Appendix 10—
The Dating of Attic Inscriptions

Harold Mattingly, in a series of important articles,^[1] has adduced various arguments to establish the propriety of a lower date for a set of inscriptions containing three-bar sigma and other archaic letter-forms. For example, he would date the so-called "Regulations for Miletus,"^[2] now rather improbably assigned to 450/49, to 426/5; he suggests that the

[1] See Mattingly, "The Athenian Coinage Decree," Historia 10 (1961), 148–88; "Athens and Euboea," JHS 81 (1961), 124–32; "The Methone Decrees," CQ 11 (1961), 154–63; "The Growth of Athenian Imperialism," Historia 12 (1963), 257–73; "The Financial Decrees of Kallias (IG i 91/2)," PACA 7 (1964), 35–55; "Athenian Imperialism and the Foundation of Brea," CQ 16 (1966), 172–92; "Periclean Imperialism," in Ehrenberg Studies, pp. 193–224; "Athens and Aegina," Historia 16 (1967), 1–5; "The Date of the Kallias Decrees," BSA 62 (1967), 14–17; "Athenian Finance in the Peloponnesian War," BCH 92 (1968), 450–85; "Epigraphically the Twenties Are Too Late," BSA 65 (1970), 129–49; "The Language of Athenian Imperialism," Epigraphica 36 (1974), 33–51; "Athens and Eleusis: Some New Ideas," Phoros, 90–103; "The Mysterious 3000 Talents of the First Kallias Decree," GRBS 16 (1975), 15–22; "Three Attic Decrees," Historia 25 (1976), 38–44; "The Tribute Quota Lists from 430 to 425 B.C. " CQ 28 (1978), 83–88; "The Athenian Decree for Miletos (IG i , 22+ = ATL II, D11): A Postscript," Historia 30 (1981), 113–17; "Coins and Amphoras-Chios, Samos, and Thasos in the Fifth Century B.C. ," JHS 101 (1981), 78–86; "The Athena Nike Temple Reconsidered," AJA 86 (1982), 381–85; "The Tribute Districts of the Athenian Empire," Historia 33 (1984), 498–99; review of IG i , D. Lewis (ed.), AJP 105 (1984), 340–57; "The Alliance of Athens with Egesta," Chiron 16 (1986), 166–70; "The Athenian Coinage Decree and the Assertion of Empire," in I. Carradice, ed., Coinage and Administration in the Athenian and Persian Empires (Oxford, 1987), pp. 65–71.

[2] IG i 21 = Fornara 92, SEG 34.7. See, e.g., Mattingly, Historia 10 (1961), 174–81; Phoros, 98–101; AJP 105 (1984), 340–48.

treaty with Hermione,^[3] with three-bar sigma, finds its proper context in 425. The Colophonian Decree^[4] he would move from c. 446 to 427, and the Egestaean Decree^[5] (with immense probability) from 458/7 to 418/17. The difficulty with these proposed dates is that they are all of them plausible, some of them likely, none of them mandatory. The traditional and relatively early context for at least the greater number of the inscriptions discussed by Mattingly is no less acceptable than the later, and it has, on its side, the inference from letter-forms. To jettison the epigraphical criterion would be unjustifiable. We must decide, case by case, whether it is reasonable or unreasonable to allow for exceptions to the rule, keeping open the possibility that Mattingly's systematic redating may yet be vindicated when all of the evidence is more thoroughly assimilated and tested.

In the meantime, it will be enough to argue that "anomalies" do indeed exist. One is the decree of Athena Nike, which has been set into the period 450–445 solely because it was cut with three-bar sigma.^[6] Significantly, other telltale letters (beta, phi, and rho) exhibit "developed" characteristics. This decree authorized construction of the temple of Athena Nike and provided for the selection of her priestess. The temple was constructed in the 420s.^[7] Moreover, a decree on the reverse side of the stele,^[8] apparently carved at the same time, or shortly thereafter, sets the procedure to be followed for the payment of the priestess's salary—a natural consequence of her creation in accordance with the terms of the decree on the obverse side of the stele. This decree is dated with certainty to the year 424/3.^[9] It would therefore be a gross

[3] IG i 31, cf. SEG 10.15, 34.9. See Mattingly, Historia 10 (1961), 173; BCH 92 (1968), 484–85; AJP 105 (1984), 348–49.

[4] IG i 37 = ML 47 = Fornara 99; cf. SEG 34.12. See Mattingly, Historia 12 (1963), 266f.; Ehrenberg Studies, pp. 210–12; Epigraphica 36 (1974), 44ff.; AJP 105 (1984), 344.

[5] IG i 11 = ML 37 = Fornara 81. See Mattingly, Historia 12 (1963), 268f.; AJP 105 (1984), 344; Chiron 16 (1986), 166–70. Cf. T. E. Wick, CP 76 (1981), 118–21.

