Book Search:  


Google full text of our books:


Roman Eyes:
Visuality and Subjectivity in Art and Text
Jas Elsner

Book Description | Endorsements | Table of Contents

COPYRIGHT NOTICE: Published by Princeton University Press and copyrighted, © 2007, by Princeton University Press. All rights reserved. No part of this book may be reproduced in any form by any electronic or mechanical means (including photocopying, recording, or information storage and retrieval) without permission in writing from the publisher, except for reading and browsing via the World Wide Web. Users are not permitted to mount this file on any network servers. Follow links for Class Use and other Permissions. For more information, send e-mail to

This file is also available in Adobe Acrobat PDF format

Chapter 1


Visuality in the Greco-Roman World

A principal argument supporting the assertion of a great divide between the arts of Classical antiquity and the Middle Ages has been an assumption about naturalism. Classical art, we have been told, is the supreme precursor of the Renaissance—not only in its search for illusionistic forms and in its celebration of the artists who led the way in creating such forms1 but also in the kinds of visuality associated with naturalistic verisimilitude. Even in the “coldly classicizing academic” copies of the Roman imperial period,2 sophisticated viewers, like the essayist Lucian or the rhetorician and historian Philostratus, were able to indulge the most complex and elegant wish-fulfillment fantasies in front of naturalistically rendered objects. The power of naturalism encouraged (and still does encourage) the imagination to believe that the visual world of a painting or sculpture is just like our world, even an extension of it. This kind of Classical visuality—leading ultimately to fantasies of (and apparently, according to our sources, even attempts at) sexual intercourse with statues so perfectly beautiful as to be better than the real thing— anticipates the frisson of Renaissance masterpieces from Michelangelo’s David to Titian’s Venus of Urbino.3 The superlative naturalism of the image—its artifice so brilliant as to disguise the fact that it is merely art, as Ovid puts it4—prompts the willing viewer to suspend his or her disbelief that the image is more than pigment or stone. Entrapped, like Narcissus, in the enchanting waters of desire and illusion, the viewer identifies with, objectifies, and may even be seen by the image into which the imagination has poured so much aspiration.5

Writing on art within the Roman empire shows extraordinary self-awareness of the problematics of visuality in relation to naturalism. Just as Narcissus sees himself reflected in the pool and is deceived into a fatal love, so we who look at his image in a painting (and at his image in the pool within the painting) are ourselves putting a toe into the dangerous waters of his visual desire. In the Elder Philostratus’ scintillating account of a painting of Narcissus, the realism is so vibrant that the writer (and his audience) cannot tell whether a “real bee has been deceived by the painted flowers or whether we are to be deceived into thinking that a painted bee is real.”6 In one sense this is a literary topos of the sort which occurs in Pliny the Elder’s chapters on art history,7 but at the same time this very dilemma (our dilemma as viewers) is a version of the fatal delusion of Narcissus himself.8 Philostratus, in his description of a painting showing huntsmen, with superior psychological insight sees the pursuit of a boar (the painting’s ostensible subject) as a sublimation of its real theme, the hunters’ pursuit of a pretty boy whom they seek simultaneously to impress by their exertions and to touch physically (1.28.1). Yet at the moment the writer discovers the image’s deeper meaning as a presentation of desire, he draws back, seeing his own desire as interpreter thwarted by the fact that naturalism is not nature, that what is realistically realized may not necessarily be real:

How I have been deceived! I was deluded by the painting into thinking that the figures were not painted but were real beings, moving and loving—at any rate I shout at them as though they could hear and I imagine I hear some response— and you [that is, Philostratus’ listeners or readers] did not utter a single word to turn me back from my mistake, being as much overcome as I was and unable to free yourself from the deception and stupefaction induced by it.9

