Dying and Rising Gods

views updated


DYING AND RISING GODS . The category of dying and rising gods, once a major topic of scholarly investigation, must now be understood to have been largely a misnomer based on imaginative reconstructions and exceedingly late or highly ambiguous texts.


As applied in the scholarly literature, "dying and rising gods" is a generic appellation for a group of male deities found in agrarian Mediterranean societies who serve as the focus of myths and rituals that allegedly narrate and annually represent their death and resurrection.

Beyond this sufficient criterion, dying and rising deities were often held by scholars to have a number of cultic associations, sometimes thought to form a "pattern." They were young male figures of fertility; the drama of their lives was often associated with mother or virgin goddesses; in some areas, they were related to the institution of sacred kingship, often expressed through rituals of sacred marriage; there were dramatic reenactments of their life, death, and putative resurrection, often accompanied by a ritual identification of either the society or given individuals with their fate.

The category of dying and rising gods, as well as the pattern of its mythic and ritual associations, received its earliest full formulation in the influential work of James G. Frazer The Golden Bough, especially in its two central volumes, The Dying God and Adonis, Attis, Osiris. Frazer offered two interpretations, one euhemerist, the other naturist. In the former, which focused on the figure of the dying god, it was held that a (sacred) king would be slain when his fertility waned. This practice, it was suggested, would be later mythologized, giving rise to a dying god. The naturist explanation, which covered the full cycle of dying and rising, held the deities to be personifications of the seasonal cycle of vegetation. The two interpretations were linked by the notion that death followed upon a loss of fertility, with a period of sterility being followed by one of rejuvenation, either in the transfer of the kingship to a successor or by the rebirth or resurrection of the deity.

There are empirical problems with the euhemerist theory. The evidence for sacral regicide is limited and ambiguous; where it appears to occur, there are no instances of a dying god figure. The naturist explanation is flawed at the level of theory. Modern scholarship has largely rejected, for good reasons, an interpretation of deities as projections of natural phenomena.

Nevertheless, the figure of the dying and rising deity has continued to be employed, largely as a preoccupation of biblical scholarship, among those working on ancient Near Eastern sacred kingship in relation to the Hebrew Bible and among those concerned with the Hellenistic mystery cults in relation to the New Testament.

Broader Categories

Despite the shock this fact may deal to modern Western religious sensibilities, it is a commonplace within the history of religions that immortality is not a prime characteristic of divinity: Gods die. Nor is the concomitant of omnipresence a widespread requisite: Gods disappear. The putative category of dying and rising deities thus takes its place within the larger category of dying gods and the even larger category of disappearing deities. Some of these divine figures simply disappear; some disappear only to return again in the near or distant future; some disappear and reappear with monotonous frequency. All the deities that have been identified as belonging to the class of dying and rising deities can be subsumed under the two larger classes of disappearing deities or dying deities. In the first case, the deities return but have not died; in the second case, the gods die but do not return. There is no unambiguous instance in the history of religions of a dying and rising deity.

The Deities

The list of specific deities to whom the appellation "dying and rising" has been attached varies. In most cases, the decipherment and interpretation of texts in the language native to the deity's cult has led to questions as to the applicability of the category. The majority of evidence for Near Eastern dying and rising deities occurs in Greek and Latin texts of late antiquity, usually post-Christian in date.


Despite the original Semitic provenance of Adonis, there is no native mythology. What is known depends on later Greek, Roman, and Christian interpretations.

There are two major forms of the Adonis myth, only brought together in late mythographical tradition (e.g., the second-century ce Bibliotheca, falsely attributed to Apollodorus of Athens). The first, which may be termed the Panyasisian form, knows only of a quarrel between two goddesses (Aphrodite and Persephone) for the affections of the infant Adonis. Zeus or Calliope decrees that Adonis should spend part of the year in the upperworld with the one, and part of the year in the lowerworld with the other. This tradition of bilocation (similar to that connected with Persephone and, perhaps, Dumuzi) has no suggestion of death and rebirth. The second, more familiar Ovidian form narrates Adonis's death by a boar and his commemoration by Aphrodite in a flower. There is no suggestion of Adonis rising. The first version lacks an account of Adonis's death; the second emphasizes the goddess's mourning and the fragility of the flower that perpetuates his memory. Even when the two versions are combined, Adonis' alternation between the upper and lower worlds precedes his death.

