N.B. I have made no specific reference to Theodor Hopfner's Fontes Historiae Religionis Aegyptiacae, 5 parts, Bonn, 1922–25, but I have depended on it constantly. Hopfner's extensive index provides a conspectus of Egyptian religion readable in itself and of a kind not available elsewhere. I am happy to acknowledge also my indebtedness to A. D. Nock for valuable comment and bibliography; without his kind intervention I might easily have overlooked the Bacchic inscription in the Metropolitan Museum, which has proved of basic importance for the interpretation of P. Mich. Inv. 4686. J. G. Winter has kindly consented to my use in this paper of our joint transcript of the papyrus, which is intended ultimately to form part of an edition of private letters on papyrus in the Michigan Collection.

This paper was read at the Annual Meeting of The American Philosophical Society on April 25, 1947.

¹ For a general survey of the Graeco-Oriental cults see Cumont, , Les Religions orientates dans le paganisme romain, 4th ed., Paris, 1929Google Scholar; more briefly, Gressmann, , Die orientalischen Religionen im hellenist.-röm. Zeitalter, Berlin-Leipzig, 1930Google Scholar.

² The exact meaning of this statement in the historical context is still open to question. See footnote 93.

³ The most recent detailed account of Sarapis is by Wilcken, , Urkunden der Ptolemäerzeit I, Berlin-Leipzig, 1927Google Scholar, introd., where all earlier views, among them theories deriving Sarapis from Babylonia, Syria, and Pontus, are reconsidered with the author's usual restraint and penetration. See also Pauly-Wissowa, R.-E. 2te Reihe, I, art. Sarapis (Roeder), 2394–2426. For the latest vue d'ensemble, executed with characteristic brilliance, see Jouguet, , Trois études sur l'hellénisme, Cairo, 1944, 120–125Google Scholar. The ancient sources are conveniently arranged by Schmidt, R.G.V.V. VIII, pt. 2, 1909, 49ff. On the preëminence of Isis see Anrich, , Antikes Mysterienwesen, Göttingen, 1894, 44Google Scholar; Dow, H.T.R. XXX, 1937, 231.

⁴ Plutarch, de Iside et Osiride 28 (361F–362B); Plutarch identified Sarapis with Osiris (ibid.; 61 [376A]). Aristides, εἰς Σάραπιν(45, ed. Keil). For the spread of the Egyptian cults through the Greek world see Rusch, De Serapide et Iside in Graecia cultis, Diss. Berlin, 1906; Brady, Missouri Studies X, 1935, No. 1 (esp. his concluding sentence: “By the time of Augustus, the cult of Sarapis and Isis had spread throughout all Greece and remained the dominant cult there, very probably, until displaced by Christianity”) ; Dow, , H.T.R. XXX, 1937, 183–232Google Scholar. By the 2nd cent. A.D. the Sarapis cult was entrenched as far north as Cologne, where he was worshipped by a cline or dining fraternity (Salač, Berl. Phil. Woch. XXXIV, 1914, 253–5). His hold on the popular mind of Egypt is reflected in private letters on papyrus written at Alexandria. In the 1st cent. A.D. a formula of obeisance to Sarapis begins to be used after the prescript, and in the 2nd cent, examples are plentiful; they taper off and finally disappear in the 3rd cent. (Ziemann, Diss. Hal. XVIII, 1911, 321–3; Exler, Diss. Cath. Univ. Amer., 1923, 108–110). Its suppression in the 4th cent. may be related to the bitter struggles of that critical period, when Egyptian paganism faced an advancing Christian front (cf. Bell, H.T.R. XXXVII, 1944, 186, 204).

⁵ Pausanias 1.42 (18.4) states that “the Egyptians have temples of Sarapis; the most magnificent belongs to the Alexandrians, and the most ancient is in Memphis.” As shown in an inscription discovered by Alan Rowe at Rhacotis in 1943, Ptolemy III erected “the temple and the temenos” of Sarapis. This need not mean that he was responsible for the original foundation of the temple, but as Jouguet rightly observes (Trois Etudes, 123, n.1): “Pour le moment, on ne peut douter que Ptolémée III n'ait ordonné des constructions importantes, peut-être les plus importantes.” For the Alexandrian temple see Calderini, , Dizionario I, 1935, 140–6Google Scholar; for the temple at Memphis, Wilcken, U.P.Z. I, pp. 7 ff.

⁶ After reciting the typical miracles of Sarapis, Aristides 45, 29–30, ed. Keil, adds that temple libraries furnish examples without limit and applies to Sarapis the opening verses of Aratus' Phaenomena: “Full are the markets, says the poet, and the harbors and the city squares of those who recount his exploits in detail.” P. Oxy. XI, 1382, 2nd cent. A.D., states that “this miracle is deposited in the libraries of Mercurius.” Cf. Weinreich, , Neue Urk. z. Sarapisrel., Tübingen, 1919, 13 f.Google Scholar; Höfler, , Sarapishymnus d. Ailios Aristeides (= Tübinger Beitr. 27, 1935) 103 f.Google Scholar; Wilcken, U.P.Z. I, p. 84.

⁷ Anrich, Antikes Mysterienwesen, 44. The long road that Sarapis had travelled since his separation from Osiris-Apis is signalized in the solar formula Ζεὺς “Ηλιος Σάραπις (Weinreich, Neue Urkunden, 9, 14, 27ff.), which cannot be traced back beyond the 2nd cent. Peterson, A.D., Εἱς θεός, Göttingen, 1926, 235f.)Google Scholar. “Ηλιος Σάραπις occurs as early as 94 A.D. (Roeder, R.-E., 2te Reihe, I, 2421). Accompanying this movement toward a universal henotheism, if we may so distinguish 4th cent, paganism from Judaeo-Christian monotheism (Nock, Conversion, London, 1933, 158f.; Nilsson, , H.T.R. XXXVI, 1943, 255f.Google Scholar, on “Pantheos”), there was from the 2nd cent, an increasing emphasis on moral purity (Cumont, Religions orientales, 85f.).

