Honor Thy Gods: Popular Religion in Greek Tragedy

By Jon D. Mikalson

In Honor Thy Gods Jon Mikalson uses the tragedies of Aeschylus, Sophocles, and Euripides to explore popular religious beliefs and practices of Athenians in the fifth and fourth centuries B.C. and examines how these playwrights portrayed, manipulated, and otherwise represented popular religion in their plays. He discusses the central role of honor in ancient Athenian piety and shows that the values of popular piety are not only reflected but also reaffirmed in tragedies.

Mikalson begins by examining what tragic characters and choruses have to say about the nature of the gods and their intervention in human affairs. Then, by tracing the fortunes of diverse characters -- among them Creon and Antigone, Ajax and Odysseus, Hippolytus, Pentheus, and even Athens and Troy -- he shows that in tragedy those who violate or challenge contemporary popular religious beliefs suffer, while those who support these beliefs are rewarded.

The beliefs considered in Mikalson's analysis include Athenians' views on matters regarding asylum, the roles of guests and hosts, oaths, the various forms of divination, health and healing, sacrifice, pollution, the religious responsibilities of parents, children, and citizens, homicide, the dead, and the afterlife. After summarizing the vairous forms of piety and impiety related to these beliefs found in the tragedies, Mikalson isolates "honoring the gods" as the fundamental concept of Greek piety. He concludes by describing the different relationships of the three tragedians to the religion of their time and their audience, arguing that the tragedies of Euripides most consistently support the values of popular religion.

• Language: English
• Publisher: The University of North Carolina Press
• Release date: Mar 19, 2014
• ISBN: 9781469617183

Jon D. Mikalson

Jon D. Mikalson is Professor of Classics at the University of Virginia. He is the author of The Sacred and Civil Calendar of the Athenian Year (1975), Athenian Popular Religion (1983), and Honor Thy Gods: Popular Religion in Greek Tragedy (1991).

Book preview

Honor Thy Gods

Honor Thy Gods

Popular Religion in Greek Tragedy

Jon D. Mikalson

The University of North Carolina Press

Chapel Hill and London

© 1991 The University of North Carolina Press

All rights reserved

Manufactured in the United States of America

The paper in this book meets the guidelines for permanence and durability of the Committee on Production Guidelines for Book Longevity of the Council on Library Resources.

95 94 93 92 91 5 4 3 2 1

Library of Congress Cataloging-in-Publication Data

Mikalson, Jon D., 1943-

Honor thy gods : popular religion in Greek tragedy/Jon D. Mikalson.

      p. cm.

Includes bibliographical references and index.

ISBN-13: 978-0-8078-2005-6 (alk. paper)

ISBN-10: 0-8078-2005-9 (alk. paper)

ISBN-13: 978-0-8078-4348-2 (pbk.: alk. paper)

ISBN-10: 0-8078-4348-2 (pbk.: alk. paper)

1. Greek drama (Tragedy)—History and criticism. 2. Gods, Greek, in literature. 3. Religion in literature. I. Title.

PA3136.M54 1991     91-50282

        CIP

τὸ πλῆθoς ὅ τι

τὸ ϕαυλότεϱον ἐνόμισε χϱη̃-

ταί τε, τόδ’ ἄν δεχοίμαν

Euripides Bacchae 430–432

Contents

Preface

Abbreviations

ONE Introduction

Mythology and Popular Religion

The Fragments of Greek Tragedy

Irony and Interpretation

The Need to Particularize

Tragedy and Cult Prehistory

TWO The Deities

The Presence of a God

Gods and Daimones

Gods and Heroes

Divine Intervention

THREE Challenges to Popular Religious Beliefs

Asylum

Xenia

Oaths and Perjury

Divination

Death, Burial, and Afterlife

Conclusion

FOUR The Pious and the Impious

Creon and Antigone

Ajax and Odysseus

Hippolytus

Pentheus, Teiresias, and Cadmus

Troy and Athens

Polyphemus

Conclusion

FIVE Piety and Honor

Pieties and Impieties

Piety and Justice

Piety, Folly, Sophrosyne, and Hybris

Piety and Tιμή

SIX The Tragedians and Popular Religion

Aeschylus

Sophocles

Euripides

Exodos

Notes

Bibliography

Index of Passages Cited from Tragedy

General Index

Preface

The religion found in Greek tragedy is, like the language of Homer, a complex hybrid, a hothouse plant which never did and probably never could exist or survive in real life. Although the components of the hybrid are heterogeneous, they have been fused so expertly by the genius of the poets that the original distinctions between them are seldom apparent. At the risk of considerable oversimplification, we may designate as major components of religion in tragedy (1) the anthropomorphic deities of the Homeric pantheon; (2) beliefs and deities once part of popular religion but in the classical period virtually extinct; (3) deities, beliefs, practices, and cults of contemporary fifth-century society; (4) a concern with the morality and justice of the gods; and, finally, (5) contemporary or recent philosophical conceptions of deity. In varying degrees each element has itself influenced or been influenced by others, and traces of one can usually be found in another. Despite this, each has, for purposes of analysis, a fairly distinct character, and taken together they provide a context for understanding religion in tragedy in relation both to the beliefs and practices of the audience and to its uses as a vehicle of literary expression.

