BMCR 2003.07.09

The Religious Life of Palmyra

, The religious life of Palmyra : a study of the social patterns of worship in the Roman period. Oriens et occidens ; Bd. 4. Stuttgart: Franz Steiner Verlag, 2002. 305 pages, 7 unnumbered pages of plates : illustrations, maps ; 25 cm.. ISBN 3515080279 EUR 64.00/sFr 102.40.

This book is a slightly revised version of the author’s doctoral thesis, written at Oxford University under Fergus Millar, and is a study of the social patterns of worship in Palmyra in the Roman period. The work pays special attention to how earlier elements of religious life and society interplayed with innovative elements as Palmyra gradually took on the outward forms of a Graeco-Roman city. Kaizer’s work does not provide definitive answers or simple generalizations. Still, this carefully nuanced discussion of Palmyrene religion does attempt to modify some of the previous understandings of the deities, rituals, sanctuaries, and civic and religious groups and institutions at Palmyra and will thus be significant for historians of religion and the Roman Near East, as well as Semitists and scholars of Palmyra. Kaizer’s work takes its place beside another recent study of Palmyrene religion in the Roman empire, The Palmyrenes of Dura-Europos: a study of religious interaction in Roman Syria, by Lucinda Dirven (Leiden: Brill, 1999).

In the introduction (pp. 13-34), Kaizer gives a brief presentation of the site of Palmyra in the Syrian desert and reviews the current state of sources, the history of the study of Palmyrene religion, various problems of approach, and the phenomenon of bilingualism at Palmyra. Kaizer announces his goal of modifying the usual description that religion at Palmyra was almost completely Semitic, having an indigenous substratum that was influenced by Babylonian and Canaanite elements in pre-Hellenistic times, followed by the infiltration of Arab and Syrian deities by the Roman period. He instead posits a dynamic model of “continuous renegotiation of old and new elements” in Palmyrene religious life (p. 27); original traditions and newly introduced elements coexisted and continually influenced each other, and new traditions quickly became as native as the original. As for the problematic nature of bilingualism at Palmyra, in that over 200 of the 3000 inscriptions from Palmyra are in both Greek and the local Aramaic (most of the rest are solely in Aramaic, while any Latin disappears early in the third century CE), Kaizer reviews the discussion of which language is the source language and which is the receptor, but does not himself come down on one side or the other.1

In ch. 1 of the book, “Palmyrene society and Palmyrene religion” (pp. 35-66), Kaizer undertakes to explore the sociopolitical structure of the city and its influence on the religious structure. In contrast to the common view that the city had “tribal” versus “civic” forms of worship (sanctuaries associated with certain tribes, with the Temple of Bel as the central temple for the community at large), Kaizer thinks that the cultural spheres of the city are best demonstrated by the two strata of its religious world. That is, the group of deities surrounding Bel such as Yarhibol and Aglibol “points to an Aramaean layer which had undergone Mesopotamian influences” (p. 56), and the deities centered around Baal-Shamin, Allat, and the sun-god Shamash, are a later addition to Palmyra from differing regions, but not necessarily Arab or nomadic. (Baal-Shamin’s origins are clearly Phoenician, while the sun-god was male at Palmyra as in Mesopotamia but female among the ancient Arabs. Allat’s origins may go back to the Mesopotamian goddess Allatu.)2 That the deities of the Baal-Shamin group are later additions to the society is illustrated by the fact that the temples to these deities are on what used to be the outskirts of the city in the north and west.

In this section Kaizer also carefully reviews the difficulties behind the idea of “tribal religion” at Palmyra. Not only is the very word “tribe” often misunderstood as indicating a static organization based solely on kinship connections, but the second century CE inscriptions referring to “the four tribes of the city each with their own sanctuary” are problematic as well. There were at least fourteen tribes at Palmyra all told, and the five designated as φυλή can not all be clearly connected to a particular temple. Kaizer concludes that, although the assembly ( δῆμος) and council ( βουλή) were in place by 74 CE, and the designation of “four tribes” was typical of the Roman division of cities into four quarters, the city’s organization was not completely that of a standard Graeco-Roman city. Several tribes pre-dated Roman rule, and the use of the term “the four tribes” in the Roman period was an artificial construct to designate the city as a whole rather than in its parts.

In ch. 2, “Sanctuaries and cults” (pp. 67-161), Kaizer reviews the religious topography of Palmyra. He first turns to sanctuaries with identified archaeological remains such as the temples of Bel, Baal-Shamin, Nebu, Allat, Bel-Hamon, and Arsu, and then to those sanctuaries or sacred sites mentioned in inscriptions for which there is little or no clear archaeological or iconographic evidence, such as the temple of Rabaseirè, the temple of Atargatis, the “Sacred Garden” of Aglibol and Malakbel, the Efqa spring near the city with its probable connection to Yarhibol, the Caesareum of the imperial cult, and the alleged “cult of the sun” as mentioned in the Historia Augusta. New temple architecture was incorporated with the old as the city became Graeco-Roman; often an older cella would be retained as the temple grew, or a new structure preserved an old form. The relationships between different deities can be hard to define as individual deities often share the same attributes in depictions and can show up in different constellations in various reliefs, and inscriptions — members of the Bel group will appear with those from the Baal-Shamin group and vice versa. Several tribes are connected to sanctuaries in the inscriptions (e.g., the Bene Mattabol with Arsu/Ares, the Bene Komare with the “Sacred Garden,” etc.), but Kaizer concludes that an act of worship could “be simultaneously ‘individual,’ ‘familial,’ ‘tribal,’ ‘societal’ and ‘civic'” (p. 161).

