STOP Early Journal Content on JSTOR, Free to Anyone in the World This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in the world byJSTOR. Known as the Early Journal Content, this set of works include research articles, news, letters, and other writings published in more than 200 of the oldest leading academic journals. The works date from the mid-seventeenth to the early twentieth centuries. We encourage people to read and share the Early Journal Content openly and to tell others that this resource exists. People may post this content online or redistribute in any way for non-commercial purposes. Read more about Early Journal Content at http://about.istor.org/participate-istor/individuals/early- journal-content . JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people discover, use, and build upon a wide range of content through a powerful research and teaching platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please contact support@jstor.org. THE ATHENIAN CALENDAR By W. S. Ferguson 333/2* 1 3S7/6*t 2 336/5 3 335/4 i 5 6 332/1 7 331/Ot 8 330/9 9 329/8* 10 328/7 11 327/6t 12 326/5 13 325/4* 14 324/3 15 323/2 16 322/1 17 321/0* 18 320/9 19 319/8 July 16 5 " 24 " 13 1 " 20 9 June 29 July 17 " 6 June 25 July 15 3 June 22 " 11 May 31 June 18 8 " 27 Phrynichos, 01 PythodelOB, I Euainetos, O Ktesikles, O Nikokrates, I Niketes, O Aristophanes, O AriBtophon, I Kephisophon, O Euthykritos, O Hegemon, i Chremes, O Antikles, O Hegeslas, O Kephisodoros,0 Philokles.I Archippos, o Neaichmos, I ApoUodoros, o 318/7t 317/6* 316/5 315/4 314/3 313/2* 312/1 311/Ot 310/9 309/8*t 308/7 307/6t 306/5 305/4*t 304/3 303/2 302/1 301/0* 3C0/9 June 16 July 5 " 23 " 12 1 " 19 Aug. 7 July 27 " 17 5 June 25 July 14 4 June 22 July 11 1 " 20 8 " 27 Archippos, I Demogenes, i Demokleides, o Praxibulos, o Nikodoros, I Theophrastos, I Polemon, o Simonides, o Hieromnemon, O Demetrios, o Kairimos, i Anaxikrates, O Koroibos, O Euxenippos, I Pherekles, O Leostratos, I Nikokles, O Klearchos, I Hegemachos, o 298/7 297/6* 296/5t 295/4 294/3 293/2*t 292/1 291/0 290/9 289/8* 288/7 287/6t 286/5 285/4* 284/3 283 2t 282/1 281/0* July 16 5 23 12 2 21 9 June 29 July 18 Aug. 6 July 25 " 14 Aug. 3 July 23 Aug. 10 July 30 Aug. 18 8 July 27 Euktemon, O Mnesidemos, i Antiphates, o Nikias, o Nikostratos, I 01ympiodoros,0 Charinos? o Philippos? i Kimon, i Dickies, O Diotimos, O Isaios, i Euthios, O Xenophon, i XTrios, O Telokles? 1 Menekles, O Nikias Otr., O Aristonymos, O 1 0=demon5trated ; o=hypothetical, ordinary year. I=d6monBtrat6d; i=hypothetical. Intercalary year. The first part of this table is compiled, with modifications, from that given by Unger in Mueller Hand- buch 12, p. 755. Seeing that 331/0 did actually begin on July 9 (ibid., p. 754) the dates of the months for the first cycle or two are approximately correct— If the intercalations were made as set forth in the table. For the calendar character of 326/5 B. o. see below, p. 388, n. 1 ; for that of 305/4 see 18. II, 238, which SundwaU (Acta »ocietati» ecientiarum fennicae, 1907) has correctly put in this year. The archons after 294/3 B. c. are dated as in the Priests of Asklepios, p. 132. For the calendar quality of the archon years see Beloch Klio I, pp. 412 ff., and Oriech. Gesch. Ill 2, pp. 52 ff. where the year of Nikias Otr., which is an ordinary year, is wrongly treated as undetermined ; also SundwaU, Ofversigt af Finska Vetenskaps-Societetens Fdrhandlingar XLIX, 1906-7, No. 9, pp. 75 ff., cited below as Untersuch, from the title of the monograph, which is Vntersuchungen iiber die attische Mllnzen nemeren Stils. 