The rituals of Adonis, held during the summer months, are everywhere described as periods of intense mourning. Only late texts, largely influenced by or written by Christians, claim that there is a subsequent day of celebration for Adonis having been raised from the dead. The earliest of these is alleged to be the second-century account of Lucian (Syrian Goddess 67) that, on the third day of the ritual, a statue of Adonis is "brought out into the light" and "addressed as if alive"; but this is an ambiguous report. Lucian goes on to say that some think the ritual is not for Adonis but rather for some Egyptian deity. The practice of addressing a statue "as if alive" is no proof of belief in resurrection; rather it is the common presupposition of any cultic activity in the Mediterranean world that uses images. Besides, Lucian reports that after the "address" women cut their hair as a sign of mourning.

Considerably later, the Christian writers Origen and Jerome, commenting on Ezekiel 8:14, and Cyril of Alexandria and Procopius of Gaza, commenting on Isaiah 18:1, clearly report joyous festivities on the third day to celebrate Adonis (identified with Tammuz) having been "raised from the dead." Whether this represents an interpretatio Christiana or whether late third- and fourth-century forms of the Adonis cult themselves developed a dying and rising mythology (possibility in imitation of the Christian myth) cannot be determined. This pattern will recur for many of the figures considered: an indigenous mythology and ritual focusing on the deity's death and rituals of lamentation, followed by a later Christian report adding the element nowhere found in the earlier native sources, that the god was resurrected.

The frequently cited "gardens of Adonis" (the kepoi) were proverbial illustrations of the brief, transitory nature of life and contain no hint of rebirth. The point is that the young plant shoots rapidly wither and die, not that the seeds have been "reborn" when they sprout.

Finally, despite scholarly fantasies, there is no evidence for the existence of any mysteries of Adonis whereby the member was identified with Adonis or his fate.

Aliyan Baal

The Ras Shamra texts (late Bronze Age) narrate the descent into the underworld of the puissant deity Aliyan Baal ("the one who prevails; the lord") and his apparent return. Unfortunately, the order of the incidents in the several different texts that have been held to form a Baal cycle is uncertain. The texts that are of greatest relevance to the question of whether Aliyan Baal is correctly to be classified as a dying and rising deity have major lacunae at the most crucial points. Although these texts have been reconstructed by some scholars using the dying and rising pattern, whether these texts are an independent witness to that pattern remains an open question.

In the major narrative cycle, Baal, having won the rulership by vanquishing the dangerous waters, is challenged by Mot, ruler of the underworld, to descend into his realm. After some initial hesitation, and after copulating with a cow, Baal accepts the challenge and goes down to the lower realm, whence it will be said of him that he is as if dead. After a gap of some forty lines, Baal is reported to have died. Anat descends and recovers his corpse, which is properly buried; a successor to Baal is then appointed, and Anat seeks out and kills Mot. After the narrative is interrupted by another forty-line gap, El declares, on the basis of a symbolic dream, that Baal still lives. After another gap of similar length, Baal is described as being in combat with a group of deities. As is apparent from this brief summary, much depends on the order of incidents. As it stands, the text appears to be one of a descent to the underworld and returna pattern not necessarily equivalent to dying and rising. Baal is "as if he is dead"; he then appears to be alive.

In another, even more fragmentary Hadad cycle (Hadad being identified with Baal), Hadad goes off to capture a group of monsters, but they, in turn, pursue him. In order to escape he hides in a bog, where he lies sick for seven years while the earth is parched and without growth. Hadad's brothers eventually find him and he is rescued. This is a disappearing-reappearing narrative. There is no suggestion of death and resurrection.

There is no evidence that any of the events narrated in these distressingly fragmentary texts were ritually reenacted. Nor is there any suggestion of an annual cycle of death and rebirth. The question whether Aliyan Baal is a dying and rising deity must remain sub judice.


The complex mythology of Attis is largely irrelevant to the question of dying and rising deities. In the old, Phrygian version, Attis is killed by being castrated, either by himself or by another; in the old Lydian version, he is killed by a boar. In neither case is there any question of his returning to life. There is a second series of later traditions that deny that Attis died of his wounds but do not narrate his subsequent death or, for that matter, his rebirth. Finally, two late, post-Christian theological reflections on the myth hint at rebirth: the complex allegory in the Naassene Sermon and the euhemerist account in Firmacus Maternus, in which a pretended resurrection is mentioned. Attis is not, in his mythology, a dying and rising deity; indeed, he is not a deity at all.

