Athena ‘brought up to the Akropolis’, a phrase which has generated some controversy over the meaning of ⋯νεν⋯νγται. Cf. Meyer, E., Forschungen zur alien Geschichte 2 (Halle, 1899), pp. 104–5Google Scholar; Kolbe, W., Thukydides im Licht der Urkunden (Stuttgart, 1930), p. 69Google Scholar; Bannier, W., ‘Zu den Beschlussen IG i²91/92’, RhM 75 (1926), 184–202, at 186Google Scholar, who argued that the phrase ⋯ναφ⋯ρειν ε⋯ς π⋯λιν applied to a grant rather than a repayment. Subsequent discussion has tended to support the notion of a freely given grant, or transfer of money, cf. Gomme, A. W., ‘Thucydides ii 13, 3’, Hisloria 2 (1953–1954), 1–21, at 12–21Google Scholar; Historical Commentary on Thucydides, ii (Oxford, 1956), pp. 26–33Google Scholar(hereinafter HCT); Bradeen, , art. cit. 478–9Google Scholar. Mattingly, who argued in 1964 (art. cit. 43) that the phrase referred to repayment of a debt, changed his position in 1974 (‘The Mysterious 3000 Talents of the First Kallias Decree’, GRBS 161 [1975], 15–22, at 18Google Scholar), noting that ‘transfer’ is a better translation [of ⋯ναφ⋯ρειν] than either ‘give’ or ‘repay'; cf. also ATL iii.281 and 327–8. In Kallias' decree, the 3000 talents are contrasted with the ⋯π⋯δοσις to ‘the Gods', rather than being part of an overall repayment procedure. Incidentally, it is important to remember that the existence of loans contracted by the state at a certain period does not necessarily imply the absence of available funds; cf. for example the situation noted by Wade-Gery, , ‘The Year of the Armistice, 423 B.C.’, CQ 24 (1930), 33–9, at 38Google Scholar, in which the state took out a loan for a military expedition just after the Dionysia although the Hellenotamiai had just received the year's tribute, and Ferguson, W. S., The Treasurers of Athena (Cambridge, Mass., 1932), pp. 153, 162–3CrossRefGoogle Scholar, But cf. Gomme, , HCT ii.433Google Scholar, who surmised that the Hellenotamiai might have handed over each year's surplus to Athena, which would have comprised the difference between the tribute total and the amount reserved for routine and predictable expenses, e.g. maintenance of the fleet, etc. This is the hypothesis which the editors of ATL and others adopt for the years 449–434, whereby in each of those years the Hellenotamiai handed over to Athena a sum of 200 talents, representing the average annual surplus (ATL iii.326–8).

⁴⁶ Lines 103–5; cf. also ‘eleven years’ (counting from 422), lines 119–20.

⁴⁷ Athena Polias: lines 98–9, 114–15; the ‘other gods’: lines 102–3, 119–23; Athena Nike: lines 106–7. 112–13; Hermes: lines 109–10.

⁴⁸ Since in each of those years, the Logistai present an annual account of loan activity and one might expect some indication were a loan repaid in the course of an annual term.

⁴⁹ Art. cit. 185.

⁵⁰ Bannier, , art. cit. 184–5Google Scholar, lists a few of these parallels.

⁵¹ Cf. also Hell. 1.5.7; 2.1.12; Anab. 1.2.11; 7.7.74; Suidas, s.v. ⋯κλ⋯γοντες κα⋯ εἰσπρ⋯ττοντες τ⋯ ⋯φειλ⋯μενα τῷ δημοσ⋯ῳ.

⁵² Cf. also IG i³.37. lines 25–6; IG i³.68, lines 16–17 (partially restored), 21–4 (almost entirely restored); IG i³.71, lines 10–11, 29 (partially restored);IG i³.99, line 9; ATL ii D15, lines 25–6 (partly restored); D7, lines 48 (restored), 71–2 (restored); i.579 (T69) section 9 (restored); Tod 154, line 16; 156, lines 16–17. The use of ⋯φε⋯λω for fines is exceedingly common, as in, e.g., IG ii².809, line 70.

⁵³ It is, however, worth keeping in mind that neither is there direct evidence for the state borrowing from the ‘other gods’ before 433 B.C., on the usual interpretation of Kallias decree A. Naturally it is easy to accept the latter situation in view of the later evidence that attests it.