280/9 279/8 278/7 O 277/6 276/j 6 275/4 7 274/3 I 8 273/2 9 272/1 10 271/0 I 11 270/9 O 12 269/8 13 268/7 O 14 267/6 15 266/5 O 16 265/4 17 264/3 18 263/2 19 262/1 Gorgias Anaxikrates Demokles — laios? Eubulos Polyeuktos Hieron Pytharatos IG. 11 5 252c. Philokrates Peithidemos Diognetos Antipatros 261/0 260/9 259/8 258/7 257/6 256/5 255/4 254/3 253/2 252/1 251/0 250/9 249/8 248/7 247/6 246/5 245/4 244/3 243/2 O Arrheneides O Kleomachos Diogeiton Olbios 01 Lysiades I Kallimedes O? Qlaukippos O Thersilochos 242/1 241/0 240/9 239/8 238/7 237/6 236/5 235/4 234/3 233/2 232/1 231/0 230/9 229/8 228/7 227/6 226/5 225/4 224/3 I Charikles Lysias Kimon I Ekphantos O Lysanias O? Diomedon Jason O Heliodoros Leochares O Theophilos O Ergochares Niketes Antiphilos 223/2 222/1 221/0 220/9 219/8 218/7 217/6 216/5 215/4 214/3 213/2 212/1 211/0 210/9 209/8 208/7 207/6 206/5 205/4 Kalli- Menekrates Thrasyphon O? Chairephon Pantiades? I Diokles Euphiletos Herakleitos OI Archelaos Aischron Philostratos I Antimachos PhanostratoB Kallistratos? [Classical Philology III, October, 1908] 386 The Athenian Calendar 387 1 20t'3 2 203/2 3 202/1 Phanarchides? 4 201/0 5 200/9 6 199/8 7 198/7 8 197/6 9 196/5 10 195/4 11 194/3 12 193/2 13 192/1 14 191/0 15 190/9 16 189/8 17 188/7 n Symmachos 18 187/6 Theoxenos 19 186/5 I Zopyros 185/4 184/3 183/2 182/1 181/0 180/9 179/8 178/7 177/6 176/5 175/4 174/3 173/2 172/1 171/0 170/9 169/8 168/7 167/6 Eupolemos 166/5 01 Achaios 165/4 O Pelops Hermogenes 164/3 Euerg — ! Timesianaz 163/2 162/1 161/0 160/9 Erastos Poseidonios Aris tolas Epainetosi Philon 159/8 158/7 Aristaicliinos Anthesterlos Hippakos 157/6 Kallistratos 01 Sonikos 156/5 Mnesitheos 155/4 154/3 Zaleukosl I Tychandros 153/2 PJiaidrias De- 152/1 Lysiades AleiandroB? 151/0 11 Archon O Eunikos 150/9 I Epikrates I Xenokles 149/8 Theaitetos Nikosthenesi 148/7 Aristophon 147/6 146/5 145/4 144/3 143/2 6 142/1 7 141/0 8 140/9 9 139/8 10 138/7 11 137/6 12 136/5 13 135/4 14 134/3 15 133/2 16 132/1 17 131/0 18 130/9 19 129/8 Metoni Dionysiosi Hagnotheos Apollodoros Timarchos Herakleitos Nikomachos 01 Metrophanes 01 Ergokles O Epikles Demostratos Lykiskos 128/7 Dionysios 127/6 Theodorides 126/5 Diotimos 125/4 Jason 124/3 Nikias & Isigenes 123/2 Demetrios 1Z2/1 O Nikodemos 121/0 Xenoni 120''9 Eumachos 119/8 Hipparchos 118/7 O Lenaios 117/6 Menoites 116/5 O Sarapion 115/4 Nausias 114/3 113/2 Paramonos 112A O Dionysios Jnly 18 111/0 i Sosikrates July 7 110/9 O Polykleitos July 26 109/8 108/7 107/6 106/5 105/4 104/3 103/2 102/1 101/0 100/9 99/8 98/7 97/6 96/5 95/4 94/3 93/2 92/1 91/0 O Jason July 14 Herakleides O Aristarchos O Agathokles Theokles Echekrates O Medeios Theodosios O Prokles Argeios Argeios Herakleitos O Demochares Dioklesi Medeios A glance at this table shows that an effort, but not a very serious one, was made to avoid the succession of three ordinary and two intercalary years. Thus the elements which normally recur in each cycle are 01, 001, and it may be that theoretically they were arranged in the following series: 01, 001, 001, 001, 001, 01, 010. But it is possible to obtain seven intercalary years in the first cycle (sixth from the promulgation by Meton in 433/2 B.C.) only by admitting three successive intercalary years between 322/1 and 318/7 b. c, as well as the four successive 388 W. S. Fergdson ordinary years which are attested in 326/5/ 325/4, 324/3, and 323/2. This no one will do except under compulsion; hence we have preferred^ to make this cycle fall a month short of its due, for which there is the additional argument that the second (or seventh) cycle began with an intercalary year. Accordingly, we have regarded the extra month of 318/7 b. c. as wiping out the defect of the preceding cycle. So far so good. But in 314/3 and 313/2 b. c. we find two successive intercalary years,' to even up which we are obliged to enter either two groups of three suc- cessive ordinary years, or one group of four ordinary years in the period between 312/1 and 306/5 B. c. In the third (or eighth) cycle, also, three ordinary years follow one another at the end (283/2-281/0 B.C.), which argues for the existence of two suc- cessive intercalary years somewhere earlier in the period (e. g., 292/1, 291/0 B.C.). We are consequently bound to deny the maintenance of any invariable sequence of ordinary and inter- calary years within the cycle. We need to recognize only an additional effort on the part of the Athenians to bring the lunar and the solar periods into the same relation to one another at the 1 1 have restored as follows the inscription of this year, which is published in 'E^. 