All of the attempts in the scholarly literature to identify Attis as a dying and rising deity depend not on the mythology but rather on the ritual, in particular a questionable interpretation of the five-day festival of Cybele on March 2227. The question of the relationship between the Day of Blood (March 24) and the Day of Joy (March 25) caught the attention of some scholars, who, employing the analogy of the relationship of Good Friday to Easter Sunday, reasoned that if among other activities on the Day of Blood there was mourning for Attis, then the object of the "joy" on the following day must be Attis's resurrection. Unfortunately, there is no evidence that this was the case. The Day of Joy is a late addition to what was once a three-day ritual in which the Day of Blood was followed by a purificatory ritual and the return of the statue of the goddess to the temple. Within the cult, the new feast of the Day of Joy celebrates Cybele. The sole text that connects the Day of Joy with Attis is a fifth-century biography of Isidore the Dialectician by the Neoplatonic philosopher Damascius, who reports that Isidore once had a dream in which he was Attis and the Day of Joy was celebrated in his honor!

Scholars have frequently cited a text in Firmacus Maternus (22.3) as referring to Attis and his resurrection on the Day of Joy: "Be of good cheer, you of the mysteries, your god is saved!" However, the god is unidentified, and the notion of "cheer" is insufficient to link this utterance to Attis and the Day of Joy. The text most probably reflects a late antique Osirian ritual.

Neither myth nor ritual offers any warrant for classifying Attis as a dying and rising deity.


The figure of the king-god of Babylon, Marduk, has been crucial to those scholars associated with the Myth and Ritual school as applied to the religions of the ancient Near East. For here, as in no other figure, the central elements of their proposed pattern appear to be brought together: the correlation of myth and ritual, the annual celebration of the dying and rising of a deity, paralleled by an annual ritual death and rebirth of the king. Marduk is the canonical instance of the Myth and Ritual pattern.

In 1921, F. Thureau-Dangin published the text, transcription, and translation of a Seleucid era text, preserved in two copies, presenting a part of the ritual for the New Year festival (the Akitu) in Babylon. Despite a large number of references to the performance of the ritual in Babylonian texts (although not always to the Akitu associated with Marduk or Babylon) and scattered mentions of individual items in the ritual, this exceedingly late cuneiform text is the only detailed description of the ritual program in Babylon to survive. It enjoins twenty-six ritual actions for the first five days of the twelve-day ceremony, including a double reading of a text entitled Enuma elish. Assuming that this reference is to some form of the text now known by that name, the "Babylonian creation epic" as reconstructed by contemporary scholarship, the ritual suggests a close link to the myth. However, not one of the twenty-six ritual actions bears the slightest resemblance to any narrative element in the myth. Whatever the significance of the recitation of the text during the Akitu festival, the myth is not reenacted in that portion of the ceremonies that has survived.

Realizing this, some proponents of the Myth and Ritual approach have argued that the first five days of the ritual were only purificatory in nature, and go on to speculate that the next three days of the festival featured a dramatic reenactment of a myth of the death and resurrection of Marduk. This sort of imaginative speculation gave rise to a new set of problems. There is no hint of Marduk's death in the triumphant account of his cosmic kingship in Enuma elish. If some such myth was enacted, it was not the one stipulated in the ritual program. Nevertheless, scholars turned to a cuneiform text that they entitled The Death and Resurrection of Bel-Marduk. The title is somewhat misleading. There are sixteen episodes in the text, which appears to narrate Marduk's imprisonment. The text is fragmentary and difficult to interpret, but it appears to be in the form of a ritual commentary in which a set of ritual gestures are correlated to events in a subtextual narrative of Marduk's capture.

For an older generation of scholars, Marduk's imprisonment was equivalent to his death, and his presumed ultimate release represented his resurrection. More recent interpretations have minimized the cosmic symbolism: Marduk has been arrested and is being held for trial. By either reading, such a narrative of the king-god's weakness or crime would appear odd in a Babylonian setting. This caution is strengthened by the fact that the text is of Assyrian provenance and is written in the Assyrian dialect. It is not a native Babylonian text and could have played no role in the central festival of Babylon.

The so-called Death and Resurrection of Bel-Marduk is most likely an Assyrian political parody of some now unrecoverable Babylonian ritual composed after the Assyrians conquered Babylon in 691/689 bce. At that time, the statue of Marduk was carried off into Assyrian captivity. From one point of view, the text has a simple, propagandistic message: Compared to the gods of Ashur, Marduk is a weak deity. More subtly, for those Assyrians who held Marduk in some reverence, the notion of his crimes would provide religious justification for his capture.