To further complicate the situation, this potpourri of beliefs, cults, deities, and myths is set in a legendary period, within a generation or two of the Trojan War, a dramatic time when gods and rituals are often in the process of becoming what was familiar to the fifth-century audience. Often the gods themselves are changing and new cults are being introduced. Several tragedies dramatize the very moments when gods, heroes, cults, and rituals are transformed from what the poet imagined them to be in a legendary past to the form that was familiar to the audience of the fifth century.

Attention to religion in tragedies has centered almost entirely upon the behavior of gods and upon the poet’s and modern critic’s evaluation of that behavior. On stage and off, gods do or do not speak or intervene in certain situations, and modern critics discuss at great length whether such behavior is moral or immoral by the standards of the critic, of the playwright, and of the fifth-century audience. The poet is judged to be religious and pious if he represents his deities as just and moral. Some deities of tragedy behave in ways which are, to say the least, problematic for a moralizing critic, and as a result their creator is, depending on the preferences of the critic, praised or damned for being a godless detractor of traditional religion. The underlying premises of such discussions of religion in tragedy are (i) that Olympian deities like Zeus, Apollo, Aphrodite, and Poseidon, as they are depicted in tragedy, were believed in and worshiped by fifth-century Athenians, and (2) that ordinary Athenians of the time included among their religious concerns the expectation that gods be just with a justice similar or identical to that expected among men. Both assumptions are open to grave doubt and require, at the least, examination in terms of what we know of fifth-century religion.

There are, however, many more elements of religion, and many elements perhaps more religious in Greek tragedy than the anthropomorphized Olympian deities whose virtues and vices the critics have meticulously weighed. My interest is the religious beliefs and attitudes of ordinary Athenians of the classical period. These are the beliefs about which they spoke and upon which they acted in daily life. In Athenian Popular Religion I collected a corpus of these beliefs as they were expressed in orators, historians, and inscriptions of the late fifth and fourth centuries B.C. The result was a set of beliefs, usually expressed casually, briefly, and starkly, beliefs for which an Athenian speaker or writer expected general acceptance among his audience. When we find similar beliefs expressed in the tragedies, we are now able to identify them, with some justification and confidence, as beliefs commonly held by the audience of tragedy. From the orators, inscriptions, and Xenophon we have the skeletal structure of this system of beliefs. Tragedies afford the possibility of adding flesh to this skeleton, because there characters in lifelike situations express these beliefs, talk of the reasons behind them, and act upon or against them.

In the plays many characters act with what would be, by the criteria of popular religion, exemplary piety or impiety. Occasionally these matters are the focus of an entire play; more often they arise only incidentally to the main action. But even short scenes allow us to discover considerably more about the nature of popular beliefs and the rationale behind them. We may know, for example, from the offhand comment of an orator or historian that it was impious to violate the rights of asylum, but when a dramatist places on stage suppliants pleading with a king to protect such rights, the abstract belief is brought to life. We hear arguments and issues, religious and other, that may be involved in such a situation. The belief, mise en scène, is given flesh and blood.

In addition, since we have knowledge of these beliefs independently from the poetic tradition, we can often isolate instances in which tragedians, individually or as a group, refashioned popular elements so as to make them something different from that commonly believed and practised by their audiences. It is at this point that we must return to the hybrid character of religion in Greek tragedy, because when a popular belief is remodeled or recast, it is usually done so under the influence of one of the other elements of that hybrid.

I consistently speak of religion in Greek tragedy and not of the religion of tragedy, because the individual dramatists employ religion in such very different ways. Aeschylus, Sophocles, and Euripides put different emphases and interpretations on religious aspects of, for example, perjury and asylum, matters which we know were of religious concern to their audiences. Often the three tragedians present beliefs and practices in just the form and context in which we find them in popular religion. When, however, they vary or refashion these popular beliefs, they may be drawing from quite different elements of the hybrid of tragic religion.

There has been an enduring interest, since antiquity, to rank the tragedians, to find the most religious, the least religious, and so forth. This quest, focusing as it has on only the treatment of the deities, has been, in terms of the practised religion of the period, misleading. Should we grant that such ranking has merit, the scores will vary considerably depending on whether we look at the behavior of the anthropomorphic gods, the presentation of the myths, the philosophical and theological theories introduced, or the popular religious beliefs expressed. For this last, my primary interest, we now are able to investigate whether a poet, in assigning rewards and punishments to characters who act, by contemporary standards, piously or impiously, is in accord with the beliefs of his audience. Consistent accord would suggest that the poet accepted, supported, and even promoted conventional religious beliefs, whereas frequent disagreement would indicate dissatisfaction and attempts at reform.

