A new book by Christiane Sourvinou-Inwood (hereafter SI) is always a feast for the mind: broad knowledge, thorough investigation and acute — sometimes surprising — conclusions are to be expected. And it must be said that her new opus, Tragedy and Athenian Religion, meets all the requirements of the genre, and much more, as was also the case for her memorable Reading Greek Culture (1991) and Reading Greek Death (1996). A foreword by Gregory Nagy, editor of the series Greek Studies: Interdisciplinary Approaches, a second foreword by Richard F. Thomas, Chair of the Department of the Classics at Harvard University, and a short Preface by the author herself introduces the reader to the prehistory of the book, viz. the Carl Newell Jackson lectures SI gave at Harvard University in 1994. The book we now have in our hands reproduces her careful explorations of that time but achieves new goals by its larger proportions. In fact, it seems that SI was so taken by her theme that she lived with it for several years, giving short versions of it in lectures in London and Oxford, and finally produced a book “that turned out to be very much longer than was expected”.
I- Tragedy, Audiences and Religion (pp. 1-66)
The book is divided into four chapters. The first one, Tragedy, Audiences and Religion is a general introduction, where SI outlines the main themes of her study: Athenian tragedies “were, among other things, a discourse of religious exploration, part of the religious discourse of the polis”. Tragedy and Athenian Religion should therefore be considered as a continuation of SI’s previous work, mainly “What is polis religion ?” in O. Murray and S. Price eds., The Greek City from Homer to Alexander (1990), but it might also be considered as a continuation of R. Parker’s Athenian Religion: a History (1996) — as a matter of fact, SI acknowledges in her Preface R. Parker’s “learned interaction and friendship” (p. XVII). Because tragedy evolves in a ritual context, it can’t be a purely theatrical experience, and gods’ appearance on stage can’t be kept apart from gods in the religious area (pp. 1-14). At the same time, even if tragedies usually focus on a heroic past, the audience doesn’t feel completely estranged: aitiologies, references to the audience’s present and the presence of a chorus of male Athenian citizens on stage roots Athenian tragedy in the Athenian soil of the fifth century. Greek tragedies might be read as opera with universal meaning; to understand them means to read them in context, and the context is a religious one (pp. 14-50). Examples from Euripides’ Erechtheus, Iphigeneia in Tauris (pp. 25-40), Archelaos and Aeschylus’ Aitnaiai are discussed at length.
The two following chapters are meant to be read separately, each one bringing its own set of demonstrations and conclusions.
II- The Ritual Context (pp. 67-200)
The Ritual Context is mainly an archaeological and iconographical study reconstructing the Great Dionysia of the Classical Period and its meanings and recreating the beginnings of tragedy. A complete résumé is helpfully given at the end (pp. 197-200). The Great Dionysia began in the third quarter of the sixth century as a festival focussing on the ritual advent of Dionysos (xenismos) and the celebration of his cult. In the earliest form, there was a komos involving ritual dining on stibades of ivy, and the central rite took place at the prytaneion in the archaic agora. Later a new stop was introduced, a shrine at the Academy. The rite of xenismos involved the sacrifice of a tragos with singing of hymns, a processional dithyramb and other dithyrambs at the altar. The statue of Dionysos was then brought back to the sanctuary of Dionysos Eleuthereus. These rites reenacted the first (mythical) xenismos of Dionysos and the establishment of his cult. Hymns and dithyrambs concentrated on myths about Dionysos, Pegasos of Eleutherai, maybe Ikarios and Erigone. Therefore the chorus of male Athenian citizens who sang hymns at each xenismos were, above all, Athenians of the present singing in honor of Dionysos, and they were at the same time invested with the persona of the Athenians of the mythological past, who offended the god by not receiving his cult with honor, were punished, and then received him with honor.
Eventually the performance part of the xenismos acquired a dynamic of its own and became spectacle. First prototragedy, then tragedy emerged, and at some point performances became competitions. At the same time, the xenismos rite with its associated performances was moved to the Agora, near the Altar of the Twelve Gods, which was the alternative to the prytaneion as center of the polis. A skene was erected at the eschara by the Altar, and became part of the performances. Then performances associated with skene moved to the theater in the sanctuary of Dionysos Eleuthereus around 500 BC, with the result that the rite of xenismos and performances became separated.