Yet this kind of sophistication, and the concomitant fascination with the sheer artistry of art—the anecdotes of famous painters, the exquisite skillfulness of technique, the works which deceived even animals and birds—are only part of the story. For if antiquity was the ancestor of the Renaissance, it was also the mother of the Middle Ages. Alongside wish-fulfillment fantasies in the aesthetic sphere of the art gallery10 went a culture of sacred images and ritual-centered viewing, in which art served within a religious sphere of experience strikingly similar to the world of icons, relics, and miracles of medieval and Byzantine piety. I will briefly sketch Roman art’s “Renaissance” visuality, and then explore the “medieval” visuality of its oracular, liturgical, and epiphanic experience of images. My question is in part how these apparently exclusive worlds could be reconciled. My answer will be that, to some extent at least, in looking at a culture that is not just foreign but also ancestral to us our own expectations and interpretations have distorted the ancient evidence and material to suit our own desires and preconceptions. The predominant trends of ancient visuality, I suggest, were stranger and less familiar than is usually supposed when we subsume the arts of antiquity into a discourse inflected by the assumptions of Renaissance naturalism.

Visualities of Naturalism

The extent of antiquity’s “Renaissance” visuality can be indicated by a quick comparison of some images and texts. Let us begin with two visual realizations of the mythological tale of Perseus and Andromeda. The first is a wall painting excavated from the villa rustica at Boscotrecase near Pompeii in the first decade of the twentieth century. It dates from about 10 b.c. and is now in the Metropolitan Museum in New York (figure 1.1).11 The second is a sculpted relief panel now in the Capitoline Museum in Rome and earlier in the Villa Doria Pamphilj and Albani collections. It dates from the mid-second century a.d., probably from the reign of Hadrian (figure 1.2).12

Beside the fresco, place the following description of a painting from the great second-century novel Leucippe and Clitophon, written by Achilles Tatius:

The girl was placed in a recess of the rock which was just her size. It seemed to suggest that this was not a man-made but a natural hollow, a concavity drawn by the artist in rough, irregular folds, just as the earth produced it. Looking more closely at her installed in her shelter, you might surmise from her beauty that she was a new and unusual icon, but the sight of her chains and the approaching monster would rather call to mind an improvised grave.
   There is a curious blend of beauty and terror on her face: fear appears on her cheeks, but a bloomlike beauty rests in her eyes. Her cheeks are not quite perfectly pale, but brushed with a light red wash; nor is the flowering quality of her eyes untouched by care—they seem like violets in the earliest stage of wilting. The artist enhanced her beauty with this touch of lovely fear.
   Her arms were spread against the rock, bound above her head by a manacle bolted in the stone. Her hands hung loose at the wrist like clusters of grapes. The color of her arms shaded from pure white to livid and her fingers looked dead. She was chained up waiting for death, wearing a wedding garment, adorned as a bride for Hades. Her robe reached the ground—the whitest of robes, delicately woven, like spider-web more than sheep’s wool, or the airy threads that Indian women draw from the trees and weave into silk. . . .
   Between the monster and the girl, Perseus was drawn descending from the air, in the direction of the beast. He was entirely naked but for a cloak thrown over his shoulders and winged sandals on his feet. A felt cap covered his head, representing Hades’ helmet of invisibility. In his left hand he held the Gorgon’s head, wielding it like a shield . . . his right hand was armed with a twin-bladed implement, a scythe and sword in one. The single hilt contains a blade that divides halfway along its extent—one part narrows to a straight tip, the other is curved; the one element begins and ends as a sword, the other is bent into a sinister sickle, so a single manoeuvre can produce both a deadly lunge and a lethal slash. This was Andromeda’s drama.13

