Normal Lengths of Prytany in the Athenian Year Meritt, Benjamin D Greek, Roman and Byzantine Studies; Summer 1976; 17, 2; ProQuest pg. 147 Normal Lengths of Prytany in the Athenian Year Benjamin D . Meritt W HEN DATES in Athenian decrees are given in terms of both the festival and the prytany calendars it has become clear since the study by Pritchett and Neugebauer in 1947 that any irregularity must be attributed to the festival calendar.1 The treat­ ment of the prytanies was quite impartial, and any restoration of an inscription which attributes more than 30 days or less than 29 days to any prytany in an ordinary year during the period of the twelve phylai is suspect. It has taken a long time to shake off the old idea of irregular prytanies which manifested itself so often in the restorations made by Kirchner in the Inscriptiones Graecae, but there is now no valid excuse for assuming such irregularities. If other restorations can be made which do not demand irregularities in lengths of prytany, they must be attempted.2 1 W. Kendrick Pritchett and O. Neugebauer, The Calendars of Athens (Cambridge [Mass.] 1947) 3-4. Their thesis is now generally accepted, with certain necessary modifications for the fourth century. In the fifth century, when there were ten prytanies, down ro and probably including the first prytany of 407/6, the prytany year approximated the solar year. Lengths of prytany were of 37 and 36 days. The first prytany of 407/6 (of 37 days) was probably the last of the quasi-solar year (cf ProcPhiiSoc 115 [1971] 114-17 with tables 2 and 4). From 407/6 down ro 308/7, still in the period of the ten prytanies, the lengths ofprytany in an ordinary year of twel ve months ,,,ere of 36 and 35 days, but their sequences were not always the normal four times 36 followed by sLx times 35 (B. D. Meritt, The Athenian Year [Berkeley and Los Angeles 1%1] 134; AJP 95 [1974] 279). After 307/6 when there were twelve phylai the lengths of prytany in an ordinary year of twelve months could be the same as the months or could be six times 30 followed by six times 29, or vice versa, or indeed any combination which allowed an impartial distribution within the year of six prytanies of 30 days and six prytanies of 29 days, in whatever order (Meritt, op.cit. 135-40). 2 Jon D. Mikalson, The Sacred and Civil Calendar of the .4thenian State (Prince ron 1975), emphasizes another criterion which must be observed in the restoration of texts. One should avoid dates for meetings of the assembly which conflict with major state festivals. This thesis is sound and admirably documented by Mikalson. As one of the first-fruits of attention to it the restoration of Boedromion 20 in 222/1 for a meeting of the assembly must be withdrawn (B. D. Meritt and John S. Trail!, The Athenian Agora, XV: Inscriptions, The Athenian Councillors [Princeron 1974] p.120 no.129 line 3). The date to be restored is [dySon £1T~ SlKa 7'pl'm Kat £JlKoCT[ £t rijc 1Tpv7'al'£lJ!l[ c], and the same retardation had taken place in the festival calendar which is evident also in the calendar equation of line 37 (op.cit. p.122). Boedromion 20 was the first full day of the Mysteries at Eleusis. 147 148 LENGTHS OF PRYTANY IN THE ATHENIAN YEAR In 1961 I suggested that in four separate years there may have been a prytany of31 days in an ordinary year.3 These suggestions must now be corrected. The first year was 258/7 B.C. when Thymochares was archon. This was represented by a text published in the Corpus (IG 112 700) as of the third prytany, with the phyle Antigonis or Demetrias and with the date [Bo7]o]po/A-tWvoc €V€t Kcd [VEC'Ct] in the festival calendar followed by a space of eighteen letters for the date by prytany. Mikalson (op.cit. p.64) gives the text again but leaves only five spaces for the name of the phyle. There is, of course, no phyle with only five letters in the genitive of its name. He also reduces the number of letters for the date by prytany from eighteen to sixteen. For the validation of any restoration these mistakes have first to be corrected. I republished the Corpus text in 1938, restoring the dates to suit an intercalary year,4 but when Pritchett and I came to draw up