⁸ Theodoret, Hist. eccl. 5.22.4–6; cf. Weinreich, Neue Urkunden, 23f. The public life of Sarapis had been prolonged by Julian's emergence as imperial champion of paganism (361–3) just when the Christian forces were on the verge of total victory, but more especially by the internecine war of Athanasius and Anus.

⁹ Nilsson, , H.T.R. XXXVI, 1943, 269Google Scholar: “At all events the mysteries of Isis are not of purely Egyptian origin. The Egyptian cult of Osiris did not include mysteries and the next life was to the Egyptians not dependent on an initiation, but on the performance of funeral rites.”

¹⁰ Nock, Conversion, chap. IX; A.J.A. XLV, 1941, 578f.; Reitzenstein, Hellenist. Mysterienrel., Leipzig-Berlin, 1910, 28f.; De Jong, de Apuleio Isiacorum mysteriorum teste, Diss. Leyden, 1900, 99–133.

¹¹ Foucart, Mystères d'Éleusis, 401–3.

¹² Gressmann, Orient. Rel., 44. It is characteristic of the broad inclusiveness of paganism that initiations could be and were sometimes multiplied (Apuleius, Apol. 55; cf. Nock, Conversion, 114f. Willoughby, Pagan Regeneration, Chicago, 1929, 31f.). Not many had the fervor or the means to justify the attention that Lucius received; he is not the ordinary initiate of public mysteries, but becomes one of a small élite (Nock, Conversion, 56f.; Nilsson, H.T.R. XXXVI, 1943, 273).

¹³ Aristides 45.17 (ed. Keil): καὶ ἱερὰ καὶ τελετὰς καὶ τιμὰς πὰσας; cf. Höfler, Sarapishymnus, 50ff.

¹⁴ Nilsson, loc. cit.

¹⁵ Aristides 49.48: τοιαῦτα ἦν τὰ τῆς τελετῆς. Widely used for the performance of a μνστήριον, τελετή is purely a religious term (Zijderveld, Τελετή, Diss. Utrecht, 1934; cf. Nock, Conversion, 28). Aristides represents his knowledge of the rites as derived from a dream. That the dream is a literary device used to escape censure for public exposure of mysteries, hardly needs demonstration, but it is instructive to compare the curious hypothetical situation presented by Sopater (5th cent. A.D.; Rhet. Gr., ed Walz, 8, 110) in which a young man dreams that he has been initiated into the Eleusinian mysteries. So far as I know, the only documentary evidence that may point to mysteries of Sarapis is Mendel, B.C.H. XXIV, 1900, 366f.: Σαράπιδι καὶ Εἴσιδι εὐχαριστήριον οἱ … μύσται καὶ δεκατισταί (Prusa, 2nd. cent. A.D.). But δεκαδισταί and other words of the same kind were used much earlier on Delos for groups of devotees who assembled on the corresponding days of the month (Roussel, Les Cultes égyptiens à Délos du III^e au I^er siècle av. J.-C, Paris-Nancy, 1915–16, pp. 100, 253). The suggestion lies close to hand, in view of the wording of the Bithynian inscription, that these also were μύσται. Although the extant evidence does not carry the mysteries of Sarapis back beyond the 2nd cent. A.D., they are probably much older. Wilcken, U.P.Z. I, p. 75, is sceptical of their existence in the Ptolemaic period and points to the total absence of evidence. He means, of course, direct evidence, but this is to deny any value to the indirect evidence in Tacitus, Hist. 4.83, and Plutarch, de Is. et Osir. 28 (362A), or to the historical probabilities. Nock, Conversion, 56f., although using a negative statement, is closer to truth: “It is not clear that initiations of the later type had formed a part of the cult outside Egypt in the Hellenistic period. … To sum up, the cult was for the majority of its worshippers not a mystery cult, and in many places in the Hellenistic period proper it was not a mystery cult at all.” Religious mysteries were deeply imbedded in Greek experience, and their chief exponent was the cult of Demeter at Eleusis in Attica. After Alexander mysteries were adopted widely by Greek cults everywhere. Nilsson, Studi e materiali di storia delle religioni X, 1934, 8, speaks of “l'inclination universelle de l'âge hellénistique pour les cultes mystiques.” According to Tacitus and Plutarch it was Ptolemy Soter who brought the Eumolpid Timotheus from Eleusis to Alexandria and used him as well as Manetho in developing the new religion of Sarapis. Unless this was a mystery-cult long before the 2nd cent. A.D., the introduction of Timotheus into the story is pointless (cf. Cumont, Religions orient., 91, 232, n. 4). Much depends on the source or sources used by these authors. The discussion, as it now stands, is not illuminating, but on two points there is an implied agreement: the tradition is perhaps as old as Sarapis himself and may be based on the official version. For the problem of the sources see Schmidt, R.G.V.V. VIII, pt. 2, 1909, chap. III.

¹⁶ Tacitus, Hist. 4.83; Plutarch, de Is. et Osir. 28 (361F).

¹⁷ Nilsson, , Griech. Feste, Leipzig, 1906, 350–2Google Scholar; Jouguet, Trois études, 122; Roeder, R.-E., 2te Reihe, I, 2425.

¹⁸ For the meaning of “Greek” in the late Hellenistic Age see the conclusion of this paper.

¹⁹ Osiris-Apis had no career outside Egypt, any more than did his counterpart, Osiris-Mnevis, the bull-god of Heliopolis. The total separation of Sarapis from the animal form of his predecessor is a much deeper break with Egyptian thought than is generally allowed. If this had been the whole story, he would have been left with slight roots in native sentiment; as it was, he became the god without a myth (Roeder, R.-E., 2te Reihe, I, 2418; Weinreich, Neue Urkunden, 10). It was through identification with Osiris that Sarapis was restored to the main stream of Egyptian tradition. Gradually Sarapis replaced Osiris at the side of Isis, and it is significant that the new mysteries did not require the old myth, although they perpetuated its central notion of immortality obtained by identification with the god (Cumont, Religions Orient., 91f.). With regard to Delos, Nock, Conversion, 53, makes the interesting observation that “the worship introduced was not the original Memphite cult of Osorapis but the hellenized cult of Sarapis and Isis: even at this early time it took the credit of possessing immemorial antiquity.” Sarapis had supplanted Osiris; Isis is the spouse of Sarapis (Roeder, R.-E., 2te Reihe, I, 2420). But like Osorapis (= Osiris-Apis), Osiris also continued his independent existence. Osiris and Sarapis were at once the same and not the same (cf. Plutarch, de Is. et Osir. 28 [362B]; Perdrizet, Terres cuites Fouquet, p. 73). At death an Egyptian became an Osiris, as before, never an Osorapis or a Sarapis. In the same way, as we know from Apuleius, Metamorph. 11, the mysteries of Osiris were not suppressed by the emergence of mysteries of Sarapis.