The nexus of dithyrambs sung at the xenismos evolved into prototragedy in the form of a hypokrites speaking in propria persona and an unmasked chorus; then into a fully mimetic prototragedy with masks; finally into tragedy as basically we know it. Interaction of others, ritual schemata (e.g., sacred drama of Eleusis) and poetic models lead to the inclusion not only of the Athenians’ own past but also of stories such as that of Pentheus, Dionysos being still at the center of religious discourse. Then other religious themes were explored, as classical tragedy shows. The great dithyrambic poet Lasos of Hermione, who was active in Athens under the Peisistratids, may have played an important role in the development of prototragedy, which would explain the Doric elements in the tragic choruses.
The ritual matrix where tragedy originates involved a discourse of religious exploration, density of religious elements, and a form structured by hymns sung by a chorus which represented at the same time a collectivity in the frame of the hymn and a chorus of Athenian citizens singing in honor of Dionysos. The classical form of the tragedy, “parodos and stasima”, reflects an original schema “prosodion and stasima at the altar” and segments of rheseis by the hypokrites or exchanges between chorus and hypokrites.
It is impossible to give a full account of each step of the demonstration, as it is complex and fully detailed. Let’s say that when one follows the reconstruction of the Great Dionysia and its beginnings, one is fully convinced, even if one would welcome further explanation of the tragos or the origins of comedy.
III- Religion and the Fifth-Century Tragedians (pp. 201-511)
Religion and the Fifth-Century Tragedians is the longest chapter in the book: the main tragedies of the fifth century are scrutinized and analysed.
Aeschylean tragedies are in fundamental ways religious, because of their setting, the centrality of religious themes, the ritual acts that shape them, the density of religious language and religious references. Aeschylean tragedy is something different from ritual enactments of a central myth, such as the sacred drama of the Eleusinian Mysteries, but it can’t be kept totally apart from them by the audience. It has three functions: honor the god, entertain the audience (and win the competition), and explore the myth’s interstices and questions relevant to the audience’s reality. Beliefs about the gods worshipped by the polis are explored. Even Persai, which does not involve a myth, involves religious exploration: it presents the Persian defeat as a punishment of hybris, problematizes hybris and shows, through the impersonations of Darios and Xerxes, how difficult it can be to know when one has crossed the line into hybris. The basic schema structuring Suppliants, Persai and Septem is acknowledged as shaped by the ritual matrix independently reconstructed in chapter II. The Oresteia is a significant development away from the ritual matrix: there is greater stress on relationships between mortals and the canvas is much more complex, but Aeschylean tragedy is still “religious”.
A study of the tragic choruses in the fifth century, through Aeschylus, Phrynichos, Sophocles and Euripides leads SI to investigate the modalities of their appearance on stage and their relationship with the tragedy’s canvas and title. The chorus represents a collectivity that is a segment of a community, which seems to be an inheritance from prototragedy, but its uses shift in the course of the century, and its importance seems to fade in the plot. From a predominance of titles derived from the chorus in the early period, we move to a rarity of such titles in later fifth-century tragedies. However, the case of Choephoroi shows that the greater or lesser role of the chorus in the plot is not the only correlative inspiring the choice of title: the chorus consists of a group defined by its performance of a ritual role, which appears to be a significant category of fifth-century choruses.
SI’s analysis now shifts to Euripides and omits Sophocles, which she has studied elsewhere. Even when rituals play a minor part in Euripidean tragedies (such as in Medea), religious problematization is at least part of the tragic discourse, which shows that the basic ritual matrix that had generated tragedy has not yet fully disappeared. The evolution is not from a more religious to a less religious tragedy; on the contrary, Medea is an early play. SI shows convincingly that around 430 BC a tendency begins to intensify religious problematization. Through Hippolytos and Andromache, a deity’s hostility towards a mortal is explored; through Hecabe, the possibility that random tyche governs human affairs is shown to be wrong, and the world proves to be ordered and governed by the gods, even if human happiness must be cast aside. Following R. Parker’s analysis of the end of the fifth century as an “age of anxiety”, SI considers the plague at Athens, the mutilation of the Herms and the Profanation of the Mysteries as correlative — if not explanatory — of the evolution of Euripidean tragedy. Anaxagorean and sophistic elements in Euripidean tragedies defeat the challenge that Anaxagoras and some sophists issued to polis religion, because they help to acknowledge “the darkness and unknowability of the cosmos and the divine, but set it all in a wider, and ultimately reassuring, perspective” (p. 408).