'Apx., 1898, p. 7, No. 2 ; cf. Plate I : ['A]vo<; ir, ^Tt? rjv Kara tov^ Al'yinrrlov'i fiia Koi K Tov <^aijLevo)0, eto? [ttj]? yevofievrj^ iirl IIoXjuAcXetTOu 2/«:t[/)0(^o/9t]a)- 1*09 lA, Ka[Ta Se tov^<; At7U7rTt[ow tow Ilauji't t?j? IA, [xaTa Se to Mi\]'ja-iop . From this document, as Rehm and Diels' have pointed out, it is clear that the 27th of June corresponded with the 21st of Phamenoth, or the 201st day of the Egyptian year, on the 13th of Skirophorion in 433/2 b. o., and that it cor- responded with the 14th of Payni, or the 281st day of the Egyp- tian year, on the 14th of Skirophorion in the archonship of Poly- kleitos. That is to say, the solstice fell 80 days later in the year because of the interval, and since it fell back one day in every four years through the well-known failure of the Egyptians to intercalate a 366th day in leap years, the interval must have been from 320 to 324 years; in other words, Polykleitos must have been archon in 113/2, 112/1, 111/0, or 110/9 B. 0. Now in his archonship the 27th of June coincided with the 14th of Skiro- phorion ; hence the 30th of Skirophorion coincided with the 13th of July and the new moon was visible on the 14th of July. But in 112 the new moon was visible on the 29th of June or the 28th of July,^ in 111 on the 18th of July, in 110 on the 7th of July, but in 109 on the 14th of July precisely. Hence Polykleitos was archon in 110/9 B. c. on astronomical grounds as well as because of the secretary canon; hence, moreover, the Athenian calendar was flawless in this year, i. e., at the end of the 12th (or nine- teenth) Metonic cycle. There is thus a certain warrant for believing that the effort was made generally to assign seven inter- ^Sitzungsb. d. berl. Akad., 1904, pp. 92ff. ''Unger JlfwHer's Handbuch P, p. 764; Sundwall Untersuch., p. 82. 390 W. S. Ferguson calary and twelve ordinary years to every cycle, but none what- ever that this was done with such regularity in the disposition of these two elements as to enable us to predict the calendar quality of any given year. In other words, the attempt of Beloch' and others to date third century B. c. archons by means of the calen- dar had from the start no chance of success; and, furthermore, the evidence which has been advanced to prove that the Metonic cycle was not introduced in the year of its discovery (433/2 B.C.) proves nothing of the kind, seeing that this cycle gave such a large measure of internal freedom. The empirical data won from the inscriptions contains nothing contradictory with the state- ment of Diodoros^ made in his record for Apseudes' archonship (433/2 B.C.): iv Se rat? 'KOrfvait; Mercoj/ o Y[avaavl,ov (lev vlbi, B€Bo^afievoi ry ivveaKaiBeKaeTr/piBi ov Biayp-ev- Bovrai T^? aXrjOeia'i. That is to say, Meton's nineteen-year cycle started at the summer solstice — 13th of Skirophorion 432/2 B. o. — a fact amply corroborated by the Milesian parapegma quoted above. The current doctrine is that this date was merely the point of departure for Meton's calendar calculations, and that his era was not actually employed in Athens till about 100 years later; that what was set forth in 432 b. c. was the table of computations alone. It cannot be denied that the passage of Diodoros will admit of this interpretation, but it does not prescribe it, and despite the great names which have given to it its present pres- tige, it has never quite ousted the common-sense interpretation of the passage in question and of the other evidence. For in the first place it is abundantly proven by the part played by Meton in ^Griech. Gesch. Ill 2, pp. 50ff. {Klio I, pp. 411fE.' 2 XII 36. The Athenian Calendak 391 the comedies of Aristophanes' and Phrynichos'' that he was a man with a popular reputation such as does not fall ordinarily to the lot of a scholar who has made a contribution to astronomical