²⁰ Deneken, de Theoxeniis, Diss. Berlin, 1881; Schmidt, R.G.V.V. VIII, pt. 2, 1909, 105–8; Wissowa, Religion u. Kultus (= Mϋller's Handb. V, pt. 4, 421ff.; Pauly-Wissowa, R.-E. A. 5. 2, art. Theoxenia (Pfister), 2256–8; Eitrem, Symb. Oslo. X, 1932, 34f. For κλῖναι of Isis see Pauly-Wissowa, R.-E. XI, art. Kline (2. Ziebarth), 861; Roussel, Cuites égyptiens, p. 285, n. 5. Nilsson, Studi e materiali di storia delle religioni X, 1934, 12, speaks of symposia and banquets as “si fréquents dans toutes les associations religieuses.”

²¹ Aristides 45.27 (ed. Keil): προιστάμενοι δαιτυμόνα αὐτὸν καὶ ἑστιάτορα.

²² Cf. Höfler, Sarapishymnus, 93–96; Nock, H.T.R. XXXVII, 1944, 150f. (n. 37, for additional references not used here; also Wilcken, Archiv. f. Pap. VI, 1920, 424). For θυσία as a sacrifice of incense in the Egyptian manner see Roussel, Cultes égyptiens, p. 286; cf. Roberts-Skeat-Nock, , H.T.R. XXIX, 1936, 49Google Scholar; Nilsson, , H.T.R. XXXVIII, 1945, 65Google Scholar.

²³ P. Oxy. I, 110; III, 523; XII, 1484; XIV, 1755; P. Oslo. III, 157 (with an important commentary).

²⁴ Milne, J.E.A. XI, 1925, 6–9, denies any religious character to the κλῖναι of the Oxyrhynchus invitations: they are dining clubs. This view is certainly extreme and rests largely on the confusion occasioned by imputing to the pagan mind our modern habit of restricting religion to “that which is not of this world.” A more reasonable view is espoused by Schubart, Einführung, 356, 367; Wilcken, Archiv. f. Pap. VI, 1920, 424; Eitrem and Amundsen, P. Oslo. III, 157, introd.

²⁵ P. Oxy. XII, 1484: ὑπὲρ μελλοκουρίων (see editors' note). The religious motivation comes out clearly in Archiv. f. Pap. II, 1903, 447, No. 76 (Commodus): Crispinus, a soldier, was commanded in a dream τὸ συνπόσιν ποιῆσαι [τοῦ κ]υρίου Σεράπιδος, and this he did εὐχαριστήσας.

²⁶ P. Oxy. III, 523: ἐν τοῖς Κλαυδ(ίου) Σαραπίω(νος), P. Oslo. III, 157: ἐν τῆ ἰδίᾳ οἰκίᾳ.

²⁷ P. Oxy. I, 110: ἐν τῷ Σαραπείῳ; XIV, 1755: ἐν τῷ οἴκῳ τοῦ Σαραπείου. The Gnomon of the Idiologus 88 (commentary in B.G.U. V, pt. 2, p. 93), mentions sacrifices [λί]νης in the consumption of which the παστοϕόροι, but not the αροϕῆται, might share (cf. Schubart, Einführung, 367). An Oxyrhynchus templeaccount (P. Oxy. VIII, 1144, 6; late 1st — early 2nd cent. A.D.) has an entry for νης ἱερᾶς κλε[νης], evidently a κλνίη sponsored by the temple. In P. Tebt. III, i, 765, 1–2, someone wished to send ε[ἰς] τό ἱερὸν κλείνην καὶ τύλην (cf. Bottigelli, Aegyptus XXII, 1942, 250).

²⁸ P. Oxy. XII, 1484; ἑν τῷ θ[ο]ηρίῳ.

²⁹ Roussel, Cultes égyptiens, pp. 100, 253, 285.

³⁰ Höfler, Sarapishymnus, 93ff.; P. Oslo. III, 157, introd.

³¹ Aristides 45.27 (ed. Keil); cf. Höfler, op. cit. 96f. For the dance as a religious practice see Oesterley, , The Sacred Dance, New York, 1923Google Scholar; Latte, R.G.V.V. XIII. pt. 3, 64ff.; Nilsson, Studi e materiali di storia delle religioni X, 1934, 13.

³² On the extensive background of the Greek libation see Tolles, Banquet-Libations of the Greeks, Bryn Mawr, 1943; for libations in the cult of Sarapis, Roussel, Cultes égyptiens, pp. 286f. Libations and prayer are stipulated for the monthly banquet of a society of Zeus Hypsistos (Roberts-Skeat-Nock, H.T.R. XXIX, 1936, 40, 1. 9; with commentary, 49). Ptolemagrius boasts of numerous libations poured to Phoebus Apollo at feasts given by him twice yearly to the whole people of Panopolis (Guéraud, Annales du service XXXIX, Cairo, 1939, 292f., Col. 4, 4f.; Milne, Cat. Gén. Caire, 1905, 50).

³³ Köchling, R.G.V.V. XIV, pt. 2, 1914, 45ff.; Breccia, Musée Égypt. III, 1915, 13–25; Perdrizet, , Terres cuites Fouquet, Paris, 1921Google Scholar, No. 336.

³⁴ A dining club of initiates may be attested by the inscription from Prusa (see footnote 15).

³⁵ Now Kom Aushim.