⁵⁴ Linders, , op. cit., pp. 12–16Google Scholar with notes; cf. Haussoullier, B., La vie municipale en Attique (Paris, 1884), pp. 35–73Google Scholar, who mentions in particular state support of the Amarysia, but without a clear ancient source. An argument against the belief that, in general, the local shrines were disconnected from the state until the time of the Kallias decree is the unlikelihood that the state would have been in a position to borrow money from the local gods at all were they completely separate.

⁵⁴ Whitehead, D., The Demes of Attica (Princeton, 1986), pp. 179–80Google Scholar; cf. also Ferguson, , op. cit., p. 162 n. 2Google Scholar.

⁵⁶ The alternative, that of simply money due to the gods, not repayment of a loan, might make better sense of the fact that nowhere in decree A is there reference to interest (τ⋯κος) to be calculated on the amount borrowed, which occurs regularly in connection with loans, e.g. in IG i³.369, passim.

⁵⁷ Wade-Gery, , art. cit. (1931), 69 n. 50Google Scholar, seemed not to be aware of this in his objections to Bannier's proposed date for the Kallias decrees, 431.

⁵⁸ The extant inventories of the Other Gods, IG i³.383, begin in 429 and look back to previous officials in control of the funds of the Other Gods: τ⋯δε παρ⋯δ[οσαν παραδεχσ⋯μενενοι] παρ⋯ τôν π[ροτ⋯ρον ταμιôν οἱς … ], lines 9–10. If these are also ταμ⋯αι τ⋯ν ἄλλων θε⋯ν, then on the basis of the extant evidence, we should date the creation of this board of treasurers to 430.

⁵⁹ Op. cit., p. 158.

⁶⁰ Art. cit. 471 n. 14.

⁶¹ It is also difficult to see how hοι ν⋯ν ἄρχοντες in lines 21–2 could be a different board from hοι ν⋯ν ταμ⋯αι κα⋯ hοι ⋯πιστ⋯ι κα⋯ hοι hιεροποιο⋯ of lines 18–19, or that they are not local officials, contra, e.g., Mattingly, , ‘Athenian Finance in the Peloponnesian War’, BCH 92 (1968), 450–85, at 458–60CrossRefGoogle Scholar, following Beloch, op. cit., p. 347, according to whom the tamiai of line 18 would be the present board of tamiai on the Akropolis. Cf. Meritt, , ‘The Second Athenian Tribute Assessment Period’, GRBS 8 (1967), 130Google Scholar, who supposes that the officials of line 18 are the local officials but does not consider the implications of this view. The alternative proposed by Linders, , op. cit., pp. 49–50Google Scholar, that ‘the present treasurers’ might have been a temporary commission which disbursed the funds that had originally been borrowed from the ‘other gods’ seems unnecessarily complicated, and, further, it is not at all clear that ‘some authority must during this time [when the money was initially borrowed from the temples around Attica] have made the disbursements’.

⁶² As is Linders', view, op. cit., pp. 11, 48Google Scholar.

⁶³ For example, prior to decree A, there must have been a vote in the assembly to specify the funds out of which the gods would be paid; the terms used in this decree are hardly adequate by themselves as a specification, especially those in lines 6–7 (κα⋯ τἅ ⋯στι το⋯τον [τô] ν ×ρεμ⋯τον …).

⁶⁴ As Bradeen thought, art. cit. 471 n. 14.

⁶⁵ The ⋯πειδ⋯ν in lines 10–11, as well as in line 30, may suggest a continuing process and not a single action. I agree with Fornara, art. cit. 194 n. 25, and Linders, op. cit., pp. 40–1, and others, that the vouchers are to corroborate, not to establish the debt.

⁶⁶ It is also clear that it will continue into the future: κα⋯ ⋯κ τες δεκ⋯τες ⋯πειδ⋯ν πραθει (line 7).

⁶⁷ Cf. e.g. de Romilly, J., Thucydides and Athenian Imperialism, trans. Thody, P. (Oxford, 1963), p. 25Google Scholar; Kagan, D., The Outbreak of the Peloponnesian War (Ithaca, 1969), esp. pp. 241–2Google Scholar.

⁶⁸ Andrewes, A., ‘Thucydides on the Causes of the War’, CQ 9 (1959), 223–39, at 236CrossRefGoogle Scholar.

⁶⁹ As Fornara points out, art. cit. 188.

⁷⁰ As R. Sealey has observed, ‘the remarkable feature of the comment [1.44.2] is that the preceding narrative of events scarcely justifies it’ (‘The Causes of the Peloponnesian War’, CP 70 [1975], 89–109, at 98)Google Scholar.