science. Meton must have succeeded in getting his scheme dis- cussed by the people of Athens. It guaranteed, what the octae- teris had failed to secure, astronomic correctness at the end of each cycle, and it set forth what was ideally the best grouping of the lunar months in the several years; but, as we have seen, it gave the ^fio^ considerable freedom of intercalation within the limits of the larger period — a freedom which it did not wait till 326/5 B. c. to use for the first time, as is clear both from the pas- sage of IQ. II 4, p. 59, quoted below,' and from the complaints of Aristophanes in the Clouds^ (414 b. o.) that the gods were not getting their sacrifices at the right times. The notoriety of Meton is thus the first argument that he published his adjustable almanac [parapegma), not on the basis of a theoretical nineteen-year cycle — which would have had little practical value if the days, months, and years of Athens depended upon a different scheme — but on the basis of a nineteen-year cycle which he had succeeded in having made the official calendar of the state. The second argu- ment is of a negative character. It is difficult to understand how the date, 13th of Skirophorion in the archonship of Apseudes, came to be preserved if the new calendar did not go into operation at that time, since, if Meton's calendar had been first introduced 100 years after his akme, however much indebtedness to him might then be acknowledged, the innovation would be accredited popularly to the nameless man of afiPairs who effected the reform, not to the ancient computer of the system. And if Meton's almanac and calendar had been published only privately in 433/2 b. o., but used popu- larly in subsequent times, there was no reason for the state which first introduced it officially in 338/7 b. c. to regard further the five cycles which had already elapsed: it would then have sufficed to establish, let us say, the 13th of Skirophorion or thereabouts 337 b. c. as the starting-point of the Metonic era. This would '^ Birds, U. 992ff. ^Monotropos quoted by Van Leeuwen on Aristophanes' Birds, loc. dt. 'P. 42. 4Iil. 615ff. 392 W. S. Ferguson have been the historically important date — the one which should have impressed itself upon our tradition; as, of course, it has not done, for the obvious reason that it never existed.' The determination of the calendar character of the years in the second century b. o. is made difficult through the existence during a period — of which the limits thus far determinable on epi- graphical evidence are 166/5 and 122/1 B. 0. — of two distinct years at the same time, each with its distinct set of months, one, however, designated war' apxovra, and the other Kara deov. When- ever a document is dated in the two systems the prytanies coin- cide with the months Kara deov, and the months ^ar' apxovra lag from a full month to one day only behind the months kuto, deov; but documents, both in the years with two sets of months, and in years in between these, are also dated simply in terms of one year, which is, moreover, so far as can be judged,^ the old lunar year. This fact tends to corroborate the view of Adolf Schmidt that the year kut^ Oeov, with which the prytanies are in relation when the two systems exist side by side, is also the old lunar year — the year which entered into the Metonic cycle; and it is on this hypothesis that the table given above has been constructed. If this is true, what is the year /car' dpxovra? The solar year, according to Adolf Schmidt.' This, however, is impossible, for in the year 122/1 b. 0., in which the archon Nikodemos is located on too many evidences to be moved on any account, the 8th embolimos of Boedromion kut apxovra coincided with the 9th of Boedromion KaTo, deov. This, however, proves, as J. Sundwall 'The practice, begun in 338/7 B.C., of dating documents by giving not only the day of the prytany, but also the day of the month on which they were passed, simply reflects a change in the management of the public archives, which, as usual, accom- panied an epoch year. There exists, therefore, no more reason for making 338/7 b. o. the first of a Metonic cycle than for inferring a calendar change thirty years earlier {IG. II 52) when the day of the prytany was first put upon the public decrees. More- over, a dating by month and prytany from as early as 346/5 b. c. is found in the Samian cleruch document published by O. Ourtius Inschriften und Stud, zur Gesch. v. Samos, Progr., Lubeck, 1877. 