³⁶ A desert road some 35 miles long ran from Karanis via Bacchias and Sakkarah to the Sarapeum west of Memphis (P. Fay., p. 195; Petrie, A Season in Egypt, 33). The distance from Karanis to Arsinoë was about half that, and a Sarapeum was established there also (Wessely, , Sitz. Akad. Wiss. Wien CXLV, 1902, 41Google Scholar; Brady, , Missouri Studies X, 1935, No. 1, 45Google Scholar. Aristides 45.32 [ed. Keil] says that Egypt had 42 Sarapea; for the possible significance of the number see Höfler, Sarapishymnus, 110). Ptolemaeus' use of ἀνα- and κατα- in compounds is suitable for either place. ἀνα- marks the ascent from the valley to the desert plateau, κατα- the descent from the desert to the low-lying valley, e.g., in going between the city of Memphis and the Sarapeum (Wilcken, U.P.Z. I, 2, 22–23; 120, 1–2, 13–14; 122, 9; also pp. 174, n. 7–8, 320, n. 10). Within the nomes, ἀνα- indicates movement from village to city, κατα- from city to village, as between Karanis and Arsinoë (Wilcken, Chrest. 495, introd.; Westermann, J.E.A. X, 1924, 136). Because of another, and the most usual, meaning of these prefixes in Egypt, ἀνα- and κατα- for up and down the river, i.e., southward and northward, Alexandria is excluded. This usage would not exclude centers of Sarapis worship like Oxyrhynchus and Abydus, but the distances are too great. Ptolemaeus is concerned for his father's health (ll. 14–15, ἵνα μὴ σὺ κοπιᾷς), and one has the impression that he asks his father to bring the wood because he is not too far away. The text leaves it uncertain whether his father has only one donkey or more (ll. 10–11, τὸ ναῦλον τῶν ὄνων 18, τῷ ὄνῳ), but however it was, a single trip from Karanis to Oxyrhynchus with five donkeys or five trips with one donkey (ll. 11–12, χρεία γάρ ἐστιν ε γάμων) would be a heavy task even for a younger man. The management of five loaded donkeys over a distance of only 35 miles, as from Karanis to Memphis, might exceed the powers of one driver (Youtie, T.A.P.A. LXXIII, 1942, 83, n. 77). The journey would require at least one day's hard travelling and so would probably extend into a second day (Pearl, T.A.P.A. LXXI, 1940, 377, n. 14). If Ptolemaeus' father has only one donkey, five round trips would be required within two months (1. 17, δίμηος). With Ptolemaeus farther away than Memphis, the project becomes impractical.

³⁷ Greek αὐτῖς without antecedent, but it doubtless refers to the persons in charge of the κλίνη.

³⁸ τόπος was as unrestricted in its range of meaning as English “place.” Bell, , H.T.R. XXXVII, 1944, 196Google Scholar, notes its use for lodgings in P. Oxy. XVIII, 2190; for a colonnade connected with a dining hall in Sammelbuch III, 6823; for a religious community in P. Oxy. XII, 1492. Roberts-Skeat-Nock, H.T.R. XXIX, 1936, 45f., have discussed in some detail its use to designate a topographical division (city, town, village, hamlet) as well as a temple or meeting place within a temple. As used in connection with a κλίνη) at a Sarapeum, it might refer to the entire precinct or only to the banqueting hall (δειπνητήριον, P. Fay., p. 33; cf. P. Oslo. III, 157, p. 247, n. 1), but a closer interpretation is possible, τόπος may indicate an individual place at a banquet, as in Luke 14.9–10. That places were assigned is clear from P. Mich. V, 243 (Tiberius), where each participant has his own place (τοῦ ἰδίου τόπου) and a small fine is imposed on anyone who attempts to occupy the place of another. A similar rule was applied by the Iobacchi (Dittenberger, S.I.G. III, 1109, 74). In P. Mich. V, 246 (early 1st cent. A.D.) 15 members of a guild of Hippocrates are assigned places in order. The list is headed by the ἡγούμενος, who contributes sacrificial offerings, while the rest contribute sums of money ranging from 10 to 24 drachmae. Note that the amount reported by Ptolemaeus for τόπος, 22 drachmae, comes within this scale of prices.

³⁹ For a brief discussion of the -τικος suffix see Palmer, , Grammar of Post-Ptolemaic Papyri, London, 1946, 37Google Scholar. σιωπητικός can be derived from σιωπάω or σιωπή.

⁴⁰ σιωπηλός, less often σιωπηρός. Both forms survive in modern Greek.

⁴¹ It is naturally omitted from the 9th edition of Liddell and Scott, which does not pretend to offer guidance for the Byzantine period; but it is listed without references in the 8th edition, where the comment is restricted to “taciturn, Byz.” Sophocles' Greek Lexicon supplies a reference to Apophth. Patrum (= Patr. Gr. LXV) 341C: ἐάν ἦς σιωπητικός, ἔξεις ἀνάπααυσιν ἐν παντὶ τόπῳ οὖ ἐὰν οἰκήσης. Stephanus' Thesaurus adds Johannes Climax 77 (= Patr. Gr. LXXXVIII, 713D): μὴ γίνου ἄλογος σιωπητικός, ὲτέροις ταραχὴν καὶ πικρίαν προξενῶν; Ephraem. Syr. (ed. Assemani) I, 240 B: μὴ κληθῆς ἀνήκοος, σκληρὸς πὲρπερος ψίθυρος … ἀλλἀ μᾶλλον ὑπήκοος, ἀληθινός σώφρων, σιωπητικός κτλ; Athanas. (ed. Lopin-Montfaucon, 1698) II, pt. 2, 363B: θἑλε δὲ τοὐς παῖδας σιωπητικούς προβιβάζειν. The reference in Stephanus to Iamblichus, de vita Pythagorae, 163, for σιωπικούς is false: the true text is σιωπηλούς, as read by Kuster, p. 139, note 27, and adopted by Deubner, p. 92.