The notion that the king undergoes an annual ritual of mimetic dying and rising is predicated on the fact that the deity, whose chief representative is the king, is believed to undergo a similar fate. If it is doubtful that Marduk was understood as a dying and rising deity, it is also doubtful that such a ritual was required of the king. Some scholars have held that the so-called ritual humiliation of the king on the fifth day of the New Year festival, with its startling portrayal of the king being dethroned, slapped, pulled by the ears, and reenthroned, is symbolic of his death and resurrection. But such an interpretation ignores both the manifest content of the ritual text and its date. During the humiliation ceremony, the king is required to recite a negative confession: that he did not overthrow his capital city of Babylon or tear down its walls, that he did not insult its protected citizens, that he did not neglect or destroy its central temple.

From one point of view, such a negative confession is ludicrous. What native Babylonian king ever contemplated, much less carried out, such actions? These were the actions of foreign kings (Assyrian, Persian, Seleucid) who gained the throne of Babylon by conquest and desecrated the native cult. However, as with Cyrus among the Israelites, so too for the Babylonians, foreign kings could be named who restored Babylon and its temple. Read in this light, the ritual humiliation of the king appears to be a piece of Babylonian nationalistic ritual rectification: Good fortune and continued kingship comes to the (foreign) king if he acts as a pious (native) king would act. If not, he will be stripped of his kingship.

This understanding is made more plausible by the date of the only surviving texts of the ritual. They are all from the Hellenistic Seleucid period, that is to say, from a period after the ending of native kingship and the installing of foreign kings on the throne. The pattern may be earlier, dating back, perhaps, to the time of Sargon II (r. 721705 bce), the earliest conqueror of Babylon to adopt consciously the Babylonian etiquette of kingship and during whose rule, for the first time, one finds legal texts guaranteeing Assyrian recognition of the rights and privileges of the "protected citizens" of Babylon. In the present text of the New Year ritual, a set of actions designed to deal with the more proximate Assyrian conquerors has been reapplied to the relatively more foreign Seleucid rulers.

There is no evidence that the Babylonian Marduk was ever understood to be a dying and rising deity, that such a myth was reenacted during the New Year festival, or that the king was believed to undergo a similar fate.


In contrast to the other deities considered above, Osiris has a thick textual dossier stretching over millennia. Although the full, connected myth is only to be found in Greek, in Plutarch's Isis and Osiris from the early second century ce, the Osirian myth can be reconstructed from the Pyramid Texts of the fifth and sixth dynasties. While the names of the actors and details of the incidents vary, this record is remarkably consistent over twenty-five hundred years. Osiris was murdered and his body dismembered and scattered. The pieces of his body were recovered and rejoined, and the god was rejuvenated. However, he did not return to his former mode of existence but rather journeyed to the underworld, where he became the powerful lord of the dead. In no sense can Osiris be said to have "risen" in the sense required by the dying and rising pattern; most certainly it was never conceived as an annual event. The repeated formula "Rise up, you have not died," whether applied to Osiris or a citizen of Egypt, signaled a new, permanent life in the realm of the dead.

Osiris was considered to be the mythical prototype for the distinctive Egyptian process of mummification. Iconographically, Osiris is always depicted in mummified form. The descriptions of the recovery and rejoining of the pieces of his body are all elaborate parallels to funerary rituals: the vigil over his corpse, the hymns of lamentation, the embalmment (usually performed by Anubis), the washing and purification of the corpse, the undertaking of the elaborate ritual of the "opening of the mouth" with its 107 separate operations, as well as other procedures for reanimation, the dressing of the body, and the pouring out of libations. Through these parallels, the individual Egyptian dead became identified with, and addressed as, Osiris (perhaps earliest in Pyramid Texts 167a168a). The myth and ritual of Osiris emphasizes the message that there is life for the dead, although it is of a different character than that of the living. What is to be feared is "dying a second time in the realm of the dead" (Book of Going Forth by Day 175176).

Osiris is a powerful god of the potent dead. In no sense can the dramatic myth of his death and reanimation be harmonized to the pattern of dying and rising gods.


The assessment of the figure of Tammuz (Sumerian, Dumuzi) as a dying and rising deity in the scholarly literature has varied more than any other deity placed in this class. For example, within a thirty-year period, one of the most significant scholars in the field, the Sumeriologist Samuel Noah Kramer, has revised his judgment regarding this question several times. Before 1950, Kramer thought it possible that Dumuzi was freed from death; between 1950 and 1965, he considered Dumuzi to be solely a dying god; since 1966, he has been willing to speak again of the "death and resurrection" of Dumuzi.