2 Sundwall Untersuch., p. 89, who, however, errs in assuming that it is a solar year where we have no means of judging. It is unthinkable that the Athenians changed from a lunar to a solar year and vice versa four or five times in the last two-thirds of this century. sjahrb.f. Phil. 1884, pp. 651 f. The Athenian Calendae 393 has clearly pointed out,' that, if either of these was a solar year, the summer solstice and the new lunar month must have practi- cally coincided in June 27th, 122 b. o. But new moons were visible in 122 b. c. on June 10th and July 9th. Consequently, kut' dpxovra does not mean solar year any more than kuto, 6e6v does."* What then does it mean? We have as yet insufficient data to answer this question positively, but the conjecture is per- missible that it was a year of which the limits were determined by the archon, i.e., by his period of service. We might then think that the term failed to jibe with the other or senatorial year through a difference of decision as to the intercalation of a month or a day, but this explanation fails to satisfy one case of double dating by months, viz., IQ. II 408, which is a decree passed in the archonship of Metrophanes (133/2 B.C.) on the 29th of Elaphebolion /car' ap')(ovTa and the 12th of Munychion Kara 0e6v. Here the difference is 12 or 13 days. How did it come that the archon year began 12 days later than the senatorial year, this is the question, since it is inconceivable that 12 days were intercalated in the archon year after it had once begun. There was nothing new in the existence of a separate archon year. In fact it had existed throughout the greater part of the fifth century b. o., as Bruno KeiP demonstrated some years ago. He claimed further that in that century the archon year had its series of lunar months, while the senatorial year had apparently only its series of prytanies, but A. Mommsen* has since shown that the archon and the senatorial year probably belong together. According to Keil's view, to which Mommsen assents in this particular, the separate senatorial year was abandoned in 410 B. o. 1 Untersuchungen, pp. 82 f . 2 This, in my judgment the only methodical conclusion, Sundwall fails to draw. Instead of doing so, he tries to change the date of Nikodemos ; and, in fact, there was a coincidence of solstice and new moon in the year 121 b. c. Hence he puts Nikode- mos in 121/0 b. c. — with the disastrous results noted below. Sundwall encounters a similar difficulty in the oases of Metrophanes and Achaios ; the first must be shifted from 133/2 B. c. to 144/3 B. c, the latter from 166/5 B. c. to 165/4 B. c. This is, however, admissible only when Nikodemos is put in 121/0 b. c. ; and not even then, since the successor of Metrophanes, Erg. (who is, doubtless, Brgokles of 132/1 b. c. pace Sund- wall), and Meton (Klio VII, p. 222) would collide in 143/2 B.C. 3 Hermes XXIX, pp. 32 ff. *Philologos 1903, pp. 348 ff. Of., however, ibid. 1902, pp. 214ff. 394 W. S. Ferguson As we have seen, the divergence appears again in the second cen- tury B. c. This time the archon year has its own series of months, and it has different limits from the lunar-calendar year, which, how- ever, has now a fixed relation with the senatorial year, as during the two centuries which had intervened. Of course, the archon year had coincided with the calendar year for the same two centuries, and, since there was still a rough coincidence between the two years, the name of the archon was still employed to designate both. The Athena-year