However a modern spectator might refrain from (admitting to) such an excessive response to a work of art, for Achilles Tatius a painting perhaps somewhat like the Boscotrecase mural was the occasion for indulging his readers in an intense sexual fantasy. The maiden, ravishing in her “curious blend of beauty and terror,” is exposed to be ravished by the viewer’s (as well as the reader’s and the monster’s) gaze—tied up, powerless, in a posture worthy of Ingres’ spectacularly voyeuristic painting of Roger and Angelica.14 The writer dwells on voyeurism, virtually caressing the young woman’s “lovely fear,” playing with descriptive pseudo–art criticism (“drawn by the artist in rough, irregular folds,” “a light red wash,” “the colour of her arms shaded from pure white to livid”) and with suggestive similes and metaphors (Andromeda’s “flowering” eyes, the “violets in the earliest stage of wilting,” the hands splayed from the wrists “like clusters of grapes”). To the brutal penetration of the male gaze—equally that of the writer, readers, and viewers as well as of both Perseus and the sea monster within the picture—the “airy threads” of her wedding garments reveal more than they disguise. It is no surprise that this passage of hypersexualized male objectification, a voyeur’s anticipation of the violence of rape, climaxes on the sword’s phallic conquest of the sea monster, which is as much a hint of the hero’s future domination of the lady as it is a description of his valorous feat.15 “A deadly lunge and a lethal slash” would indeed be “Andromeda’s drama.” She is “spread out against the rock” as an erotic vision to satisfy and excite the viewer of the picture and the reader of the text as well as the viewers (Perseus and the monster) within the image. The strategies of description are enticing us to identify with the twin-bladed hero in anticipation of both his conquests, the monster and the girl.

Turning back to the Pompeian fresco from the erotic intensity of Achilles’ description, one might be forgiven for wondering at the extent of the novelist’s rhetorical “reading in.” The painting broadly represents a version of the iconography Achilles expected his readers to bring to mind, but its narrative takes place in the distant spaciousness of landscape, while Achilles’ story is all about impassioned identification with characters whose emotions and deeds loom larger than the everyday.16 Yet “reading in” is a key aspect of the rhetorical nature of ekphrasis, the literary device of describing people, situations, or works of art in such a way as to bring them vividly to mind in the reader’s or listener’s mind’s eye, as well as being an essential invitation of the visuality of naturalism.17 As we are enticed by a picture to tell its story, which is always, to some extent, our story or at least a story plausible to us, so we identify with, allegorize, and fantasize about the image, thereby transforming its content into a narrative which suits us. Achilles’ description of Andromeda certainly suits the voyeurism, violence, and sexuality of his novel, even if someone else might offer a very different (and differently flavored) account from looking at the Boscotrecase fresco. Interestingly, the erotic focus of the Boscotrecase fresco is significantly stronger than that of a somewhat later painting deriving from the same basic model from the Casa di Sacerdos Amandus at Pompeii (I.7.7, figure 1.3).18 The latter replaces the still potentially erotic maiden caressed by wispy draperies with a fully dressed Roman matron going through the motions of Andromeda’s drama.

The Capitoline panel (figure 1.2) enters the myth from another point in the story, as does the painting which Philostratus describes in his Imagines (1.29). In the Capitoline panel, the contest is over and the maiden has been rescued. The triangular complex of the Boscotrecase painting, where man and monster fight over a woman chained and passive at the picture’s center, is a world away from the atmosphere of relief after a crisis as boy and girl gaze into each other’s eyes. Yet this immediacy of erotic entanglement (something any viewer can instantly identify with) is set against a certain formal academicism, as the nude hero with his elegantly draped cloak is adapted from a late classical type of Hermes that seems derived from the work of Praxiteles, while Andromeda in her swirling draperies is a neo-Attic dancing girl, related to a lost fifth-century b.c. prototype of a dancing Maenad that was popular with Roman copyists.19 As with the two frescoes which belonged in rooms with other pictures on amatory themes, the panel’s meanings would be affected if it were part of a group or a program, like the so-called Spada reliefs of roughly the same date.20

Here is how Philostratus, writing in the early third century a.d.,21 describes a painting in some ways similar to the Capitoline relief:

The contest is already finished and the monster lies stretched out on the strand, weltering in streams of blood—the reason the sea is red. Eros frees Andromeda from her bonds. He is painted with wings as usual, but here— unusually—he is a young man, panting and still showing the effects of his toil; for before the deed Perseus put up a prayer to Eros that he should come and swoop down with him upon the beast, and Eros came for he heard the Greek’s prayer. The maiden is charming in that she is fair of skin though in Ethiopia, and charming is the very beauty of her form; she would surpass a Lydian girl in daintiness, an Attic girl in stateliness, a Spartan in sturdiness. Her beauty is enhanced by the circumstances of the moment; for she seems to be incredulous, her joy is mingled with fear, and as she gazes at Perseus she begins to send a smile towards him. . . . He lifts his chest, filled with breath through panting, and keeps his gaze upon the maiden. . . . Beautiful as he is and ruddy of face, his bloom has been enhanced by his toil and his veins are swollen, as is wont to happen when the breath comes quickly. Much gratitude also does he win from the maiden.22

Although Philostratus here shares Achilles Tatius’ trope of beauty mixed with fear, his subject is not, however, the violence of anticipated dismemberment and sexual satisfaction, but the coy theme of boy and girl falling in love. Even the slaying of the monster is presented as an appeal to Eros. Eros’ descriptive presence is a reflection of Perseus; he appears as the very double of the hero, a young man (rather than a putto) panting and sweating from the toil of battle, swooping down with Perseus in accomplishing the deed. Instead of a thinly disguised metaphor of the sex act, the battle is the preliminary of seduction; instead of a narrative of rape, we are offered the intimations of foreplay. The description of the girl is a superlative literary transformation of the “academic classicism” of the painting’s style, with its clear indebtedness to earlier models (like the Maenad of the Capitoline relief ). Philostratus turns this into a set of comparisons by which Andromeda surpasses the girls of Ethiopia, Lydia, Attica, and Sparta (whose forms in earlier Classical art her painter may have borrowed). Like Achilles’ description, all this is a “take” on the myth—a “reading in” to the picture of a series of cliché expectations of what happens when boys and girls are thrown together in unusual circumstances, when their gazes meet.

The viewer implied by Philostratus, like the viewer of the Capitoline relief, is offered the sight of lovers transfixed. It is ambiguous whether the panting of the hero’s chest is due to his exertions in slaying the monster or to his anticipation of getting the girl. The Capitoline panel, though different in some aspects of its iconography (the absence of Eros, the fact that Philostratus’ Perseus is lying on the grass), offers ample potential for this kind of voyeuristic viewing. On the relief, Andromeda’s otherwise profuse draperies cling fortuitously, virtually see-through (“like spider-web”) around her breasts and thighs. Her erotic nudity is displayed full frontal (to Perseus and to the viewer) through this apology for clothing, while he stands nude before her, the sea monster quelled beneath them. The maiden’s eyes are cast modestly down, looking with apprehension at the monstrous fate from which she has just been saved, or is she glancing upward to transfix her future lover’s gaze, as in Philostratus? Does “she [begin] to send a smile towards him”? What are the rewards for the “gratitude” he wins “from the maiden”?

The naturalism of this kind of art, not just in its forms but in its imitation (and intimations) of “real-life” desires and fantasies on an idealizing mythological level, brings a confident assertion of a particular visuality. It is the form of ancient Greco-Roman visuality most familiar to us, since it is precisely the kind of viewing which post-Renaissance art and art history have identified with and practiced. This is our version of “reading in,” signaled at crucial junctures of the modern art historical enterprise by the great ekphraseis (themselves deeply influenced by antiquity) which punctuate the work of the founders of the discipline, especially Vasari and Winckelmann.23 But it is my claim that this visuality—of identification, objectification, ultimately erotic desire—is only one part of antiquity’s armory of the visual.