⁷¹ One might escape this objection by dating the Kallias Decrees to 434/3 but after the Kerkyraian debate on the assumption that the Athenians then realized the likely consequences of alliance with Kerkyra. However, not only does this result in an extremely tight chronological framework for the change in attitude and policy which it would imply, but for reasons which shall become clear I find any date in 434/3 unlikely.

⁷² ξυγκρο⋯ειν δ⋯ ὅτι μ⋯λιοτα αὐτοὺς ⋯λλ⋯λοις ἴνα ⋯σθενεστ⋯ροις οὐσιν …(1.44.2).

⁷³ Mattingly in 1968, art. cit. 456, well brought out the importance of this passage for the question of the date of the Kallias decree, yet his arguments have largely gone ignored.

⁷⁴ Perikles ends his account of Athens' financial resources with mention of the gold plates from the statue of Athena on the Akropolis, but it is clear that he has now moved on to consider the resources available in a ‘worst-case’ scenario, and therefore it does not affect his previous distinction between what is on the Akropolis and what is not.

⁷⁵ HCT ii, loc. cit.

⁷⁶ Art. cit. (1968), 456.

⁷⁷ Cf. ATL iii.333, whose editors agree that the temples mentioned in this passage are local, but suppose that their treasures were never concentrated, e.g., such as Rhamnous. But there is no evidence that the state ever used, or was in any easy position to use, the money from the temples to which they suggest 2.13.5 refers; cf. Linders, , op. cit., p. 5Google Scholar.

⁷⁸ An additional point, admittedly indecisive but worth making, is that the historian dwells in 2.16 on the reluctance of the Athenians to forsake their hereditary temples, after Perikles urged the residents of Athens and Attica to evacuate the countryside and move within the city walls; he seems particularly interested in their traditional religious feelings. The entire passage from 2.13 to 2.16 perhaps loses some of its force if, three years before, the inhabitants of Athens and Attica had watched the treasures of their gods hauled away to the Akropolis in expectation of, and in preparation for, war.

⁷⁹ Bannier, , art. cit. 184–202Google Scholar; of course 431 is the date when those who argued for a later post-Archidamian war date for the decrees believed that the concentration of the treasures would naturally have occurred, e.g. Wade-Gery, , art. cit. (1931), 67Google Scholar, before he changed his mind in favour of 434; Mattingly, art. cit.; Fornara, art. cit.

⁸⁰ I agree fully that the decree itself contains absolutely no sense of impending crisis; for the contrary view, cf. Woodhead, , op. cit., p. 383Google Scholar.

⁸¹ Art. cit. 487.

⁸² Cf. also Perikles' comments in Thuc. 1.141.3–5, 142.1. The enormous expense which the Peloponnesian war would entail and therefore the need not only for a regular source of capital but more so for reserves were points which the perspicacious Archidamos recognized as well, Thuc. 1.83.2; cf. also 1.80.3–4, 81.4, 82.1, 83.2.

⁸³ The date of this vote, and the source(s) and nature of the amount, are unknown, and therefore, a fertile field for hypothesis. The editors of ATL worked out an elaborate reconstruction, linking together several important and highly problematic documents, of which the Kallias decree, necessarily dated to 434/3, was the linchpin. They argued that the Hellenotamiai made an annual grant of tribute to the treasury of Athena over a period of 15 years, beginning in 449 and ending in 434, when a total of 3000 talents was achieved (op. cit. iii.326–8). This reconstruction was based partly on the inference, erroneous in my opinion, that the Hellenotamiai had to have been the source of the 3000 talents in the Kallias decree because they are the only paymasters specified in the beginning of the decree, and partly on their interpretation of the so-called Papyrus Decree, for the fullest treatment of which see Wade-Gery, H. T., Meritt, B. D., ‘Athenian Resources’, Hesperia 26 (1957), 163–97CrossRefGoogle Scholar. I argue elsewhere (see ‘Did Tribute Fund the Parthenon?’, CA 8 [1989])Google Scholar that this view rests on exceedingly tenuous grounds and lacks direct evidence. Most important for our purposes, if the Kallias decree does not date to 434/3, their reconstruction is untenable, including for example the hypothesis that the Hellenotamiai had no reserves of their own, because they had systematically turned over the annual tribute surplus to the treasurers of Athena.