⁴² There appears to be no trace of σιωπητικός otherwise through the entire history of the Greek language. What happened is sufficiently clear. The strong tendency exhibited by post-classical Greek to form adjectives in -κος (Palmer, Grammar, 34–39) produced σιωπητικός, which came into use sporadically fromthe 3rd to the 6th century but was not able to make serious headway against σιωπηλός, itself not too frequently used. Preisigke, Wörterbuch d. gr. Papyri, does not list a single example of either word. The history of σιγηλός and σιγητικός presents a certain parallel, so far as may be judged from the limited evidence. Liddell-Scott-Jones, 9th ed., attests σιγητικός only from Hp. Decent. 3: πρός τὰςἀναστασιὰς σιγητικός; this work was composed in the middle of the 4th cent. B.C. (Pauly-Wissowa, R.-E. VIII, art. Hippokrates [Gossen], 1813.) The type-example of words that rode the wave of the koine but failed to survive is πρός τὰςἀναστασιὰς σιγητικός, which was “practically the only form in use throughout the Greek-speaking world during iii/B.C. and the first half of ii/B.C.” (Thackeray, Grammar of the O.T. in Greek, Cambridge, 1909, 58–62).

⁴³ The Byzantine silentiarius (= ἡανχοποιός, Loewe, Corp. Gloss. Lat. II, 325), whose duty it was to maintain order and silence at the meetings of the imperial consistory (Dunlap, Mich. Hum. Stud. XIV, 1924, 220; Pauly-Wissowa, R.-E., 2te Reihe, III, art. Silentiarius [Seeck], 57f.) comes to mind in any consideration of a word that could pass as a translation of his title. Adjectives in -τικος may be transitive or intransitive (Debrunner, Wortbildungslehre, 199f.; Palmer, Grammar, 37. The transitive force is probably original in the root; cf. Glotta XII, 1923, 28f.; XIII, 1924, 135. Similar but not identical is the use of σιγηλός for Narcissus; his sad fate inspired a fearful silence in those who passed his tomb [Strabo 9.2.10, ἐπειδὴ σιγῶσι παριόντες Alciphron 3. 58 (22). 3, τρέμε ἐνδακὼν τὸ χεῖλος, ὡς οἱτὸν Σίγηλον ἤρω παριόντες]). His office, however, was not instituted before Diocletian, so that the name given to him cannot have influenced Greek usage in the early part of the third century. Cumont, , A.J.A. XXXVII, 1933, 262Google Scholar, likewise rejects an equivalence of the Bacchic σιγηταί with the private silentiarii, “les esclaves chargés de clore la bouche à leurs compagnons,” an older and unofficial use of the word, on the ground that the 23 σιγηταί listed in the Bacchic inscription in the Metropolitan Museum are too many monitors for one society. At the same time, one is tempted to use the underlying idea, “he who imposes silence,” as an explanation of σιωπητικός, because banquets sponsored by individuals, cult societies, trade guilds, and social clubs in Egypt had moderators, regularly the president (ἐπιμελητήρ, ἡγεμών, ἡγούμενος, κεϕαλαιωτής, κλινάρχης, προστάτης) or his assistant (βοθός, ὑπηρέτης). See Boak, , T.A.P.A. LXVIII, 1937, 213fGoogle Scholar.; Roberts-Skeat-Nock, , H.T.R. XXIX, 1936, 50Google Scholar, 83; Nicolò, San, Ägypt. Vereinswesen, Munich, 1915, II, 77fGoogle Scholar; Westermann, J.E.A. XVIII, 1932, 23. At two annual festivals of Naucratis (Hermias ap. Athen. 4.149D-F) the banquet was supervised by a ίεροποιός, “manager of the festival” (Gulick), evidently here a municipal magistrate (Pauly-Wissowa, R.-E. VIII, art. Hieropoioi [Oehler], 1583ff.). The same title was used by an insignificant servants' club at Philadelphia (Fayyum) in the later Ptolemaic period: συνήχθησαν … διὰ ὶεροποιῦ (Edgar, Raccolta Lumbroso, 369–376), and by a club at Tebtunis, whose members make contributions for ίεροποΐα (P. Tebt. III, ii, 894, introd.; ca. 114 B.C.). See also footnote 55. At semi-annual banquets given in honor of Apollo at Panopolis by Ptolemagrius, the owner of a flourishing grove, to the people of the town, the moderators (ἄρχοντες) were priests (Guéraud, Annales du service XXXIX, 1939, 293, Col. TV, 3f.). The ritual silence imposed by the Iobacchi could be broken only with priestly permission (Dittenberger, S.I.G. Ill, 1109, 107–110): μηδείς δ’ ἔπος ϕωνείτω μὴ ἐπιτρέψαντος τοῦ ίερέως ἣ τοῦἀνθιερέως Cf. San Nicolò, op. cit. 57f. None of these officials, however, held so highly specialized a title as would the moderator of the banquet in our papyrus. It reminds us rather of the rôle played by a lesser personage, the εὔκοσμος, in the society of the Iobacchi at Athens, whose duty it was to place the thyrsus beside members guilty of unseemly conduct and thus mark them for expulsion from the meeting place (Dittenberger, III, 1109, 136–139, with note 49). The ῥαβδοϕόροι performed a like function at Andania (ibid. II, 736, x). Philo, adv. Flaccum 17, probably does not exaggerate the tumult of the usual social banquet at its height, although he is a prejudiced observer. Roberts-Skeat-Nock, op. cit. 53, point to a confirmatory passage in Athenaeus 10.420E-F. Athenaeus is approximately contemporary with the writer of our papyrus, and a little later Tertullian, Apol. 39.14–19, holds a very low opinion of the moral quality of pagan banquets, among which he includes those of Sarapis. Nothing was so necessary as a strong man to keep peace among the diners. And yet a minor official was not likely to give his name to one of two contributions, and the larger one at that, demanded of those who wished to attend a banquet. In rejecting σιωπητικός= “he who imposes silence” we may take account also of Cumont's remark on σιγηταί “Faut-il comprendre ceux qui imposent ou ceux qui observent le silence? Grammaticalement les deux sens sont admissibles, bien que le second soit le plus probable” (A.J.A. XXXVII, 1933, 262Google Scholar); but see also Pauly-Wissowa, R.-E. XVI, pt. 2, art. Mysterien (Hopfner), 1310.

⁴⁴ Vogliano-Cumont-Alexander, , A.J.A. XXXVII, 1933, 215–270Google Scholar; cf. Nilsson, Studi e materiali di storia delle religioni X, 1934, 1–18.

