Article

1. The parabiblical narrative, written ca. 600, shows the author’s awareness of Hellenised oriental cults still surviving at Harran and Baalbek, of Zoroastrian fire-worship, and of astrology.
2. [2:42 PM]6th century is seen as the final twilight of Greco-Roman religion. With the emperor Justinian prohibiting the performance of pagan rites even in private, the religious tradition of ancient Greece and Rome practically had come to an end (K. W. Harl, “Sacrifice and Pagan Belief in Fifth- and Sixth-century Byzantium,” PastPres 128 (1990) 7–27, here 22–26). Whereas public displays of pagan beliefs had been banned from the early fifth century onward, Justinian’s measures “made it against the law even to be a pagan” (Alan Cameron, “Paganism in Sixth-century Byzantium,” in Wandering Poets and Other Essays on Late Greek Literature and Philosophy [Oxford 2016] 255–286, here 255). On the anti-pagan legislation of Theodosius II in the first half of the fifth century (see J. Hahn, “Gesetze als Waffe? Die kaiserliche Religionspolitik und die Zerstörung der Tempel,” in Spätantiker Staat und religiöser Konflikt (Berlin 2011) 201–220). The closing of the Neoplatonic Academy at Athens in 529 and subsequent flight of its last scholarch Damascius to Persia effectively destroyed if not the theoretical basis of paganism then at least an important part of its educational infrastructure. On Alexandrian Neoplatonism’s adaptation to Christianity and its increasingly antiquarian interest in the Hellenic past (see C. Wildberg, “Three Neoplatonic Introductions to Philosophy: Ammonius, David and Elias,” Hermathena 149 (1990) 33–51, esp. 33–34, 38, 44–45; H. Tarrant, “Olympiodorus and the Surrender of Paganism,” ByzF 24 (1997) 181–192, esp. 182, 185, 191–192).

1. At the same time the demolition of temples or their rededication as churches had become endemic in all parts of the empire. In spite of this, the Syriac Book of the Cave of Treasures (henceforth CT ), a Christian apocryphal work of the late sixth or early seventh century written on Persian ground, bears witness to an apparently still vibrant practice of Hellenised oriental cults alongside the indigenous Zoroastrian religion in the border area between Byzantium and the Sasanid empire. As CT has so far not been used as a source-text in the reconstruction of late antique paganism, it seems prima facie worthwhile to collect and analyse whatever can be gleaned from the text in terms of contemporary pagan practices.
2. The Cave of Treasures, which in some manuscripts is attributed to the Syrian theologian Ephrem of Nisibis (fourth century), in all likelihood was written towards the beginning of the seventh century during the final years of the Sasanian monarchy but before the rise of Islam. This can be inferred with reasonable certainty from the use of Sasanian royal names for the magi attending Jesus’ birth (CT 45.19)—among them “Parwezdad,” which seems to refer to Khosrau II Aparvez (590–628)—and from the fact that CT is first quoted in the Revelations of Ps.- Methodius in the mid-seventh century and shows no trace of knowledge of the Muslim conquest. In combination this makes a rather strong case for the work being written sometime during or shortly after the reign of Khosrau II, most likely on Persian soil, by an author belonging to the (“Nestorian”) Church of the East. The book was written in Syriac, the language used by Persian Christians for ecclesiastical and scholarly purposes, and has been translated into Arabic, Ethiopic, Georgian and possibly Coptic, as well (“The Cave of Treasures. A New Translation and Introduction,” in R. Bauckham et al. (eds.), Old Testament Pseudepigrapha. More Noncanonical Scriptures I (Grand Rapids 2013) 531–584, here 532–535). On the CT’s Syriac background (cf. S. Minov, “The Cave of Treasures and the Formation of Syriac Christian Identity in Late Antique Mesopotamia,” in B. Bitton-Ashkelony et al. (eds.), Between Personal and Institutional Religion (Turnhout 2013) 155–194). The Syriac text is preserved in two recensions which reflect the confessional split of Syrian Christians into a western (“Jacobite” monophysite) and an eastern (“Nestorian” dyophysite) branch. The differences between the recensions are mostly doctrinal in nature.