There must be another explanation for the 3000 talents, but any attempt to provide one will be necessarily speculative. In this spirit, exempli gratia, I offer one of my own in order to show that other conjectures besides ATL's are possible. First, it is reasonable to suppose that the initial decision to increase Athena's treasury by 3000 talents was taken some years before (unless the 3000 talents were one lump sum transferred from one treasury off the Akropolis, i.e. the Demosion), since the sum is substantial. A plausible time for such a decision may have been shortly after the conclusion of the Samian war (440–39), quite a costly venture, amounting to some 1200 talents at least (IG i³.363), paid for from the funds of the treasury of Athena. This came on the heels of a difficult decade for Athens of revolt and disaffection among her allies. In addition, Athena's treasury was the chief contributor to the building programme (from what is extant: for the Parthenon, IG i³.437, 439, 440, 442, 444, 449 [though entirely or partly restored];, for the Propylaia, IG i³463, 465, 466?). I suggest that the need to replenish the state reserves in F order to face further imperial challenges thus became manifest and enabled Perikles to convince the Athenians of the priority of augmenting Athens' treasury by 3000 talents. (It is interesting that Athena's treasury apparently did not contribute to the Parthenon from 438 to 434, possibly suggesting that the need to replenish the treasury took priority; there are, of course, other possible explanations.) The relatively quiet years of the early to mid 430s will have allowed this j sum to be achieved by 431, at which time repayment to ‘the Gods’ could commence. I see no reason why the 3000 talents could not have come from a number of sources, both domestic and imperial, and not solely tribute. But it would be foolhardy and misleading to pretend that a convincing explanation for the 3000 talents is ready at hand, since we possess not a single piece of evidence in the mire of pre-Peloponnesian war public finance in Athens that could answer this specific question, and we are reliant upon conjecture and inference.

⁸⁴ Where does this leave the date of decree B? As mentioned above (n. 36), it is not possible to treat this question here, but a few comments can be made. First, decree B should come after decree A, since it contains the ‘later’ term, οἱ ἄλλοι θοι θεοι(cf. above, pp. 96, 107–108); and it cannot date before the treasury of the Other Gods was established, i.e. possibly 430, certainly 429. If the decrees are in fact related in content to the extent that the sum of money in decree A voted for repayment, or payment on the alternative view (cf. above, pp. 103–104), was the amount specified in decree B (line 22), then the latter should perhaps not be much later than decree A; 430 in that case is a likely choice. If the decrees are not so related, the later dates proposed for both decrees remain possible for decree B: of the two candidates proposed, 418 perhaps more so, because of the appearance of a vote of dSeia in that year (IG i³.37O) (although not until the second prytany), a procedure prescribed in decree B. The Greater Panathenaic dates 426/5, 422/ 1, and 418/17 would follow the date when, according to Thucydides 3.19.1, an eisphora was first instituted, i.e. 427, a measure to which decree B alludes twice (lines 17, 19), in both cases, however, partially restored. (A possible reference to eisphora in IG i³.41, line 58, concerning the Hestiaians does not, in my view, affect Thucydides' comment since there is no suggestion that the tax applied to Athenians. I agree with Sealey, ‘The Tetralogies Ascribed to Antiphon’, TAPA 114 [1984], 77–80Google Scholar, that Thucydides means the debut of eisphora and not the first during the war, or the first to bring in 200 talents, as Griffith, J. G., ‘A Note on the First Eisphora at Athens’, AJAH 2 [1977), 3–7Google Scholar, has recently argued. Of great relevance here is that the suggestion, that Thucydides meant that, although eisphorai had been introduced long before, only in 428 did they collect as much as 200 talents, was only prompted by the belief that the Kallias Decrees, or, more specifically, the second decree, dated to 434, contradicted Thucydides.) If one finds convincing the possible appearance of eisphora in decree B, then there is some support for a post 427 date for decree B. It is important, however, to keep in mind one significant difference between this decree (and others) and Thucydides' History: the former only provides a record of a decision; but the latter attempts to inform us of what actually happened. An apparent reference in B (line 3) to the Propylaia suggests an early no more than a later date: on the usual interpretation, the decree provided for its completion (παντελôς, line 4); since this never happened, this reference is of quite uncertain value in dating the decree.

⁸⁵ The ideas and arguments presented here originated in a paper written for Professor R. S. Stroud's seminar on Greek epigraphy at Berkeley, and were developed and offered in brief form in an appendix to my doctoral dissertation and in a talk given at the 1987 Annual Meeting of the APA. Special thanks are due to Professor Stroud for many discussions, general encouragement, and comments on earlier drafts. I also thank Professors R. Sealey, E. S. Gruen, and M. Ostwald for reading and commenting on previous versions. To the anonymous referee of CQ are due considerable improvements of content. I thank Professor R. M. Kallet-Marx for his editorial expertise at the final stage of preparation, and for examining IG i³.52B with me. None of the above should be presumed to agree with my arguments and conclusions.