began 28 days after the calendar year, and the ephebe year two months later; so that there was no reason why the archon should not have his separate year also. But it is curious that it should have a different set of months, i.e., should stand in no fixed relation to the calendar year. This shows that it had its own cycle and its own reason for its beginning and ending, just as the archon-senatorial year had in the fifth cen- tury. In the earlier period, according to Brimo Keil's reckoning,' it had an ordinary length of 360 days and an intercalary length of 390 days. There were seven intercalary years in every forty. This may or may not be correct; at any rate the archon-senatorial year did not fit in between the solstices, as is clear from the fact that in 411 b. o. it began on the 14th of Skirophorion,'' whereas in this year the solstice occurred, not 15 or 16 days before the new moon which was visible at the first of Hekatombaion, but 27 days before that event.* Hence we need not be surprised that the archon-year in the 2d century b. c. disregarded the solstices also. We have no means of establishing the cycle which was followed in this epoch ; for nothing of any value can be obtained from the double datings in 133/2 b. c. and 122/1 b. o. so long as the begin- ning point of neither set of months is determinable. The upper epigraphical limit of the divergence between the archon and the calendar or senatorial year has been set at 166/5 b. o. in the con- viction that Sundwall* is right in placing lO. II 4516 ii, and II 433 in the same year; in which case both, and with them the archon Achaios and the secretary 'H/aa/cXe'wi' Nai'[i/]a«:ow Wmvpi- Br}<;, belong to 166/5 b. 0. If this were approximately the real '^Hermes loc. ait., pp. 321ft. "Unger MUllers Handbuch P, p. 752. 2 Aristotle Const, of Athens 32. * Op. cit., pp. 86 ft. The Athenian Calendar 395 upper limit we might think that the acquisition of Delos in the winter of 167/6 b. c. and the steps taken thereafter to shift the beginning of the Delian year round from Gamelion to Hekatom- baion had something to do with the establishment of a separate series of archon years. But our knowledge of this effort depends as yet upon HomoUe's interpretation' of certain unpublished Delian documents, so that we are not at present in a position to discuss its details. Moreover, Sundwall has discovered that there existed from ca. 180 b. o. forward a different year from the calen- dar year, which, moreover, possessed a distinct set of months or subdivisions, and was used to define the terms of office of the so-called third mint magistrates. This was, doubtless, the archon year, both because we know of no other it can have been, and because Sundwall has made it very probable^ that the third mint magistrates were Areopagites; i. e., ex-archons, to whom the use of the archon year would, accordingly, be most natural. The data collected by Sundwall from the coins shows that the archon year began almost, if not altogether, as frequently after the senatorial calendar year as before it (10 to 12)* — a relation which is expli- cable on the hypothesis that either of them is the solar year only by assuming that the Metonic or Kallippic cycle was regularly violated during this century; since in it the lunar year began prior to the solar year only twice in every nineteen years. In the four cases* known to us from the inscriptions the archon year began after the senatorial or calendar year. "We have, accord- ingly, to maintain, on any interpretation, that the relation thus suggested is accidental, since the coins show one year to have begun before the other only slightly more frequently than after it. From one of the inscriptional cases already referred to we learn that the archon year began more than a month later than the cal- endar year;^ that is to say, precisely as in the year to which the monetary series Herakleides-Eukles belonged." Is it thinkable ^BCH. V 25 ; XXVII 63, n. 1. 2 Op. cit., pp. 10, 69f. 8J6id., pp. 93ff.