The ritual evidence is unambiguously negative. During the summer month of Tammuz, there was a period of wailing and lamentation for the dead deity. A substantial number of cultic hymns of mourning, going back to the second millennium bce, have been recovered; by the sixth century bce, the ritual was practiced in Jerusalem (Ez. 8:14); in Syria, it is witnessed to as late as the fifth century ce and, in variations, persisted through medieval times. If third-century Christian authors are to be trusted, the figure of Tammuz interacted with that of Adonis in Asia Minor. In all of these varied reports, the character of the ritual is the same. It is a relentlessly funereal cult. The young Tammuz is dead, and he is mourned. His life was like that of the shoot of a tender plant. It grows quickly and then withers away. It was a life that is "no more"a persistent refrain in the lamentations. There is no evidence for any cultic celebration of a rebirth of Tammuz apart from late Christian texts where he is identified with Adonis.

Given the predilection of scholars concerned with Christian origins for a pre-Christian pattern of dying and rising deities, it comes as no surprise that, despite the lack of cultic evidence, it was widely supposed that the period of mourning for Tammuz must have been followed by a festival of rejoicing. This speculative conclusion seemed to gain support with the publication of the Akkadian Descent of Ishtar from the library of Ashurbanipal in Nineveh. The text narrates the descent of the goddess into the underworld and her return. However, the concluding nine lines of the text contain a series of enigmatic references to Tammuz, Ishtar's youthful lover, in the land of the dead. Although the text nowhere mentions it, scholars supposed that the purpose of Ishtar's descent was to bring Tammuz up. If so, this would place Tammuz securely within the dying and rising pattern.

Even on the basis of the Akkadian text alone, such an interpretation is unlikely. There is no connection stated in the text between Ishtar's descent and Tammuz. (Indeed, some scholars have suggested that the last lines referring to Tammuz were originally independent and added to the Descent as a scribal gloss.) Even more detrimental to the dying and rising hypothesis, the actions performed on Tammuz in these three strophes are elements from the funeral ritual. Ishtar is treating Tammuz as a corpse. Finally, the line rendered in the earlier translations as "on the day when Tammuz comes up" has been shown to be a mistranslation. It either refers to Tammuz greeting Ishtar (i. e., coming up to her) in the underworld, or it is a reference to the month Tammuz. In the Akkadian version, Tammuz is dead and remains so. Such an understanding is witnessed to in other Akkadian texts. For example, in the Epic of Gilgamesh (6.4650), the hero insults and scorns Ishtar, reminding her that all her previous loversTammuz heads the listhave died as a result of their relationship to her.

Such considerations seemed to become purely academic with the publication of the Sumerian prototype of the Akkadian text, Inanna's Descent to the Netherworld (Inanna is the Sumerian form of Ishtar) and the closely related Death of Dumuzi. These early texts made clear that the goddess did not descend to the realm of the dead to rescue her consort. Rather it was her descent that was responsible for his death.

Inanna, the queen of heaven, sought to extend her power over the underworld, ruled by her sister, Ereshkigal. As in the Akkadian text, Inanna descends through seven gates, at each removing an article of clothing or royal regalia until, after passing through the seventh gate, she is naked and powerless. She is killed and her corpse hung on a hook. Through a stratagem planned before her descent, she is revived, but she may not return above unless she can find a substitute to take her place. She reascends, accompanied by a force of demons who will return her to the land of the dead if she fails. After allowing two possible candidates to escape, she comes to Erech, where Dumuzi, the shepherd king who is her consort, appears to be rejoicing over her fate. She sets the demons on him, and after he escapes several times, he is captured, killed, and carried off to the underworld to replace Inanna. In this narrative, Dumuzi is a dying god.

In 1963 a new portion of the Descent of Inanna was announced. Here, it would appear, there is yet a further episode. Inanna, in response to Dumuzi's weeping, decrees an arrangement whereby Dumuzi will take her place for half the year in the underworld and then return to the realm of the living; his sister, Geshtinanna, will then take Dumuzi's place in the underworld for the other half of the year, and, likewise, return.

For some scholars, this new conclusion to Inanna's Descent was sufficient to restore Dumuzi/Tammuz to the class of dying and rising gods. Such an understanding is unlikely. The myth emphasizes the inalterable power of the realm of the dead, not triumph over it. No one ascends from the land of the dead unless someone takes his or her place. The pattern of alternationhalf a year below, half a year aboveis familiar from other myths of the underworld in which there is no question of the presence of a dying and rising deity (e.g., Persephone, as in Ovid, Fasti 4.6134, or the youthful Adonis as described above), and is related, as well, to wider folkloristic themes of death delayed if a substitute can be found (e.g., Stith Thompson, Motif-Index A 316; D 1855.2; P 316). Such alternation is not what is usually meant in the literature when speaking of a deity's "rising."