My focus and emphasis throughout are on the religious beliefs and practices which can be shown to be those of the audience of the fifth and fourth centuries. I do not propose to describe synoptically the religion of tragedy, a task which may well be impossible given the differences between the tragedians and even between plays by the same poet. I hope to delineate better the religion of the Athenian people of the classical period, both as that religion was practised and as it was represented in their theater. Tragedy often allows us to get beyond the simple statement of a belief to ideas lying behind it and to actions resulting from it. We also have in tragedies the opportunity to examine how the poets manipulated popular beliefs for particular literary and dramatic purposes. Despite their presentation in religious festivals, most tragedies are not, fundamentally, about religion. They are about exceptional men and women in exceptional situations. There is to most of these situations a religious dimension; it is only one of several dimensions, but the one, I think, least understood and explicated by modern critics. It is upon this and upon fifth-century Athenian popular religion that I would hope to shed a little light.

This study is the product of several years’ work, the most pleasant and profitable of which I owe to Glen Bowersock, Christian Habicht, and their colleagues at the Institute for Advanced Study in Princeton. P. David Kovacs, Richard Hunter, Robert Garland, and Mary Lefkowitz have, in an encouraging and gentle way, offered many criticisms and suggestions, and where I have been sufficiently wise to accept them, the book has been considerably improved. I express my gratitude to them, and to Gail Moore and her staff, who have provided much valuable assistance and expertise in the modern-day mechanics of manuscript preparation. I thank Mary and Meli Mikalson for their careful and patient help with the proof and indexes.

A Note on Spelling

The names of ancient authors and their works and the spellings of the names of characters of the tragedies are, with a few exceptions, those of the second edition of the Oxford Classical Dictionary (1970). The epithets of deities and some place names are, however, in the Greek, not Latinized, style.

Abbreviations

Full references for all works cited appear in the Bibliography. Textual editions are commonly indicated in citations by author’s last name only (see Editions Cited in the Bibliography). Journals, major reference works, and ancient works are cited in the general style of abbreviations in L’année phil-ologique and the second edition of the Oxford Classical Dictionary.

The Tragedies

A. Ag. Aeschylus Agamemnon Ch. Choephoroi Eum. Eumenides Pers. Persae (The Persian Women) Pr. Prometheus Bound Suppl. Supplices (The Suppliants) Th. Septem contra Thebas (Seven against Thebes) E. Alc. Euripides Alcestis Andr. Andromache Ba. Bacchae Beller. Bellerophon Cycl. Cyclops El. Electra Erech. Erechtheus Hec. Hecuba Hel. Helen Heraclid. Heraclidae HF Hercules Furens (Heracles) Hipp. Hippolytus Hyp. Hypsipyle IA Iphigeneia Aulidensis (Iphigeneia in Aulis) Ion Ion IT Iphigeneia Taurica (Iphigenia among the Taurians) Med. Medea Or. Orestes Ph. Phoenissae (The Phoenician Women) Sthen. Stheneboea Suppl. Supplices (The Suppliants) Tr. Troades (The Trojan Women) Rh. Rhesus S. Aj. Sophocles Ajax Ant. Antigone El. Electra OC Oedipus Coloneus (Oedipus at Colonus) OT Oedipus Tyrannus Ph. Philoctetes Tr. Trachiniae (The Trachinian Women)

Modern Works

A Austin, Nova Fragmenta Euripidea APR Mikalson, Athenian Popular Religion Burkert, GR Burkert, Greek Religion DK Diehls and Kranz, Die Fragmente der Vorsokratiker IG Inscriptiones Graecae Jacoby, FGrHist Jacoby, Die Fragmente der griechischen Historiker K Kannicht and Snell, Tragicorum Graecorum Fragmenta, vol. 2 N Nauck, Tragicorum Graecorum Fragmenta² (with supplement by B. Snell) P Page, Greek Literary Papyri R Radt, Tragicorum Graecorum Fragmenta, vol. 3 (Aeschylus) and 4 (Sophocles) RAC Reallexikon für Antike und Christentum RE Realencyklopädie der classischen Altertumswissenschaft

Journals

ABSA Annual of the British School at Athens AJAH American Journal of Ancient History AJP American Journal of Philology BICS Bulletin of the Institute of Classical Studies CJ Classical Journal CP Classical Philology CQ Classical Quarterly GR Greece and Rome GRBS Greek, Roman and Byzantine Studies HSCP Harvard Studies in Classical Philology HThR Harvard Theological Review JHS Journal of Hellenic Studies REG Revue des études grecques RhM Rheinische Museum SIFC Studi italiani di filogia classica TAPA Transactions of the American Philological Association YCS Yale Classical Studies ZPE Zeitschrift für Papyrologie und Epigraphie

One: Introduction

In Athenian Popular Religion I collected and described religious beliefs of ordinary Athenians of the late fifth and fourth centuries B.C. as they appeared in the writings of contemporary orators and historians and in dedications, epitaphs, sacred laws, and cult regulations. My attention centered not on poets and philosophers but on the townspeople of Athens, average Athenian citizens, those who, no doubt, constituted the majority of the audience of fifth-century Greek tragedy and comedy. A corpus emerged, not surely of all religious beliefs of the time, but of a good number which Athenians expressed publicly and for which they expected to find widespread acceptance.