⁴⁵ Ibid., 262f Cumont notes the parallel with new disciples of Pythagoras, who were pledged to silence for five years (ibid., 263 and n. 6); Plutarch, Moral. 519C: ἒταξε τοῖς νέοις πενταετῆ σιωπήν, έχεμνθίαν προσαγορεύσας; Suidas, s.v. σιωπή; Casel, R.G.V.V. XVI, pt. 2, 54f On sacred silence see Mensching, R.G.V.V. XX, pt. 2, 86 (Greek), 102 (Roman); Arist. Jud. 95 (Jewish); Schmidt, R.G.V.V. IV, pt. 1, 66ff.; with special reference to Rome, Savage, T.A.P.A. LXXVI, 1945, 157–165. The mystic silence is illustrated with a number of Greek passages by Cumont, op. cit. 262, nn. 3–4. On the rôle of silence in late paganism see Nilsson, Rev. of Religion, Jan. 1947, 119; in Hermetic thought, Willoughby, Pagan Regeneration, 215.

⁴⁶ There is no record of σιωπητής.

⁴⁷ In omitting the preposition ὑπέρ with σιωπητικοῦ and τόπον the writer has been influenced by the syntax characteristic of lists and accounts (Völker, Artikel, Münster, 1903, 8f.).

⁴⁸ Philo, adv. Flaccum 17, equates σύνοδοι and κλῖναι with θίασοι Cf. Poland, Griech. Vereinswesen, 152.

⁴⁹ Anrich, Antikes Mysterienwesen, 9f. (esp. 9, n. 3).

⁵⁰ P. Oxy. XIII, 1612, introd.; Deubner, Sitz. Heid. Akad., 1919, Abh. 17, 10. Cf. Foucart, Mystères d'Éleusis, 263f. Otto, Priester u. Tempel, 265, n. 1, rejects the assumption that the Eleusinian mysteries were introduced into Egypt, but does not deny that the cult of Demeter was popular there. Nilsson (Griech. Feste, 350; H.T.R. XXXVI, 1943, 270) takes the common view that the Eleusinian mysteries came to Alexandria with Timotheus, seemingly a misinterpretation of Tacitus, Hist. 4.83.

⁵¹ For Eleusinian influence on the Dionysiac cult, see Cumont, A.J.A. XXXVII, 1933, 232f., 239, 241, 243, 253; on Samothrace, Nock, A.J.A. XLV, 1941, 577.

⁵² See footnotes 13, 15.

⁵³ Foucart, Mystères d'Éleusis, 272; Cumont, A.J.A. XXXVII, 1933, 258, 263; id., Monuments relatifs au Culte de Mithra, Bruxelles, 1899, I, 314f.

⁵⁴ A probationary period for the mystes of Isis is clearly implied in Metamorph. 11.19: aedibusque conductis intra consaeptum templi Larem temporarium mihi constituo deae ministeriis adhuc privatis appositus contuberniisque sacerdotum individuus et numinis magni cultor inseparabilis. Cf. Reitzenstein's illuminating discussion of Lucius' status during this period (Hellenist. Mysterienrel., 25ft.), although his attempt to see a catochus in Lucius cannot be sustained (Wilcken, U.P.Z. I, p. 75).

⁵⁵ Polycrates writes to his father Cleon, the royal engineer of the Fayyum, in the middle of the 3rd cent. B.C.: γίνωσκέ με τὴν ἱεροποίαν ὠικονομημέ[νον] (Mahaffy, P. Petrie II, 11 (2) = Wilcken, Chrest. 223). On the hieropoios see footnote 43. Polycrates' appointment was also probably for a single festival and he may have received some equivalent of his services, as did Ptolemaeus (Mahaffy, ibid.; Witkowski, Epist. priv., 2).

⁵⁶ The value of the double portions is much increased by the fact that his assumption of the agoranomia exempts him from payment. The technical term is ἀσύμβολος, and the exemption is ὰτέλεια (Dittenberger, S.I.G. III, 1109, 6). Normally this privilege was conferred as an honor in view of past services. A society of Tyrian Heracles (2nd cent. B.C.) decreed that one of its members should be free of dues in perpetuity (Foucart, Assoc. Relig., 39f., 224 [44], cf. 227 [45, 114]). In P. Petrie III, 136, i, 11–12, no σνμβολή is recorded for the ἰερεύς and the δεύτερος ἱερεύς (cf. Wilcken, U.P.Z. I, p. 439, n. 1). These ἀσύμβολοι honoris causa are to be distinguished from the visitors at village clubs who were ἀσύμβολοι or ἀϕέσιμοι because their hosts paid for their dinners (cf. Edgar, Raccolta Lumbroso, 372; San Nicolò, Vereinswesen, II, 167, n. 2.).

⁵⁷ Hermias ap. Athen. 4.149D: τούτων γὰρ ἑκατέρῳ διπλοῦς ὁ οῖνος μετὰ καὶ τῶν ἄλλων μερίδων δίδοται On μερίς and μέρος see Poland, Vereinswesen, 258. Nock, H.T.R. XXXVII, 1944, 141, refers to Plutarch, Prov. Alex. (ed. Crusius, Progr. Tub., 1887), pp. 11f. (text; cf. preface, xvii), for an amusing account of how the custom of portions originated. For a divergent explanation preserved in Athenaeus see Crusius' commentary, 53.

⁵⁸ διμοιρία: Dittenberger, O.G.I.S. 78 (221–205 B.C.).

⁵⁹ Hula-Szanto, Sitz. Akad. Wiss. Wien CXXXII, 1895, 23.

⁶⁰ Foucart, Mystères d'Éleusis, 222.

⁶¹ Ferguson, H.T.R. XXXVII, 1944, 74f. and n. 17.

⁶² This would be true whether the banquet was to take place at the great Sarapeum west of Memphis, which lay on the edge of the desert and was devoid of flora as it was of water (Wilcken, U.P.Z. I, pp. 17, 382), or at Arsinoë, the capital of the nome, where wood might be bought but certainly not cheaply.