our archon list for the mid third century we found that the sequence of intercalary and ordinary years suggested strongly (if it did not prove) that the year of Thymochares should be ordinary,S and we made it so by assuming that one of the first two prytanies must have had 31 days. This violates the first principles of impartial lengths of prytany. In the somewhat later study by Pritchett and Neugebauer the authors held to an ordinary year but avoided any irregularity in the lengths of prytanies.6 They did, however, assume a possible irregularity in the stoichedon order of fifty letters on a line, and they allowed themselves a liberty in restoration by noting that frequently uninscribed spaces exist on the stone between phrases.7 Neither the assumption ofirregu­ larity in spacing nor of uninscribed spaces in the line where the date occurs is good if some better solution can be found. They also sug­ gested as an alternative that the date should be £VCXTH Ked. ElKocni, when, in addition to assuming an uninscribed space before the day, they thought it possible that the first three months had only 29 days 3 Op.cit. (supra n.l) 81 with n.9. 4 Hesperia 7 (1938) 110-12, 113. o William Kendrick Pritchett and Benjamin Dean Meritt, The Chronology of Hellenistic Athens (Cambridge [Mass.] 1940) 97. I Op.cit. (supra n.l) 81 with n.9. 7 There is much less disturbance in the order than Kirchner thought, as was noted in Hesperia 7 (1938) 113, and there is no evidence for any irregularity in the opening lines. There are indeed uninscribed spaces between phrases, but the one place where such an epi­ graphical phenomenon is never known to occur is between the date by month and the date by prytany (cf. AJP 95 (1974] 274-75). BENJAMIN D. MERITT 149 each. The assumption of three hollow months in succession depends on a theory, for which there is no evidence in historic times, that lengths of months were determined by empirical observation of the new moon.s In 1961 I still held to an ordinary year, with one of the first two prytanies having 31 days, but cited the alternative (which would be correct epigraphically) for an intercalary year. The year of Thymo­ chares appeared in my table of archons as ordinary.9 The matter could not be left there. After considering what the possibilities were for a normal restoration, epigraphically, to yield an ordinary year I later proposed the following restoration of the formula of date in lines 2-4 of the inscription: [Bo7Jo]pofLuvvoe EVEL Ka~ [V€Lat oyOOEL Kat €iKOeT€t Tfje 1TpVTaV€Lae],lO or with the spelling vEaL retained the extra letter needed to fill the stoichedon line might have been supplied by reading Oy80LEL instead of oySO€t. Both variants are well attested. The last day of Boedromion was the 88th day of the year (29+ 30+ 29), and this was equated with the 28th day of the third prytany (30+ 30+ 28). There has been no need to assume any irregularity of prytany length. I note again that the year was ordinary to clarify the confusion in my presentation of 1961. The second year in which I suggested a possible prytany of 31 days was 195/4, in a text first published in 1940.n Here the calendar equa­ tion was read as [B07JOPofLLwvoe EVEL Ka~ vEm] JfJOOfLEL Ka~ El [KOeT€L Tije 1TpvTaveLae]. The authors noted that the context seemed to call for the restoration of the month Boedromion. Since the decree praises epheboi "the probability is great" that this month name should be restored. The 27th day of the prytany is sure, and the question now is to find which date by month can be equated with it. The restoration of the name of the month as Boedromion would require that one of the first three prytanies should have had 31 days. As was the case when the assumption was made for IG 1I2 700 (see above) such a restoration would violate the law of impartiality in the lengths of prytanies. In an ordinary year, which this clearly is, the last day of no month before Posideon can be equated with the 27th day of any prytany. When the text was published in 1940 we were not so 8 Cf. Meritt, op.cit. (supra n.1) 1-15, 16-37,44; TAPA 95 (1964) 228-34; ArchEph 1968,87-88. 8 Meritt, op.cit. (supra n.l) 140-42,233. 10 Hesperia 38 (1969) 112-13. 