At that time I excluded tragedy and comedy as reliable sources of information for popular belief, not because they had little to offer,¹ but because I found it impossible to isolate elements of popular religion in these poetic genres until I had developed a rather precise notion of what was—and what was not—popular religion. When a character in a play expresses an opinion on a religious topic such as divination, oaths, or prayer, there had been little, apart from the instincts of the scholar, to determine whether this character’s view was one widely held by members of the audience or was somehow at variance with the usual view. Too often the procedure, especially among Euripidean scholars, was simply to dismiss statements of belief in the gods or in their justice and benevolence as popular or traditional and to label criticisms of myths, affirmations of atheism, or statements even faintly resembling current philosophical theories as enlightened.² Crudely put, what appealed to the scholar’s conception of what religion should be was praised as enlightened. The rest, often the object of disdain, was merely traditional or popular.

Statements about gods and religion abound in tragedy and comedy, and it would have been relatively easy to list or describe in Athenian Popular Religion poetic passages which parallel, to some degree, what is to be found in the prose sources. I chose not to do this, because even a cursory examination of the material warned me that the examination of popular religion in drama was no simple matter. The poets have interwoven these elements into the fabrics of their plays and, for the most part, they cannot simply be torn from their contexts for citation. A religious belief or concept may be expressed or even challenged in a play, but its full import or a response to the challenge may appear only hundreds of lines later. Often too the personal character, good, bad, or indifferent, of the individual expressing the belief is important in our evaluation. But it is not simply that evil characters speak irreligiously and good characters piously. Occasionally wicked characters use, in distorted ways, widely held religious beliefs to justify evil actions. The context—both the immediate context of the passage and the play as a whole—must be taken into consideration in interpreting and judging the import and significance of statements of a religious as well as of a moral nature.

Also, in contrast to orators, historians, legislators, and common citizens, tragic poets present religious views and topics in the framework of a legendary, largely mythical past. These are the days of kings, of Agamemnon, Oedipus, and Theseus. The resulting social and religious order is, in several ways, more akin to that of Homer than to that of fifth-century Athens. Into this legendary past are fit moral, political, and religious concepts and values of the fifth century,³ but in the time shift they too can be transmuted in subtle and significant ways, and we must take cognizance of this.

But perhaps the most significant difference, and one that most seemed to recommend two separate studies, is that in the Greek tradition poets, unlike orators, historians, and governmental and religious authorities, had the inclination and license to alter, transform, and challenge current and traditional religious views and practices. Here we must be alert to how, and for what purposes, a poet is presenting or manipulating a religious datum known from other less inventive or polemic sources.

In understanding this process of poetic adaptation of current religious beliefs, we must give some preliminary attention to (1) the relationship of poetic gods, mythology, and popular religion, (2) the appropriate use of fragments from lost plays, (3) ironical interpretations of plays and scenes, (4) the tendency of Greek religion to particularize deities by locale and function, and (5) the depiction of cultic prehistory in tragedy.

Mythology and Popular Religion

Scholarly discussions of religion in Greek tragedy have treated almost exclusively the morality and the justice of individual gods. Is Zeus just or unjust, moral or immoral, in his neglect of his son Heracles in Sophocles’ Trachiniae? What are we to think of Athena, who punishes mercilessly her former favorites in Sophocles’ Ajax and Euripides’ Troades? Is Aphrodite justified and is she wise in her vengeance on Hippolytus in Euripides’ play of the same name? What of Apollo in the Ion, a god who raped a mortal woman, seemingly abandoned her and his child, gave a false oracle, and then lacked the courage to face the music as deus ex machina? How much responsibility and guilt should be assigned this same Apollo in the versions the dramatists give of Orestes’ murder of his mother? Almost universally, real or presumed moral censures of gods in the plays are viewed as attacks by the poets on religion and on contemporary religious beliefs. This, however, is true if, and only if, these gods with their human personalities and passions formed part of the religious beliefs of the fifth-century Athenian public.