⁶³ The Greek is ambiguous. From whom and for what will the father receive the δαπάνη The reimbursement must come from Ptolemaeus, for when he writes χορηγῶ αὐτοῖς ξύλα, this can only mean that he will supply wood to the temple — αὐτοῖς has no expressed antecedent — without compensation. Whether Ptolemaeus promises to pay his father for the wood or only for the transportation, does not emerge readily from the text but is of less importance.δαπάνη

⁶⁴ Schnebel, , Landwirtschaft im hellenist. Ägypten (= Münch. Beiträge VII, 1925), 292Google Scholar.

⁶⁵ Poland, Vereinswesen, 258 (n. 2), 393, 466; cf. Dittenberger, S.I.G., II, 736, xxii.

⁶⁶ P. Cair. Zen. II, 59154 (= Select Papyri I, 90): κορμοὺς ὄτι πλείστους καὶ παχυτάτους; 59191: σχίζας ὄτι πλ[ε]ίστας.

⁶⁷ I.G. XII (5), 606. Cf. P. Oxy. VIII, 1144, 15: ξύλων εἰς θνσίαν.

⁶⁸ We have no reason here to concern ourselves with the notarial agoranomi, functionaries established in the metropolis of each nome for the registration of legal documents from the 3rd cent. B.C. (Gerhard, Philologus LXIII, 503; Mitteis, Grundzüge, 58). They are without parallel in earlier Greek history and are not to be confused with the municipal agoranomi (see footnote 70).

⁶⁹ For a lively, imaginative picture of the bustle and confusion of the agora, see Glotz, Ancient Greece at Work, New York, 1926, 289f.: “the Agoranomoi do not know to whom to listen.” For the agora as a center of activity at the time of a great fair or festival (panegyris), see Nilsson, Greek Popular Religion, 100.

⁷⁰ Häderli, Jahrb. f. cl. Ph., Supplb. 15, 1887, 47–94; Jouguet, Vie munidpale, 327f.; Pauly-Wissowa, R.-E. I, art. Agoranomoi (Oehler), 884; Oertel, Liturgie, 332f. The two kinds of agoranomi are distinguished by Jouguet as “agoranomenotaire” and “agoranome-édile.” When Septimius Severus revised the organization of the Egyptian nomes in 200 A.D. (Milne, Egypt under Roman Rule, London, 1924, 141ft., 289ff.), the two agoranomic types — recorder and aedile — were probably brought under a single jurisdiction both in the nomes and in the Greek cities.

⁷¹ Dittenberger, S.I.G. II, 736, xx–xxii: άγορᾶς … ὔδατος … άλείμματος καὶ λουτροῦ.

⁷² Dittenberger-Purgold, Inscriften von Olympia, 436, 437.

⁷³ Dittenberger, S.I.G. II, 596 (ca. 200 B.C.): κατασταθ]εὶς ὐπὸ τοῦ δήμο[υ τοῦ IIαριανῶν] ἀγορανόμος εἰς τὰ [νέα IIαναθήν] αια.

⁷⁴ The appointment of a private agoranomus by a temple is unlikely since the title is unknown outside of municipal government. The Thessalian agoranomus, while he corresponds to neither of the types referred to above, is nevertheless a concity official. Keil, Hermes XXXIV, 1899, 196f.Google Scholar, has shown him to be the counterpart of the Athenian ἐπιστάτης.

⁷⁵ Oertel, Liturgie, 334. The municipal agoranomia was by the time of our papyrus a compulsory liturgy in Egypt (ibid., 332). There is also evidence pointing to a shorter term of 4 months. See Pauly-Wissowa, R.-E. I, art. Agoranomoi (Oehler), 883f.; A, 6, 2, art. Tralleis (Ruge), 2110.

⁷⁶ A festival would be the most opportune occasion for gathering people together at a large kline. Cf. Nilsson, Popular Greek Religion, 101: “The cult of the gods provided opportunities for assembling and feasting and for mutual intercourse between people from neighboring towns and even from all Greek countries.”

⁷⁷ West, Class. Phil. XI, 305. In the middle of the 3rd cent. A.D., and so practically contemporary with Ptolemaeus, Heroninus was manager of an estate of Appian at Theadelphia and drew a monthly salary of 40 dr. (P. Lond. III, 1226, pp. 103f.; cf. Johnson, Roman Egypt, Baltimore, 1936, 303). It is instructive to compare the high wages of musicians and actors working on contract for festivals. Their earning capacity was anywhere from 5 to 18 dr. a day, plus food and transportation (Westermann, J.E.A. X, 1924, 141).

⁷⁸ On the flow of pilgrims to the Memphite Sarapeum see Wilcken, U.P.Z. I, pp. 50f.; Höfler, Sarapishymnus, 72; to the Sarapeum at Abydus, Perdrizet-Lefebvre, Graffites grecs d'Abydos, Paris, 1919, xv. Typical are Harmais, who came once a year from his home in the Heracleopolite nome to offer a sacrifice at Memphis (U.P.Z. I, 122; 157 B.C.), and Hephaestion, who came in 168 B.C. together with his comrades, on his way home at the conclusion of the war with Antiochus IV, to show his gratitude to Sarapis for coming through safely (ibid. pp. 36, 51, Nos. 59, 60).

⁷⁹ Merchants were of course plentiful at Arsinoë (cf. Wessely, Sitz. Akad. Wiss. Wien CXLV, 1902, 43ff.), and the population of the great Sarapeum near Memphis was sufficient to support a number of resident merchants. Demotic contracts of 116–5 B.C. mention “Kaufleuten des Hauses des Osorapis (d.h. des Osiris-Apis-Tempels), die innerhalb des Tempelbezirks Häuser besassen” (Wilcken, U.P.Z. I, pp. 17. 52).

⁸⁰ The Sarapeum and the city of Memphis were naturally prepared for these occasions. Apart from others, there were lodgings in the Anubium at the Sarapeum and next to the Aphrodisium in Memphis (U.P.Z. I, 120, 122; cf. p. 51).

⁸¹ This situation is illuminated by the Ptolemaic accounts from the Sarapeum, in which the catochus Ptolemaeus is seen providing wood for cult purposes and receiving water supplied to him by the temple (Wilcken, U.P.Z. I, pp. 17, 382). In connection with our immediate interest compare especially Sammelbuch V, 7617, 55: [ὕ]δωρ ταῖς ἡορταῖς.