The enormous temple of Bel was “the centre of the religious life of Palmyra” (p. 67) and had a communal function; it is often called “the house of their gods” in inscriptions. Kaizer is sympathetic to the idea that there was a so-called “triad of Bel” which included the supreme god Bel along with two attendants Yarhibol and Aglibol, but other deities are mentioned or depicted alongside the three, especially Astarte. The building had several phases beginning in the early first century CE (although the actual cult of Bel was older); all were funded by local Palmyrenes, and its installations were dedicated only to Palmyrene gods.3 Society and religion at Palmyra took on aspects of Graeco-Roman patterns, but, as Kaizer notes later in the book, “direct Roman influence on Palmyrene religion was almost completely absent, but for the presence of the imperial cult” (p. 261).

Baal-Shamin was paradoxically also a supreme deity like Bel, and, like Bel, is rendered as Zeus in the Greek inscriptions. The temple of Bel-Hamon was on top of Jebel Muntar to the west but close to the Efqa spring and was served by members of the Bene Agrud who also seemed to have been in the service of Bel; Kaizer concedes that Bel-Hamon might be a manifestation of Bel. The “Sacred Garden” of Aglibol and Malakbel is listed as one of the four sanctuaries of the city in some inscriptions, but, if there was a real physical building or temple at the garden, it has not been found in the archaeological record yet. Kaizer does not believe that Aglibol and Malakbel should be thought of as forming a triad with Baal-Shamin or the so-called “anonymous God,” as has been previously suggested. Concerning evidence of the god Yarhibol and his relation to the Efqa spring, there are hardly any traces of a sanctuary at the source, and there are no dedications to Yarhibol there. On the other hand, inscriptions elsewhere in the city and in other places such as Dura-Europos connect Yarhibol to the spring (as does the meaning of his name, “Lord of the Source”). At the Efqa spring there are instead numerous altars dedicated to an unnamed god, “he whose name is blessed forever.” This so-called “anonymous God,” who shares the main epithets of Baal-Shamin, does not have a sanctuary of his own, and Kaizer agrees with others that his cult was a development toward more personal religion.

In ch. 3, “The rhythm of religious life” (pp. 163-211), Kaizer’s focus is on the various rituals performed at Palmyra. He states that it is now impossible to know which activities were part of any indigenous, original core, or which were developments in the Roman period. There are only six inscriptions regarding cultic regulations at Palmyra, and, as there are no myths, prayers or liturgies extant, any information about religious life must come from the dedicatory and other kinds of inscriptions, as well as the iconography. As for the sacrificial acts at Palmyra, the most commonly depicted in the iconography is the burning of incense in a censer, while the second is the offering of libations from bowls or jars. Other rituals at Palmyra are less frequently depicted and included offerings of animals, birds and even fruit. In the temple of Bel there was a lustration basin, but little more is known about conditions of purity or ritual bathing.

With regard to sacrificial terminology at Palmyra, most words are not derived from Greek, but from Semitic languages, especially Akkadian. Kaizer says it is hard to tell what this means for the rituals themselves. The beds or couches dedicated to deities at Palmyra in inscriptions indicate divine dining paraphernalia and not a “sacred marriage” ritual. There is some evidence for processions in the city, but probably not, as previously understood, on a route down the Central Colonnade between the temples of Bel and Allat. Kaizer discusses the matter of festivals, especially the possibility of a Mesopotamian Akitu spring festival and a Kanunu winter festival. The evidence for an Akitu festival is that the temple of Bel at Palmyra was dedicated on the sixth of April (April, or Nisan, is the month in which the Akitu festival was celebrated in Mesopotamia, at which time temples were sometimes dedicated), and there is a relief in that temple probably depicting the slaughter of Tiamat. This is the goddess known from the Babylonian Enuma Elish, the story of creation, which was probably recited on the fourth day of Nisan. On the other hand, the main god in the relief is Nebu, and not Bel, but this may be a Palmyrene peculiarity. Finally, Kaizer notes that the overwhelming number of reliefs of the gods in contrast to statues at Palmyra seems to indicate there was no “dressing” of the gods as at Dura-Europos. He supposes the reliefs might serve the purpose of prompting what Drijvers calls “cultic acts of reflection.” Kaizer does not believe rituals were used to enforce or transmit group identities, but instead were a means of bringing the community together.