What most surprised and in fact disappointed me as I gathered the sources for Athenian Popular Religion was how very separate from these gods of poetry were the everyday beliefs, practices, and religious concerns of fifth-and fourth-century Athenians.⁴ The gods and individual gods were often spoken of, but never in those terms familiar to us from Homer and other poetry. The relationship of the deities of poetry to the deities of cult and worship is ill-defined and problematic, largely because it has not been systematically investigated in either theoretical or practical terms. Students of Greek religious cults occasionally note the differences between the gods of cult and those of myth and literature, whereas most literary critics presume that, for example, an Athena of a Sophoclean play is to be identified, at some level, with the Athena whom Athenians worshiped in their city. But what is this level? The gods of cult and poetry shared names, and this of course suggests some identification, but, to put it simply, they shared first names only. We do not know whether an Athenian, as he made his morning offering at the little shrine of Zeus Ktesios in his house, thought of Homer’s thunder-bearing, cloud-gathering Zeus. There is no evidence that he did, and the two deities, both named Zeus, are very different in both appearance and function.⁵ Athena Polias was the namesake and mistress of Athens, with two major sanctuaries, the Erechtheum and the Parthenon, on the Acropolis,⁶ but one is hard put to demonstrate that in cult she was assigned any of those features which most characterize the Athena of Sophocles’ Ajax or Euripides’ Troades. Likewise the Aphrodite of Euripides’ Hippolytus appears quite unlike the Aphrodite worshiped in Athens.⁷ And, conversely, there are among the major divine actors of tragedy no deities similar to Zeus Ktesios, Zeus Herkeios, Demeter, Athena Hygieia, Asclepios, and most other deities central to Athenian worship.

Because of the limited evidence available, it is unlikely that we shall ever satisfactorily understand the relationship of cult deities to those of poetry, but the most promising avenues of approach are iconography and aetio-logical myths. Cult statues, dedications, and vase paintings often do represent cult deities in the manner and garb we know from the poetic literature, and we may expect that worshipers visualized some of their gods in this form. But whether this visual image carried over to the attribution of the functions and virtues and vices that these deities are given in poetry is by no means certain. Here we may turn to aetiological myths, myths which provide an explanation of a local deity, cult, institution, or practice. Localized aetiological myths are the point at which worship and mythology intersect, and the myth, we may assume, reveals something of what local people thought about the deity, ritual, or cult.⁸ Some such aetiological myths were dramatized, as, for example, that of Athena in Aeschylus’ Eumenides and of Apollo in Euripides’ Ion, and we shall consider these in turn. But aetiological myths form, if not in origin then at least by the fifth century, a group quite distinct from the myths taken from the epic cycles with their Panhellenic, anthropomorphic, and free-spirited Olympian gods and goddesses. My work in Athenian cult and belief inclines me to see the differences between these deities of the Panhellenic poetic tradition and the deities worshiped in Athens. The deities of poetry were well known, they were loved and hated in the literary context, and they were praised or criticized by poets and philosophers for ethical and theological reasons, but they were not worshiped as cult deities were. In the form that Homer and the tragedians present them, they did not have temples, sanctuaries, or altars in Greek cities. They did not receive dedications, sacrifices, or prayers. The gods of poetry are, I would claim, the products of literary fantasy and genius, not of the Greek religious spirit. Criticisms of these gods and of the myths encompassing them need not be criticisms of contemporary religion and of its beliefs and practices.⁹

Others may understandably judge differently the gods of poetry and their relationship to the deities of cult and may be reluctant to deny the deities they know to the religious beliefs of the Athenians. But all will agree, I trust, that in the wealth of tragedy there are many elements of religious belief apart from these anthropomorphic gods, elements which have been overlooked in our concentration on the actions and morality of the individual deities. We know, for example, that piety consisted of belief in and reverence for the gods of the city, maintenance of oaths, respect for the rights of asylum and hospitality, observance of tradition and law in cult sacrifices and in tendance of the dead, loyalty to one’s country, and proper care of one’s living parents. The person who violated oaths, maltreated those having asylum, violated traditional practices in sacrifice or the rites of the dead, betrayed his country, neglected his parents, or committed murder or other similar acts was impious and incurred the hostility of the gods, a hostility which might manifest itself in the suffering and misfortune of him, of his family, or even of his country.¹⁰ It is this complex of ideas—the corpus of Athenian popular religious beliefs—and its presentation in tragedy to which I wish to direct attention. Apollo, Athena, and Zeus will appear now and then, but this is not a study of the gods of Greek tragedy. That is more properly a topic of mythog-raphy, literary criticism, and intellectual history.

The Fragments of Greek Tragedy

Whether they refer to religious beliefs or anything else, two or three lines of a play require a context in order to be interpreted. Even with that context, the interpretation is sometimes problematical. Interpretation becomes virtually impossible when those two or three lines are all that we possess of a play. Well over three thousand fragments of lost plays by Aeschylus, Sophocles, and Euripides have survived, discovered on papyri or quoted by grammarians, scholiasts, anthologists, and other late sources. Some of these consist of only a single word, others of single lines, short passages, whole speeches, or even entire scenes. Many were copied out by anthologists just because they contain a complete, pointed, and epigrammatic statement of a moral or religious idea. The following examples, from Euripides, are characteristic of this latter kind:

The gods turn the errors of parents upon their children.