As to its content, CT essentially presents a re-reading of Old Testament history with a focus on pre-diluvian times and an emphasis on genealogical tables leading in an uninterrupted path from Adam to Jesus. Its author makes use of a variety of sources of mostly Christian provenance which are strung together in the well-known scheme of septimana mundi, i.e an attempt to structure history according to a framework of seven times 1000 years (ee W. Witakowski, “The Idea of Septimana Mundi and the Millenarian Typology of the Creation Week in Syriac Tradition,” in R. Lavenant (ed.), V Symposium Syriacum 1988 (Rome 1990) 93–109). The coherent use of chronology and the fact that so far no independent source-texts outside of CT have been identified strongly suggest that the text in its entirety goes back to the seventh-century author instead of being an assemblage of earlier unrelated works (C. Leonhard, “Observations on the Date of the Syriac Cave of Treasures,” in P. M. M. Daviau et al. (eds.), The World of the Aramaeans III: Studies in Language and Literature in Honour of Paul-Eugène Dion (Sheffield 2001) 255–293). Whatever its sources may have been, the book had a long afterlife in both Christian and Muslim authors, mostly historians and polymaths such as Ps.-Dionysius of Tell-Mahre (late eighth century), al-Jaqubi (late ninth century), Tabari and Eutychius of Alexandria (both ninth/tenth century), Michael the Syrian (late twelfth century), Barhebraeus and Solomon of Basra (both thirteenth century) (see A. Goetze, “Die Nachwirkung der Schatzhöhle,” ZS 2 (1923) 51–94 and 3 (1924) 53–71, 153– 177). Its author, however, was witness not only to the Zoroastrian religion of Sasanid Persia but also to Greco-Roman religious practices in the eastern Mediterranean and northern Mesopotamia.

1. Euhemeristic explanation of pagan cults
2. References to pagan cults are sparse in the Cave of Treasures. After all, the text, being part of the rewritten-bible genre, first and foremost is concerned with retelling biblical history ab initio mundi until Pentecost, while giving the account some characteristic twists, the background and motives of which cannot always be fully understood. As is common among Christian authors of late antiquity, the origin of pagan cults is explained by CT in a euhemeristic way. In the course of its relecture of biblical history, the ‘invention’ of image-worship is narrated in the context of the prehistory of Abraham (25.13–14). The latter passage is interesting in several respects. Whereas the euhemeristic theme is common in several earlier Christian authors (e.g. Ps.-Clem. Rom. Hom. 5.23; Clem. Alex. Protr. 2.24.2; Min. Fel. 21.1–5; Lact. Div.Inst. 1.1; Arnob. 4.29), CT here connects the origin of image-worship with the idea that images are inhabited by oracle-giving demons, thereby tying traditional pagan religion intimately to “magic, incantations, divination, Chaldaean arts, fortune-telling, augury, and omens.”

Both themes—the idea that statues are vehicles of divine presence and the link between traditional religion and theurgy— are of course well known from Neoplatonism and at least the idea that statues were inhibited by demons seems to have been current in Byzantium as late as the ninth century (C. Mango, “Antique Statuary and the Byzantine Beholder,” DOP 17 (1963) 53–75, here 56–59; H. Saradi-Mendelovici, “Christian Attitudes toward Pagan Monuments in Late Antiquity and their Legacy in Late Byzantine Centuries,” DOP 44 (1990) 47–61, here 56–58; I. Jacobs, “Production to Destruction? Pagan and Mythological Statuary in Asia Minor,” AJA 114 (2010) 267–303, here 267–273, 279–282 (crosses carved on the foreheads of statues as a form of Christianization or exorcism), 285–286 (burial of statues as a means “to neutralize their powers”), 291–292). For an account of the belief in animated statues from archaic through imperial times see (J. N. Bremmer, “The Agency of Greek and Roman Statues,” Opuscula 6 (2013) 7–21, here 7– 14). Among Christian authors already Eusebius Dem.Evang. 6.20.11 regards cult statues as being inhabited by demons. The author of CT sees these demons as evil spirits; “conversely, in the eyes of fourth-century Neoplatonists, idols were animated with divine presence.”T hus, while for Plotinus (Enn. 4.3.11) statues simply mediate the presence of the gods, they can predict the future according to Proclus (In Ti., III 155.18–22 Diehl). The theme of human sacrifice which is introduced by CT in this context is attributed to Julian in the Historia Ecclesiastica of Theodoret (3.26.2–3). Writing in the first half of the fifth century, the church historian recounts a visit of Julian to the northern Mesopotamian city of Harran during his campaign against the Persians. While the visit and a sacrifice in the temple of the moon god Sin are historical (Amm. Marc. 23.3.2), Theodoret has him sacrifice a woman for oracular purposes. Even though this story is most likely a fabrication of Julian’s Christian adversaries, there seem to have been human sacrifices down to the sixth century in connection with the Arab goddess al-῾Uzza (T. Green, “The Presence of the Goddess in Harran,” in E. N. Lane (ed.), Cybele, Attis, and Related Cults (Leiden 1996) 87–100, here 97; Th. Hainthaler, “Die Töchter Allahs in vorislamischer Zeit mit besonderer Berücksichtigung von al-῾Uzzā,” in G. Augustin et al. (eds.), Christentum im Dialog (Freiburg i.Br. 2014) 297–310, here 306–308). As punishment for the rampant idolatry in Abraham’s time, the author of CT then introduces a “deluge of wind,” which is based upon a similar account of Josephus (AJ 1.118) (S.-M. Ri, Commentaire de la Caverne des Trésors (Louvain 2000) 316). Whereas Josephus has the tower of Babel destroyed by this cataclysmic event, CT employs it to wipe out idol worship. The account thus closes with a reference to archaeology (26.11, 17–18).