As the above examples make plain, the category of dying and rising deities is exceedingly dubious. It has been based largely on Christian interest and tenuous evidence. As such, the category is of more interest to the history of scholarship than to the history of religions.

See Also



The classic formulation of the dying and rising pattern was made by James G. Frazer in The Golden Bough, 3d ed., 12 vols. (London, 19111915), esp. vol. 4, The Dying God (1912), p. 6. Frazer cites a representative sample of the older scholarly literature. A full bibliography, from the perspective of Old Testament scholarship, is supplied in Karl-Heinz Bernhardt's Das Problem der altorientalischen Königsideologie im Alten Testament (Leiden, 1961). Günter Wagner's Pauline Baptism and the Pagan Mysteries, translated by J. P. Smith (Edinburgh, 1967), offers not only a full bibliography from the perspective of New Testament research but also a brilliant critique of the notion of dying and rising deities.

For Adonis, the old collection of all the relevant texts and testimonia by W. W. Baudissin, Adonis und Esmun (Leipzig, 1911), has been partially superseded by Wahib Atallah's Adonis dans la littérature et l'art grecs (Paris, 1966). The most consistently critical position toward Adonis as a "rising god" is in Pierre Lambrechts's "La 'resurrection' d'Adonis," in Mélanges I. Lévy (Brussels, 1955), pp. 207240; compare Lambrecht's Over Griekse en Oosterse mysteriogodsdiensten: De zogenannte Adonismysteries (Brussels, 1954).

The relevant texts on Aliyan Baal are collected and translated in Cyrus H. Gordon's Ugaritic Literature (Rome, 1949) and Godfrey R. Driver's Canaanite Myths and Legends (Edinburgh, 1956), both of which reject the dying and rising pattern. Theodor H. Gaster is thoroughly convinced of its applicability; see his Thespis: Ritual, Myth, and Drama in the Ancient Near East, 2d rev. ed. (1961; reprint, New York, 1977). Arvid S. Kapelrud is more cautious; see his Baal in the Ras Shamra Texts (Copenhagen, 1952).

Hugo Hepding's old collection and typology of sources for Attis, Attis: Seine Mythen und sein Kulte (1903; reprint, Berlin, 1967), remains standard. The fundamental work on Attis as a dying and rising god is a series of publications by Lambrechts: "Les fêtes phrygiennes de Cybèle et d'Attis,"Bulletin de l'Institut Historique Belge de Rome 27 (1952): 141170; Attis: Van herdersknaap tot god (Brussels, 1962); and Attis en het feest der Hilariën (Amsterdam, 1967).

For Marduk, the text of the New Year ritual is available in English translation by A. Sachs as "Temple Program for the New Year's Festival at Babylon," in Ancient Near Eastern Texts relating to the Old Testament, 2d ed., edited by J. B. Pritchard (Princeton, 1955), pp. 331334. The Death and Resurrection of Bel-Marduk is available in a less adequate translation by Stephen H. Langdon: The Babylonian Epic of Creation (Oxford, 1923), pp. 3449. A shorter recension, with an important essay that challenges the parodic interpretation, has been translated by Tikva Frymer-Kensky: "The Tribulations of Marduk: The So-Called 'Marduk Ordeal Text,'" Journal of the American Oriental Society 103 (JanuaryMarch 1983): 131141. The major critical treatment of Marduk as a dying and rising god is Wolfram von Soden's "Gibt es ein Zeugnis dafür das die Babylonier an die Wiederauferstehung Marduks geglaubt haben?" Zeitschrift für Assyriologie 51 (May 1955): 130166.

The most useful treatment of Osiris, with full critical bibliography, is contained in the notes and commentary of J. Gwyn Griffiths's Plutarch's De Iside et Osiride (Cardiff, 1970). The newest material on Dumuzi and Inanna, with bibliography for the older, is found in Samuel Noah Kramer's The Sacred Mar riage Rite: Aspects of Faith, Myth, and Ritual in Ancient Sumer (Bloomington, Ind., 1969). Kramer and Diane Wolkstein's Inanna (New York, 1983) provides a highly literary translation.

Jonathan Z. Smith (1987)