(980[N])

Even if someone laughs at the story, Zeus and the gods exist, looking upon human sufferings.

(991[N])

The gods, who are more powerful than we, trip us up with many forms of clever tricks.

(972[N])

The prophet who guesses well is best.

(973[N])

The Mind (Nous) is a god in each of us.

(1018[N])

Such fragments have played, vis-à-vis the complete plays, a disproportionately large role in discussions of religion in tragedy, particularly of Euripides’ own religious beliefs and his criticisms of mythology and contemporary religion. Wilhelm Nestle, for example, in his highly influential study of the Theologie of Euripides regularly argues from these fragments to his conclusion that Euripides, strongly influenced by Heraclitus and contemporary philosophy, rejected popular and traditional religion, developed his own consistent theology, and himself became the poet of the Greek enlightenment.¹¹ Nestle then fits, sometimes with considerable difficulty, complete plays into a scheme which he has constructed from the fragments. This he sometimes is able to do only by invoking Verrallian irony, assuming that, to those few in the audience as sophisticated as the poet and the critic, the play meant something different from or opposite to its apparent meaning. In their emphasis on the fragments many other scholars who have written on Euripides are little more circumspect and judicious than Nestle.¹²

To scholars, as to anthologists, these fragments are attractive as precise and elegant statements of religious principle or belief. They also offer the engaging prospect of composing, from them and the myth cycle, one’s own Greek play. But—and for our purposes this is the critical flaw—these fragments lack context. We are never able to know what motivated the statement, whether, for example, it is a momentary doubt expressed by a basically pious character or a reasoned argument by a blatantly impious individual, or whether, if it is a challenge to conventional belief, it was ultimately justified or refuted.¹³ We may guess, of course, but in a study of religion in Greek tragedy there is little need for such guesswork when we have at hand thirty complete or nearly complete plays.

A brief but telling example of the unreliability of fragments is Hippolytus’ famous line, spoken when he recognizes the nature of the secret (his stepmother’s passion for him) which he had sworn not to reveal: My tongue has sworn the oath, my mind has not (E. Hipp. 612). Maintenance of one’s oath was considered an important, perhaps the key, element of personal piety.¹⁴ Had the play been lost and only this line, quoted by Aristotle (Rh. 1416a32), survived, it would be touted as an example of Euripides’ cynical treatment of contemporary religion and morality. In fact it has been treated just this way by critics from Aristophanes to the present day,¹⁵ despite the fact that the whole play survives and we learn that Hippolytus never acts upon these words. Quite the contrary. Four hundred lines later he says expressly that for religious and practical reasons he will not violate his oath (1060–1063; cf. 657–658).¹⁶ When a line like this, standing in the midst of a complete play, can be so misused by critics, we can hardly put our trust in lines or short passages where the context is altogether lacking or, at best, can be only hypothetically restored.¹⁷

My own experience in epigraphy suggests that it is a mistake to argue from restorations of fragmentary sources to or against complete texts. We must begin with the whole plays, for there we learn the character of the individual speaking the lines, the circumstances in which he spoke them, and, if that is made an issue, their ultimate resolution. What the Euripidean fragments best illustrate is something we can also see clearly from complete plays: that Euripides was willing, somewhat more willing than either Aeschylus or Sophocles, to put into the mouths of characters statements that were blasphemous, impious, or otherwise somehow unconventional. As George Grube warned us years ago, the occurrence of a religious statement in a fragment indicates only that such a view could be expressed by some character or chorus in some setting in tragedy.¹⁸ That and little more we have to learn from the fragments. This study has been based on complete or nearly complete plays, and fragments are introduced only as parallels, in thought and language, to passages about which we are better informed.

Irony and Interpretation

Since the end of the nineteenth century a major concern of literary critics, particularly those of Euripides, has been the detection of irony.¹⁹ This is not the dramatic irony so characteristic of Sophocles, whereby the audience has been told explicitly something of which a major character is unaware and whereby that character’s words take on, unbeknownst to him, double or treble meanings. This broader irony, deep irony, so important to modern criticism but unknown to theoreticians of antiquity, consists of, to give Dr. Johnson’s definition, Saying one and giving to understand the contrarye.²⁰ Apparently straightforward statements and claims in the plays are belied by the use of subtly undercutting words, by allusions to other plays, to philosophical theories, or to variant forms of the myth, by deliberate but often slight contradictions within or between scenes, by the use of images and symbols opposed to words or actions of characters, or in the expectation that members of the audience assumed, to give pertinent examples, that all prophets are corrupt and lie, or that "on the Euripidean stage whatever is said by a divinity is to be regarded, in general, as ipso facto, discredited."²¹ Modern ironologists call into question the apparent meaning of virtually every scene and every play, and since Verrall’s day much of the irony they have detected has been directed against gods, mythology, and conventional morality and religion.