[6] IG i 35 = ML 44 = Fornara 93; cf. SEG 32.2. For the dating, see Meritt and Wade-Gery, JHS 83 (1963), 109–11; ML, pp. 109–11, Meiggs, AE, pp. 496–503; Contra, Mattingly, Historia 10 (1961), 169–71, and AJA 86 (1982), 381–85. The presumption that the plans for the temple had been used for another (the Ilissos temple) built earlier is no longer tenable. On the possible identification of the hand, see S. V. Tracy, "Hands in Fifty-Century B.C. Attic Inscriptions," in Dow Studies, pp. 277–82.

[7] See Meiggs, AE, pp. 496–503, esp. 501; R. Carpenter, The Architects of the Parthenon (London, 1970), p. 85; Mattingly, AJA 86 (1982), 381 n. 1.

[8] IG i 36 = ML 71 = Fornara 139.

[9] The decree is dated by the secretary (Neokleides); cf. ML 70; Wade-Gery, Essays, pp. 308f.; ML, p. 205.

understatement to declare that, by all the rules we customarily apply to the assessment of data, the decree of Athena Nike belongs in the 420s. The architect Callicrates has received his commission; financial officials are given the date on which to farm out the contract and commence the building. These details compel the obvious inference, which even without them would seem necessary—namely, that a precise correlation existed between the passing of this resolution and its effectuation.

Explanations to justify a date in the early forties, entailed by the archaic sigma, seem contrived merely to neutralize the straightforward implications of the evidence. We know, for example, that the first priestess of Athena Nike died sometime after 411,^[10] which makes the assumption improbable that she had served in office since the early forties—a tenure of almost forty years. The response is not cogent—when we try to assess the sum of the evidence in order to judge the probabilities—that this priestess must have enjoyed unusual longevity, like the priestess of Hera at Argos who served for more than fifty years.^[11] In any event, our priestess, on that assumption, would have been without her temple for some twenty years. Again, the regulation of 424/3 makes the Kolakretai the paymasters of the priestess, referring explicitly to the decree in question. The counterargument is here that some alteration was effected in substance or procedure at that time, though it is neither implied by the sentence nor otherwise known to us.^[12] In fact, the decree of 424 states that "the fifty drachmas designated on the stele [i.e., on the obverse side] are to be dispensed by the Kolakretai." The mutual relation of these two decrees is natural and complementary; the assumption of a hiatus of twenty years is not.

How, above all, is the twenty-year wait for the building of the temple to be explained? It has been suggested that construction was delayed because the architect was involved in other programs.^[13] The argument is special pleading. The magnitude of the Athenian building program is reflective of optimism and affluence. Money was in abundant supply;

[10] J. Papademetriou, "ATTIKA I," Arch. Eph. (1948/49), 146–53; D. M. Lewis, "Notes on Attic Inscriptions (I)," BSA 50 (1955), 1–7; ML, p. 109. The first priestess, Myrrhine, is identified with the Myrrhine of Ar. Lys., produced in 411. See Mattingly, AJA 86 (1982), 381–85.

[11] Thuc. 2.2.1, 4.133. See Meritt and Wade-Gery, JHS 83 (1963), 110; ML, p. 109. Cf. Pliny NH 34.76, with Lewis, BSA 50 (1955), 4–6, for a tenure of office of 64 years for Lysimache, priestess of Athena.

[12] Suggested by Meritt and Wade-Gery, JHS 83 (1963), 111; cf. ML, pp. 204–5.

[13] Carpenter, Architects of the Parthenon, pp. 84–85, holds that the temple was voted to honor Cimon's victory and Callias's diplomacy and then scrapped by Pericles after the Cimon's death (!). Cf. Meiggs, AE, p. 597.

public works were planned and executed with more élan than at any other time in Athenian history. Did the Athenians build the Parthenon and the Propylaea at the price of the tiny temple of Athena Nike? The adoption of this resolution, whatever its date, is the best possible evidence of the people's commitment to bear the expense and commence construction. The proper moment for the cancellation of a project is not after it has been decreed but when the pressure of other burdens comes to make its completion inexpedient. A pinch of this kind did not remotely afflict the Athenians until just before the outbreak of the Peloponnesian War when, indeed, they left the Propylaea unfinished.

It is more realistic to allow for an apparent epigraphical anomaly than to make Athenian behavior arbitrary and fitful in a matter bearing all traces of studied concern. A resolution by the people is a law; the substance of that decree is weightier evidence than one of the letters comprising it. And, indeed, simple reflection should assure us that the style of engraving cannot have been as uniform and unvarying as the present doctrine requires. As W. K. Pritchett writes:

We know of no ancient law that Athenian masons must give up the use of the three-barred sigma precisely in the year 446 B.C.  . . . There is no parallel in the history of the Attic script for such precision in the abandonment, not of a style of lettering, but of one letter-form in one particular year.^[14]

Certainly the style of engraving must depend on idiosyncratic practice as well as on changing trends. Too many unknowns dictated the choice of a workshop or engraver for modern scholars to presume, on the basis of a random sample,^[15] what could or could not have been done at a specific time. The Poletai no doubt preferred certain masons over others when they let the contracts, which may help to explain the comparative homogeneity of style, now elevated to the status of an ironclad rule. What are we to suppose occurred, however, in a busy year or as the result of illness, death, or military conscription? What of older stone masons who had learned their craft in the fifties? What of old-fashioned

[14] W. K. Pritchett, BCH 89 (1965), 425.