Ch. 4, “Groups of worshippers, priests and benefactors” (213-259), reviews the various groups of worshippers and how they functioned in the stratification of society. It is here that Kaizer discusses more fully the kinship terminology, such as the bny X, or “sons of” X, indicating the various tribal groups who dedicated items or performed sacred acts in the sanctuaries. These designations are on the over one thousand tesserae, or tokens of admission for the sacred banquets, found mostly in the temple of Bel.

The most important groups of worshippers at Palmyra were those involved in drinking and banqueting confraternities, called mrzH (vocalized marzeaH) in Aramaic, or συμπόσιον in Greek, which can also refer to the “banqueting room” itself. Palmyrene )drwn) is often used too, perhaps to be understood as an Aramaic transliteration of Greek ἀνδρών, meaning “dining hall,” but perhaps etymologically from the Sumerian word for “heavenly abode” anduruna, although Kaizer notes that the two options are not mutually exclusive. Nonetheless, the Sumerian option, first suggested by Stephanie Dalley,4 may be more problematic. It is true that the Sumerian sequence an dur2-ru-na seems lexicalized in some instances, as in the Akkadian Enuma Elish (I.34). However, it does not normally mean “heavenly abode” (a usage clearly attested only in one incantation) but rather “the place in which the god An/Heaven dwells” ( ki an dur2-ru-na).5 Four banquet halls have been found: one in the temple of Baal-shamin, two outside the temple complexes, and the largest one in the temenos of the temple of Bel, where most of the tesserae were found. The tesserae and reliefs speak of the marzeah’s of different deities, and one notes that individual Palmyrenes seemed to be involved in more than one cult. Kaizer rightly notes that any funerary aspect of the marzeah is lacking at Palmyra, although there may be a funerary aspect to older versions of it, as at second millennium BCE Ugarit (although the cautious view there now predominates6).

The presidents of the confraternities and the priests seem to have been important citizens who also held civic leadership. Unlike elsewhere in the empire and in contrast to the establishment of Roman civil offices at Palmyra, nothing in the priesthood seems to have been based on Roman models. Kaizer also discusses funerary foundations or testamentary gifts, that is, the practice of leaving money to provision the cult. Here there was definite interplay between city and cult, in that the city could pay for dedications to temples and the council and assembly could set up a statue for a benefactor who had performed civic acts as well as donated to a temple.

After a very brief summary of the book’s main points (pp. 261-264), the book concludes with a bibliography (pp. 265-293) and indexes of inscriptions, place-names, divine names, personal names, and general topics. Unfortunately, the index of personal names is highly selective, and the tribal names are not included in any index — this is a pity since the purpose of the study is to relate the city’s social groupings and institutions with its religious ones. There is some irregularity in the rendering of Aramaic personal names in the book in general: mkymw as Moqimu but also Moqimo and Moqeimu; (gylw as Ogeilu usually, but also Ogeilo; yrHy as Yarhai but also Iarhai; and Hdwdn as Haddoudan but also Hadudan. There are seven plates of maps and reliefs, and one would wish for more of them to illustrate the various depictions and groupings of deities.

Kaizer is to be congratulated for emphasizing the complexity of this ancient local Syrian religion and society, with influences from not only its Near Eastern milieu, but also from the Greco-Roman world as the city became a Roman colony. On the other hand, this complexity is often unwieldy to present; for instance, the concept of an “indigenous” layer is hard for Kaizer or anyone to avoid, but, as Kaizer admits, one can not really isolate an original religious or cultural substratum at Palmyra. Kaizer’s work succeeds in thoroughly questioning easy generalizations about religious life in this Syrian city.


1. Concerning the bi/trilingualism of Latin and Aramaic by Palmyrenes, see J. N. Adams’ Bilingualism and the Latin Language (Cambridge: Cambridge University Press, 2003), pp. 247-271.

2. On the problems inherent in the alleged first mention of Arabian Allat found in Herodotus and on the etymology of Allat in general, see now Jaakko Hämeen-Anttila and Robert Rollinger, “Herodot und die arabische Göttin ‘Alilat,'” Journal of Ancient Near Eastern Religions 1 (2001):84-99.

3. On Bel, Belti, Nabu, Nanaya, in first and second century CE Mesopotamia, see Christa Müller-Kessler and Karlheinz Kessler, “Spätbabylonische Gottheiten in spätantiken mandäischen Texten,” Zeitschrift für Assyriologie 89 (1999): 65-87.

4. Stephanie Dalley, “Bel at Palmyra and elsewhere in the Parthian period,” Aram 7 (1995): 137-151 (esp. 140).

5. Wayne Horowitz, Mesopotamian cosmic geography (Winona Lake, Ind.: Eisenbrauns, 1998), pp. 109, 225.

6. Dennis Pardee, “MarziHu, kispu, and the Ugaritic funerary cult: A minimalist view,” in Ugarit, religion and culture, ed. N. Wyatt et al. ( Fs. J.C.L. Gibson. Münster: Ugarit-Verlag, 1996), pp. 273-287.