11 Pritchett and Meritt, op.cit. (supra n.5) Ill. 150 LENGTHS OF PRYTANY IN THE ATHENIAN YEAR averse as we were after 1947 to an irregular length of prytany. Now we know better, and I restore: Chronology, p.ll1 [E7T' ......... apxov'T ]oe E7T/. 'Tije O[lv€iDOe €K'T"fJC 7TpV'TaV€L] [ac 7jL -------] Kpa'TOV l:'T/l-'axt[D'T/C Eypal-'l-'a'TEVEv' IIo] [CL8Ewvoe €VEL Kat VEaL] efJ861-'€L Kat €t[ KOC'T€L TfjC 7TPVTCCV€tCCC] K'T'\. While it is true that one expects the honors for the epheboi to be voted in Boedromion, none the less there are examples of votes longer delayed. A vote of praise for the epheboi of 185(4 was made in the tenth prytany of the following year (IG 112 900), and honors for the epheboi of 246(5 were voted in the year 244(3, after a delay of thirteen months, in the month of Pyanopsion.12 Surely, it is better to accept a delay of three months, as here suggested, than to violate the law of impartiality in lengths of prytany which a prytany of 31 days would mean. Here the first six prytanies were of 30 days each and the vote was passed on the 177th day of the yearP The third year for which I suggested a prytany of 31 days was 176(5 when Hippakos was archon. a As the inscription in question of this year was first published the suggested date in the prytany was I-'UXL Kat [7pLaKOe'TEi], but this in itself is not enough to assure that anyone prytany had 31 days, for the restoration I-'tiiL Kat [ElKOC'T€i] is apt and is equally possible. If the date was the 21st of the ninth prytany the day of the year was the 255th, granted that the first six prytanies were of 29 days each and the last six of 30 days each. This brings the month date to Elaphebolion, the one month par excellence when days were added extra ordinem to the festival calendar.Is The 9th day of Ela­ phebolion would normally have been the 245th day of the year. It was delayed by the addition of ten extra days, and in consequence the 12 See Xapte-rr7ptov Ele 'AvaC'TaetoV K. 'OpAavSov, A' (Athens 1%5) 194-%. 13 This decree is cited by Mikalson, op.cit. (supra n.2) 65, from Chrolwlogy for a restored date in Boedromion. 14 Pritchett and Meritt, op.cit. (supra n.5) 120. 15 In 271/0 the 9th of Elaphebolion was intercalated four times (Hesperia 23 [1950] 299) and at that time the festival date had been retarded in all by eight days. In 145/4 the Dionysia in Elaphebolion had been postponed as much as nineteen or twenty days (Agora XV pp.191-92 no.238; cf TAPA 95 [1964] 255). In general see also Meritt, op.cit. (supra n.1) 33,147-48,161-65,208. BENJAMIN D. MERITT 151 calendar equation may be restored to read [}L1Jv<)c 'El\acP1J{3oAU»VOC EvaTE:L] ECLOVTOC ILL«L Ka~ [ElKocni TijC 1TpvTavELac]. This is probable; other solutions are possible, but there is no longer reason to believe that any prytany in this year had 31 days. I choose to restore the 9th rather than the 10th day of Elaphebolion because of the improbability of a meeting of the ekklesia on Elaphebolion 10.16 The fourth year for which I suggested a prytany of 31 days in an ordinary year was 168/7 when Xenokles was archon (IG II2 945). This year is best taken as intercalary with the calendar equation EKLPO­ cpOpLWVOC EVEt Kat vEaL ILL«L Kat TpLaKoc-rei TijC 1TpV[ Tavetac ] representing the next to last (383rd) day of the year. The day was in fact EKLPO­ cpOpLWVOC EV1J Ka~ vEa1Tpo-r/.pa, but the distinguising epithet 1TpoTEpa (or E/l-{3oAL/l-OC) was not always added in a published text.!' This year will have had the normal sequence of prytanies of 32 days each. The same impartiality in lengths of prytany must be assumed for the fourth century when there were only ten phylai. I had taken the correspondences between the calendar equations of two inscriptions of the year 327/6 to show that the final prytany probably had 37 days.18 No prytany during the period of the ten phylai in the fourth century should have had more than 36 days or less than 35 days in an ordinary year. If the first five prytanies had 35 days each the equation in IG II2 357 is correct for Prytany VI 3, the 178th day of the year: EVEL [ " ,- J' K aL VEaL TpLTEt T1JC 1TpVTav Etac. 30 29 30 29 30 30=178 days 35 35 35 35 35+ 3=178 days