1. Harran
2. That the statues of old were not just buried beneath the tells of Syria and Mesopotamia, but that paganism was still very much alive at the author’s own time, is shown by two later accounts of Harran and Baalbek-Heliopolis, both strongholds of traditional worship towards the end of the sixth century (Harl, PastPres 128 (1990) 14; E. Watts, “Where to Live the Philosophical Life in the Sixth Century? Damascius, Simplicius, and the Return from Persia,” GRBS 45 (2005) 285–315, here 302–305). Especially Harran, located at the crossing of two trade routes connecting Antioch with Nineveh and Melitene with Babylon respectively, gained notoriety well into Islamic times for being home to the Hermetic religion of the Sabians (See generally W. Cramer, “Harran,” RAC 13 (1986) 634–650; M. Tardieu, “Ṣābiens coraniques et ‘Ṣābiens’ de Ḥarrān,” JA 274 (1986) 1–44; J. van Ess, Theologie und Gesellschaft im 2. und 3. Jahrhundert Hidschra (Berlin 1992) 442–460; U. Possekel, “The Transformation of Harran from a Pagan Cult Center to a Christian Pilgrimage Site,” POr 36 (2011) 345–356). Christian authors from the fourth century on were highly suspicious of it, and the acts of the Council of Chalcedon in fact mention Harran as Ἑλλήνων πόλις (ACO II.3 25.3). In the reign of the emperor Maurice (582–602) there was a persecution which led to the martyrdom of the prefect Acindynus for failing to convince the authorities of the sincerity of his conversion to Christianity (Cf. Cramer, RAC 13 (1986) 646–647; D. Pingree, “The Ṣābians of Ḥarrān and the Classical Tradition,” IJCT 9 (2002) 8–35).

In the briefest possible manner the author here gives an overview of the Harranian cult and mythology as it presumably was at the end of the sixth century. To begin with, under its euhemeristic guise emerge the ancient moon god Sin and his wife Nikkal/ Ningal, here called Harranit, a compositum of the nomen locis with the Semitic feminine ending -it. Worship of her husband Sin is documented at least from the fourteenth century BCE and seems to have ceased only with the final destruction of his temple by fire in 1035 CE (Cf. Cramer, RAC 13 (1986) 636, 642; T. Green, The City of the Moon God (Leiden 1992) 62–64). As mentioned above, Julian visited the temple in 363 to obtain an oracle. Sin’s consort Ningal, on the other hand, “is the most clearly formulated female aspect of the moon” (Green, in Cybele 89), and indeed there is a certain amount of confusion in Greek and Latin authors as to whether Harran’s main deity should be seen as male or female. Thus Herodian (4.13) and Ammianus Marcellinus (23.3.2) refer to Selene and Luna, respectively, while the HA (Caracalla 7.3) mentions Lunus as a secret name alongside the commonly-used feminine version Luna.