⁸² This is an inference drawn from his acceptance of the duties of the agoranomus, which would certainly involve him in considerable preparation before the festival begins, and from 11. 12–16 of the Greek text, which carry the implication that he would leave his post only to fulfill the demands of Sarapis.

⁸³ Although the life of the catochi at Memphis, so far as it is known from the sources, and the various theories concerning them have been carefully described and analyzed by Wilcken (U.P.Z. I, pp. 52–77), we remain in ignorance concerning their function in the religion of Sarapis because we do not know with certainty the religious mechanism by which they became catochi (ibid., p. 66) or how they differed essentially from others who lived at the Sarapeum for a longer or shorter time. We know only that unlike the priests and the pilgrims they did not leave the precinct of the temple (ibid., p. 65). Wilcken's final word: non liquet.

⁸⁴ Ibid., p. 75.

⁸⁵ Cf. Reitzenstein, Hellenist. Mysterienrel., 66ff.

⁸⁶ U.P.Z. I, pp. 62–65.

⁸⁷ Ibid., 70–72. The literary sources show a late trend toward ecstatic manifestations and asceticism; these are obviously foreign to Ptolemaeus.

⁸⁸ Apuleius, Metamorph. xi. igff.

⁸⁹ Cumont, AJ.A. XXXVII, 1933, 232f.

⁹⁰ Conceivably only minor changes might be required to turn an Egyptian into a Greek festival. Herodotus (2.48) saw a close resemblance between the Egyptian festival of Osiris and the Greek festival of Dionysus (Westermann, , J.E.A. XVIII, 1932, 18Google Scholar). The κλίνη like the practice of incubation and the ἐγκατοχή, is almost certainly Greek (Wilcken, U.P.Z. I, p. 95; Archiv. f. Pap. VI, 1920, 424).

⁹¹ Wilcken, U.P.Z. I, p. 95, appears to distinguish the Egyptian cult at Memphis from the Greek cult at Alexandria, but his own list of concessions made at Memphis to the spirit of Hellenization that was initiated at Alexandria is impressive. It is useful in this connection to insist on the formal distinction between Hellenic and Hellenistic, between old Greek and later cosmopolitan Greek custom. After Alexander, the real situation is obscured by excluding native elements from the definition of res Graecae. Hellenistic civilization was precisely a more or less unstable fusion of things Greek and non-Greek. For an excellent characterization of Hellenism see Schubart, Religiöse Haltung d. früh. Hellenismus (= Der Alte Orient XXXV, pt. 2, 1937), 3–5.

The view that the Egyptians obliterated the worship of Sarapis in the 2nd cent. B.C. by “identifying him with Osiris and Apis and thus destroying his Greek nature” (Brady, Missouri Studies X, 1935, No. 1, 29) is a purely statistical reading of the evidence. Gods as well as men submit to varying fortunes. Sarapis was a new god and put in a long apprenticeship at the side of Isis, but everyone, whether Greek or Egyptian, knew from the first that he was the Interpretatio Graeca of Osiris-Apis (Wilcken, U.P.Z. I, p. 88). His name was not an invention; it was the form that Osor-hapi had taken in Greek. Ptolemy and his advisers were not cloaking a deception with a barbarous name (ibid., p. 87). By the second century it was at least a hundred years too late for calling attention to what had never been a secret.

⁹² Brady, Missouri Studies X, 1935, No. 1, 43.

⁹³ Jouguet, Trois études, 55: “Dès le III^e siècle on jurait par ces dieux, Isis et Sarapis, dans le serment dit ‘royal’.” Cf. Seidl, Der Eid, Diss. Munich, 1929, 12ff. Scholars have commonly sought to explain Ptolemy's fateful creation of the Greek Sarapis as an attempt to bridge the cultural and religious gulf between the Greek conquerors and the subject Egyptians (cf. Wilcken, U.P.Z. I, pp. 83–85), i.e., as a policy of amalgamation, but this would have suited Alexander (Tarn, Proc. Brit. Acad. XIX, 1933, 123ff.Google Scholar; for limits of Alexander's ideas on racial equality, Jouguet, op. cit., 41.) far better than his successors, who turned away from equalitarian ideas with the ingrained sense of Greek superiority (Schubart, Einführung, 339). There had come, however, to be another side to Greek dignity: the Hellene was not such by virtue of blood, but every man who spoke and thought with a Greek mind was a Hellene (Jüthner, Hellenen u. Barbaren, 44.ff; Jouguet, op. cit., 39). This view, which was one, and not the least important, of the consequences of Alexander's life, left the door wide open for the ambitious or enlightened “barbarian,” and it was a view peculiarly suited to the political needs of the new Greek kingdoms established on foreign soil. With the establishment of Sarapis as the dynastic deity of Egypt (Gressmann, Oriental. Rel., 35), the Egyptian was offered, on equal terms with the Greek, a god properly his own but in Hellenistic dress. Those who accepted him would to that extent have achieved, whether consciously or not, an adaptation to the dominant Hellenism of the country and the time. The separation from native custom would be carried further by membership in a cult society, which was a usual complement to divine worship according to the Greek pattern. Westermann has written with thoughtful insight that “the organization of the private clubs, which appear in such numbers in Egypt, as elsewhere, after Alexander's time, whether connected with some cult or purely social and dissociated from religious practice, represents an element in the Mediterranean life of the Hellenistic period which has its source and inspiration in the imitation of Greek life” (J.E.A. XVIII, 1932, 27Google Scholar).

Nock, , H.T.R. XXIX, 1936, 78ff.Google Scholar, has found the Egyptian elements in cult societies so impressive that he describes the guild of Zeus Hypsistos at Philadelphia (?) as “an Egyptian form of organization.” Rostovtzeff, Soc. and Econ. Hist, of the Hellenistic World, III, 1590, gives it as his opinion “that in the late Ptolemaic period the two types influenced each other, while in pre-Ptolemaic and early Ptolemaic times they may have existed side by side.”

For a recent treatment, along the lines laid down by Tarn, of Alexander's cosmopolitan response to the problems of imperial government, see Robinson, C. A., Alexander the Great, New York, 1947Google Scholar.