The discovery of this deep irony requires an exceptionally close reading of the text (or, in some cases, of the subtext) of the play, and for those less devoted to literary theory and more interested in the Greeks as a people the question naturally arises, to whom in antiquity was this irony apparent? The critics who concern themselves with this question seem to agree that only a small group (in addition to the poet, of course) of enlightened, sophisticated, and intelligent spectators would grasp the ironic or full meaning of the play.²² The rest, the great mass of the Athenian public, as Nestle puts it, would have to content themselves with the play’s apparent meaning.²³

My interest happens to be with this large group of average Athenians, the great mass of the Athenian public, and I am most concerned to understand how they saw their own religious beliefs dramatized upon the stage. They may have lacked sufficient literary training and imagination to follow the ironic subtext of, say, Euripides’ Suppliants, but they did have their own religious beliefs and every year saw these represented, in various and sundry ways, in their theater. Since my topic is the average Athenian citizen, his religious beliefs, and their presentation on the Athenian stage, I may perhaps be forgiven for not embracing or even repeating interpretations of plays, passages, and characters which even their proponents consider intelligible to only a handful of the audience. Those few enlightened Athenians, if they existed, would themselves be a fascinating subject of inquiry, but they have little to contribute to the subject of Athenian popular religion.

The Need to Particularize

The Helen of Euripides’ Helen, produced in 412, spent the years of the war chastely in Egypt, while only an image of her was in Troy, sleeping with Paris. The Helen of Euripides’ Troades, a play produced just three years earlier, was with Paris at Troy and is a real Helen; no thought of an image or of a Helen in Egypt is possible. In Sophocles’ Antigone Creon defies the laws of the gods and lacks basic human sympathies. In the Oedipus Tyr annus he is not heroic, to be sure, but full of respect for the gods and of concern for his fellow man. In the Oedipus at Colonus we find him again impious and villainous. The Agamemnon of Sophocles’ Ajax is as impious and unfeeling as his brother Menelaus in attempting to prevent Ajax’ burial, but in Euripides’ Hecuba an Agamemnon with some moral authority assists Hecuba in her vengeance on the immoral, impious Polymestor. Such variations in conception of the same myth cycle and the same character between plays could be multiplied manyfold. Tragic poets had the license, within rather broad limits, to choose or create for each play one version of the myth and one conception of their characters. We can expect in this regard to find consistency within a play, but have no right to expect it—without evidence to the contrary—between plays, even between plays by the same poet.

This principle is acknowledged for human characters and myths but is not nearly so well observed when gods, seers, rituals, or other aspects of religion become the topic of discussion or interpretation. A common assumption in critical literature, particularly that concerning Sophocles and Euripides, is that Zeus is Zeus, Apollo is Apollo, Artemis is Artemis, and that’s that. Deities are treated as fixtures with stable characters and characteristics, and what ill or good is said of a deity in one play may therefore be applied to the same deity in another play. This, of course, often results in the discovery of irony. But in fact characteristics of deities and often their fundamental natures vary markedly from play to play. The Artemises of Euripides’ Hippolytus and Iphigeneia among the Taurians are each carefully delineated but have almost nothing—apart from their names—in common. So too the Athenas of Aeschylus’ Eumenides and Sophocles’ Ajax, or the Zeuses of the Agamemnon and Prometheus. The tragic poet makes of a god, as of a myth or a human character, what he wants him to be for his purposes in a given play, and in this process he has, for at least two reasons, considerable latitude.

First, the gods of tragedy are, like all gods of Greek literature, hybrids with various and sundry components from poetic (especially epic) and philosophical traditions, from cult, and from the poets’ own inventiveness. The gods’ myths are much more broad, layered, and variegated than those of individual human characters such as Agamemnon and Creon, and the poet therefore had a far larger pool of material from which to create his god.

The second factor is the audience’s own familiarity with significantly different conceptions of the same deity in practised religion as well as in literature. Greek popular religion was highly particularized: in a particular place one appealed to a particular deity to fulfill a particular need.²⁴ As place (even within the same city) or need changed, so did the deity, often in basic nature and function if not in name. To Athenians Athena Polias, Athena Skiras, and Athena Hygieia were separate, for all practical purposes independent deities. They had separate myths, sanctuaries, cult officials, festivals, and rites, and they provided different services. In literature, where the possibilities of portrayal were so much more extensive, it was incumbent upon the poet to define the deity he chose to represent.²⁵

The deities of cult also varied significantly from one city-state to another, and dramatists, I think, paid more attention to this than is generally recognized. When Euripides set his Medea in Corinth, he adopted basically a Corinthian scheme of deities and cults. In Aeschylus’ Oresteia different deities become immediately concerned as the scene moves from Argos (in the Agamemnon) to Delphi and Athens (in the Eumenides). From the abundant and often contradictory cult and literary mythology of each god the poet chose what suited the locale and purposes of his play. For the next play he might choose considerably different or even contradictory elements of the same god. As a result we must take care not to assume that, for example, every Athena in tragedy is the Athenian Athena Polias, or that the audience would take each Athena in tragedy to be their Athena. If the dramatist wants an Athenian Athena, he must make her such, as Aeschylus does in the Eumenides and Euripides does in the Ion. The Athenas of the Troades and Ajax, however, are given no Athenian associations, and criticisms of them should not be transferred to the Athenas believed in and worshiped by Athenians.