1. CT’s description of the cult of Harran, sparse and truncated as it is, thus seems to hint at a pattern of the dying god as it is well known from the Adonis cult as well as the mythologies of Isis and Osiris and the Eleusinian mysteries, all of which were popular well into late antiquity.
2. Baalbek-Heliopolis
3. The Syrian city Baalbek/Heliopolis, which was one of the last remaining centres of paganism alongside Harran in the outgoing sixth century, is likewise mentioned in the Cave of Treasures. Its foundation is attributed to Solomon (35.18–21). Such a cult was not only implied by the city’s Hellenistic name “Heliopolis,” but is also attested in Macrobius’ description of Baalbek (Sat. 1.23.10–21). There Baalbek’s main deity Jupiter Heliopolitanus is expressly identified with the sun god whose statue is made of the solar metal, gold. While this does not correspond to the god’s true historical pedigree ( Jupiter Heliopolitanus seems to have been derived from the Syro-Palestinian storm god Baal-Hadad), it is quite in line with the general tendency of late antiquity to reduce male divinities to the sun god (Cf. W. Liebeschuetz, “The Significance of the Speech of Praetextatus,” in P. Athanassiadi et al. (eds.), Pagan Monotheism in Late Antiquity (Oxford 1999) 185–205, here 186–194). Although not mentioned by the Syriac author, the actual cult at Heliopolis was dedicated to a triad, Jupiter, Venus, and Mercury (Hajjar, ANRW II18.4 (1990) 2462–2465).
4. Zoroastrianism
5. It is Nimrod, however, who in the Cave of Treasures is not only presented as king and founder of cities but also as the father of Zoroastrianism and an aspirant to astronomical and astrological knowledge. Identified with Zoroaster by Christian authors early on (W. Bousset, Hauptprobleme der Gnosis (Göttingen 1907) 147; A. M. Schilling, Die Anbetung der Magier und die Taufe der Sasaniden (Louvain 2008) 150– 152), he is introduced in CT as the “first king upon earth” and ruler of Babylon (24.24–26). Again, Nimrod-Zoroaster is firmly embedded in biblical history by the reference to “Reu’s days” in Gen 11:20 as well as by the epithet “hero” which reflects the Hebrew gibbor “warrior” of Gen 10:8. At the same time the author of CT here establishes a first link between Nimrod and Zoroastrianism by introducing the “weaver Sisan,” whose name sounds conspicuously similar to “Sasan,” name-giver of the Sasanid dynasty of late antique Iran. The heavenly crown which Sisan—perhaps not unintentionally in the role of servant—prepares for Nimrod makes an appearance in the Syriac Romance of Julian. There it is treated by Julian’s opponent Shapur as a token of the Persian kings’ oracular powers, which have been bestowed upon Nimrod on the occasion of receiving “the crown from heaven” (S. Minov, Syriac Christian Identity in Late Sasanian Mesopotamia (diss. Hebrew Univ. Jerusalem 2013) 267–268, with a fuller exploration of the motif’s Iranian background at 268–273. On the anti-pagan background of the Romance of Julian see J. W. Drijvers, “Julian the Apostate and the City of Rome: Pagan-Christian Polemics in the Syriac Julian Romance,” in W. J. van Bekkum et al. (eds.), Syriac Polemics (Louvain 2007) 1–20; D. L. Schwartz, “Religious Violence and Eschatology in the Syriac Julian Romance,” JECS 19 (2011) 565–587). One reason why it might have been important for a Christian author of the late sixth/early seventh century to detach NimrodZoroaster from actual Persian religious practice can be seen in the fact that various sources have Zoroaster predict the birth of Christ (Bidez and Cumont, Les mages I 46–49). Because of their identification of Zoroaster with Bileam, several patristic authors (Origen, Jerome, Ambrose, Diodorus of Tarsus) trace the magi’s knowledge to the Moabite prophet. The early Christian evidence is treated extensively in D. Hannah, “The Star of the Magi and the Prophecy of Balaam in Earliest Christianity, with special attention to the lost Books of Balaam,” in P. Barthel et al. (eds.), The Star of Bethlehem and the Magi (Leiden 2015) 433–462. On Bileam in early Jewish literature see J. Zsengellér, “Changes in the Balaam Interpretation in the Hellenistic Jewish Literature,” in H. Lichtenberger et al. (eds.), Biblical Figures in Deuterocanonical and Cognate Literature (Berlin 2008) 487–506. In CT such a tradition is markedly present in the description (45–46) of the magi’s visit to Bethlehem.47 According to this there obviously was a tradition of astronomical observation in the “land of Nimrod,” i.e. Persia (45.7).