The chapter ends with a study of divine appearance in Aeschylus, Sophocles and Euripides. In Aeschylus there is direct interaction between deities and mortals, whereas in the extant Euripidean tragedies there is a distant interaction: gods are unseen by the tragic character. Sophocles’ tragedies are in between these two poles. The Euripidean model is therefore the divine epiphany, something that comes directly from the religious experience of the audience (if not by direct knowledge). Divine epiphany on stage is therefore another link with sacred drama, even if the great distance between sacred drama and tragedy can’t be reduced. Divine epiphany is ambiguous, as it shows a god on stage and at the same time stresses the distance between god and mortal. In a way this could be compared with the otherness of gods shown by the chryselephantine statues of that time. SI concludes with the thought that Euripidean tragedy is pedagogical: as a group of individuals, the audience compares itself with the tragic characters, because they have the same limited understanding of their world. But at the same time, the audience has the same knowledge of the tragic world as the gods on stage: “tragic spectators (…) are guided to think in a broader perspective, shown how, almost trained to learn to think in that perspective. This can be seen as one of the ways in which the tragedian was a teacher to the polis” (p. 496).
A Summary of the Central Conclusion (pp. 513-518) sums up the major points of the demonstration.
This book is fundamental both for the history of Greek religion and for the history of tragedy, as well as for readings and studies of classical tragedy. I have generally been convinced by SI’s demonstrations, even if I didn’t fully agree with some very small parts of it (e. g. with the interpretation of Euripides’ Heracles, p. 365). The book is well produced, and I have only minor quibbles to deal with now. I noticed just one typo, p. 104 (“commmon”), and had some difficulties in locating the photographs of vases discussed in chapter II.1: the List of Photographs at the beginning of the book doesn’t give the page number, and they are inconveniently located after chapter III.2, between pp. 290 and 291, without any page numbers. In the two Appendices pp. 410-422, the titles of tragedies are abbreviated, whereas they are written in their full form in the rest of the book (e. g. IT for Iphigeneia in Tauris pp. 412 and 415). Finally, Greek texts, even when they are not included in the main English text, are transliterated, without accents (pp. 221, 235, 296, 309, etc.). It is a pity not to follow the way in which the first Greek text is written page 96, with Greek alphabet and accents. A translation of those long quotations would also be welcome for those readers who don’t understand Greek at first sight.
Should this book have been shorter? Surely not, even if sometimes some explanations may appear tedious or longwinded (for example pp. 70-99, and above all pp. 76-78 where I got completely lost). The fact is that SI’s demonstrations are usually based on a perfect logical frame: she warns her readers that “circularity” and “cultural determinism” must by all means by avoided, she succeeds in avoiding them perfectly, and she repeatedly and unnecessarily points out that she is succeeding in avoiding them. Moreover, SI goes carefully into every single detail that would endanger her argument, to make her demonstrations irreproachable. Therefore the reader is on sure ground, but he/she is at the same time drowning in a sea of information, dissenting voices and refutations that compel him/her to read the demonstration with a pen and a piece of paper to follow the main line of the argument. Some of these parts might have been relegated to the end of the chapters, as SI did indeed with a study on Lenaia and Lenaion (pp. 120-123), a brief critique of readings of Euripides’ Orestes (Appendix 1, pp. 410-413) and of Dunn’s theories about Euripidean strategies of closure (Appendix 2, pp. 414-422). Finally, as SI is constantly aware of her own demonstration, of its logic and its reception, she introduces each one of her chapters with a programmatic introduction, spreads short résumés and temporary conclusions throughout the text, and concludes with a new summary. This is nice and helps the reader, insofar as he/she may go directly to these short pieces of exposition. However, the minute repetition of formulas, sentences and advice make uninterrupted reading disagreeable. The feast for the mind turns out to be occasionally a nightmare for the stomach.