The economy of Greek drama also affects the portrayal of deities. While an individual in real life might, in a year, turn to forty, fifty, or more deities to satisfy his religious needs, the tragedian had no wish to clutter the stage or diffuse attention by involving fifty, or ten, or even five separate deities in the action of one play. The major deity to play a role in each story was usually determined by the mythological tradition. For a tale of Heracles it must be Zeus and Hera. An Orestes would have his Apollo, an Ajax his Athena. But once the central deity was decided, there was a subtle but detectable tendency to concentrate upon him most or all the religious concerns of the poet and of the characters in the play, concerns which in real life fell into the bailiwick of quite different members of the pantheon. By limiting the number of deities in a given play to one, two, or a handful, the poets often loaded upon the chosen few deities functions and responsibilities which these gods appear not to have borne in practised religion. And, of course, a deity, once created, like a human character was intended to serve only one play. The same poet might, next year, attribute to a deity of the same name significantly different characteristics, functions, and concerns.

A similar variety, not so extensive but arising for many of the same reasons, may be found in the treatment of seers and religious rituals. By convention a seer in a Theban play must be Teiresias, as the Greek seer in a Trojan play must be Calchas. But that does not mean that in each play we have before us the same Teiresias or seer. The Teiresias of Sophocles’ Oedipus Tyrannus differs sharply from that of Euripides’ Bacchae, both as a seer and as a person. So too there is not just one prophetess figure. Theonoë of Euripides’ Helen has little resemblance to the Delphic Pythia of the Ion or the Cassandra of the Troades. Each of these characters has been particularized, not only in personality but also as prophet and prophetess, and criticisms of one in one play cannot simply be transferred to another in another play. Each was created for particular circumstances in a particular play.

Religious rituals, especially those of burial, death, and sacrifice, also afforded to the poet a rich and varied combination of elements, any one or group of which he could select to describe or emphasize. The rituals become, in each play, what the poet chooses to make them. In Euripides’ Troades Hecuba, Andromache, and the chorus invoke for help their deceased loved ones. No help comes; the prayers are in vain. In Aeschylus’ Persae the Persian queen invokes her dead husband Darius, and in response he appears—on the stage. This does not necessarily indicate a change in religious belief in the fifty-seven years between 472 and 415 B.C., or even differing beliefs of the two poets. Aeschylus wanted Darius present; Euripides did not want Priam, Hector, and a swarm of dead Trojan men in his play. Likewise Cly-temnestra’s killing of Agamemnon in Aeschylus’ Agamemnon, Aegisthus’ sacrifice in Euripides’ Electra, Heracles’ killing of his children in the Heracles, and the sacrifice of Iphigeneia in Iphigeneia at Aulis are each drawn from the same sacrificial ritual, but the effect of each is very different. For rituals interpretation has become complicated by the modern notion that certain rituals, qua rituals, provoked an automatic response in the audience. If, for example, the Greeks assumed sacrificial ritual was an expression of social solidarity, of the horror of violence, of the cosmic hierarchy of god and man, and of such things, then the ritual, qua ritual, becomes an independent determinant in interpretation of a play, a determinant which critics often find in opposition to words and actions of the characters and thus contributing substantially to the ironic subtext of the play.²⁶ There is, however, scarcely any evidence that the Greeks recognized consciously or subconsciously in their rituals what modern theoreticians make of them, and even less evidence from the plays themselves that the tragedians then played upon the resultant feelings in the composition of their plays. Until these connections are made, this new theoretical movement has little to tell us of popular religion in tragedy.

The gods, religious officials, and rituals are, within rather broad limits, what the poet chose to make them in each play. We must be careful how we argue from one play what they meant in another. We are on safer ground here, I think, to argue from what we know to be widely held religious beliefs, because they would apply equally to all the plays. But if we wish to understand how religion in tragedy is related to the practised religion of its time, we must attempt to see each deity, seer, and ritual in the context of its own play and not generalize overhastily what is in fact particular.

Tragedy and Cult Prehistory

The surviving tragedies, apart from Aeschylus’ Persae, are set in a legendary, prehistorical past. The dramatic time of most is the Trojan War, give or take a few generations. One, the Prometheus, is in much earlier days of mankind. These are the times of heroes past, times when, in several plays, Greek religion is in flux. New cults are being introduced, old cults are being changed, and functions and prerogatives of deities are being established. In most instances of change the movement is from