[15] E.g., Meiggs, JHS 86 (1966). As he recognized, the small sample also hindered A. S. Henry's attempt to provide firm stylistic criteria for dating from the decree prescripts ("The Dating of Fifth-Century Attic Inscriptions," CSCA 11 [1978], 75–108). However, Henry's observations, here and elsewhere (e.g., The Prescripts of Athenian Decrees, Mnemosyne Suppl. 49 [1977]; "Negative Coordination in Attic Decrees," JHS 97 [1977], 155–58; "Archon-Dating in Fifth Century Attic Decrees: The 421 Rule," Chiron 9 [1979], 23ff.; and Honours and Privileges in Athenian Decrees [Hildesheim, 1983], etc.) possess sound evidential value. Cf. Mattingly, AJP 105 (1984), 355–57.

curmudgeons? Did everyone prefer the appearance of the sigma with the extra bar?

There is no reason to believe that the masons marched in lockstep or that the dictates of style were sharply defined and rigorously followed. The expectations of the Athenians, official and ordinary citizen alike, were indulgent in matters of style. The importance the Athenians attached to the massive stele containing the first series of the tribute-quota lists perhaps suggests that the presence of four-bar sigma in the list cut for the year 453/2^[16] was due primarily to aesthetic considerations; and it may similarly have been chosen for style when it was used in the Erechtheid casualty list of 459—though its appearance there is in itself an "anomaly." But the Athenians were usually indifferent. No other explanation will cover the indiscriminate use of Ionic and Attic together in the same inscription, the appearance of different forms of the same letter, though often carved by a single person, and other like discordancies.^[17] There is a strong possibility that our own appreciation of the aesthetics of Attic inscriptions has caused us subliminally to postulate the refined and self-conscious progression of style more appropriately ascribed to the practicing sculptor. Although one may willingly agree, therefore, that the style of significant letters is a good index of the relative date of Attic decrees, one must insist at the same time that no internal necessity guarantees its universal validity. When ordinary rules of probability indicate a historical context later than that suggested by the exceptional letter-form, it is sounder method to assume an abrogation of our stylistic criteria than to reshuffle the historical sequence.

This conclusion is hardly radical. Too many exceptions to stylistic norms exist to make any one of them invariably decisive. Stone masons carved decrees in the Ionic alphabet fifty years before the archonship of Eucleides (403/2), when the practice became standard.^[18] Indeed, some inscriptions nineteenth–century scholars dated to the fourth century in deference to that criterion have now found their proper place because

[16] ATL, vol. 1, figs. 5, 8, 11, 12. Cf. Meiggs, JHS 86, 91–92, Mattingly, AJP 105 (1984), 340, and the following note.

[17] IG i 929 = ML 33 = Fornara 78; cf. SEG 33.34. The line containing four-bar sigma is considered a different hand from that preceding it (ML, p. 73). Other cases of four-bar sigmas in the corpus dated before 445 are collected and briefly discussed by Mattingly, AJP 105 (1984), 340. For examples of variant letter forms in a single inscription, see, from the Attic deme Rhamnous, ML 53 (dated c. 450–440 B.C. ), and the Chalcis Decree, ML 52 (dated 446/5).

[18] E.g., ML 51.

of their substantive requirements.^[19] Such anomalies are surely as "unallowable"—even harder to explain—as those involving letter-forms. It has already been observed that consistency is lacking in some decrees carved by one hand alone—Ionic letters are used randomly, as are different forms of the same letter. And as to style and the relative maturity of lettering, the inscription on the altar of Pythian Apollo^[20] displays maturity of execution twenty years ahead of its time. The inscription is not unique.^[21] We must therefore not be misled by the apparently "objective" evidence of the letter-forms into assuming its intrinsic superiority to other kinds of evidence. For "history," institutional and political development, is broadly governed by its own internal logic. For example, extreme measures of coercion in the ancient world presuppose an originally more moderate approach; complicated bureaucratic procedures do not precede but follow more rudimentary ones. The "tone" of a decree acquires its timbre from a people's developing attitudes, and this tone is subject to discrimination. Such tone is no less "objective" a criterion of date than the shape of a letter, even though the latter can be measured and denoted schematically, while the former requires more subtle analysis and exposition. A perceptible consistency in attitude and expression unifies the decrees of a specific period and demands less peremptory judgments than one based primarily on stylistic criteria.

[19] Ibid., and p. 137.

[20] IG i 761 = ML 11 = Fornara 37; cf. SEG 31.31. See ML, p. 20.

[21] See, e.g., the descriptions of the epigraphic styles in ML 31, 33, 37, 44, 45, and 52. Especially note the phrase "developed Attic letters except [three-bar sigma]" (pp. 80, 107, 111); see also nn. 16, 17, 18 above.