Religion and the Primacy of Ritual

Beside the visual culture of the art gallery, with its hyper-realized celebration of naturalistic visuality in rhetorical set-pieces and erotic fiction, antiquity offered a world of sacred images. It was, for instance, not always possible to differentiate the deity from his or her statue.24 In the Greek language this gives rise to the interesting ambiguity that, for example, “Artemis” can mean equally the goddess herself or an image of her.25 In the handling of images, this ambiguity afforded an edge of danger, since incorrect treatment of a statue could be construed as an assault on the deity embodied in it.26 In the context of a temple, the statues adorning its sacred enclosure, including the cult image inside the temple, were themselves part of a broad culture of ritual (similar in many respects to the cultivation of icons in Byzantium and medieval Italy).27 Statues might be dressed, paraded, washed, fed, and worshipped;28 they were imagined to have volition and magical power (oracular, talismanic, healing, or malevolent);29 and the more important statues were even capable of intervening in legal problems or power politics—by granting sanctuary to fugitives, for instance.30

These functions were all highly varied and nuanced by the particular local conditions, myths, and traditions within which any one image might be worshipped. Instead of a broad survey, however, I shall concentrate on the issue of visuality by focusing initially on some remarkable texts. First, I will use Pausanias’ Description of Greece, the richest account of religious images and their myths and cultivation surviving from antiquity, to outline some aspects of ritual-centeredness. A native Greek-speaker, Pausanias was born in Lydia in Asia Minor, and seems to have traveled for several years between about a.d. 135 and 180 throughout mainland Greece (the Roman province of Achaea), observing rituals, local myths, and works of art in remarkably painstaking and careful detail and writing as he went.31 His text is a fascinating combination of what we would call antiquarianism as well as genuine art historical (even connoisseurial) expertise with a pilgrim’s precise interest in the religious nature and sacred details of myths, rituals, and statues.32 I will then turn to a discussion of cult images in Lucian’s brief pilgrimage narrative De Dea Syria, in order to explore the visuality of the cult image in its ancient ritual setting. Writing in Greek at about the same time as Pausanias, the Syrian-born Lucian is antiquity’s finest surviving satirical essayist.33 His De Dea Syria, which describes with little apparent irony the process of pilgrimage to the shrine of Atargatis in Hierapolis in Roman Syria, was thought by many in the past to be spurious, but is now accepted as genuine.34

Of course one might impute the differences I am suggesting between the texts in the previous section and the following ones as amounting to differences of narrativity and genre rather than differences in visuality. And it is of course the case that fiction and the traditions of ekphrasis engage rhetorical and narrative techniques that are different from those of travel literature or pilgrimage tracts. But I would argue that the genre chosen for the description of art presupposes a kind of visuality within which images are to be received. Whether a particular form of visuality is one of the reasons a writer chooses a particular narrative style, or whether the form of visuality may be said to arise as a result of the literary genre chosen, matters less (for my purposes here) than the claim that more than one kind of visuality existed in antiquity. Here the case of Lucian is significant, since at different times he chose to write in all the genres touched upon here. He was a master of rhetorical ekphrasis (for instance in De Domo or Zeuxis), of outrageous fiction (especially the Verae Historiae), and of religious polemic (Alexander, Peregrinus), but in the De Dea Syria he deliberately chose a different kind of genre—that of the pilgrim’s travel book (which many have found uncomfortably sincere within his corpus of writing)—in order to express a form of piety and religious visuality not possible in his more usual satirical style.

The Ritual Setting

I concentrate first on one of the two sites which Pausanias claims to be the most special in all Greece. At 5.10.1, he announces, “Many are the sights to be seen in Greece, and many the wonders to be heard; but on nothing does heaven bestow more care than on the Eleusinian rites and the Olympic games.”35 Within the Altis, the “sacred grove of Zeus” (5.10.1) which was Olympia’s holiest spot (figure 1.4), Pausanias describes in accurate detail the two great temples of Zeus (5.10.1–11.11) and Hera (5.16.1–20.6) with their cult images, decorations, and offerings.36 Sandwiched between them is a remarkable description of the altars within the Altis, on which I wish to focus (5.13.8–15.12).37 Although ancient altars are not figural images, they are nonetheless handcrafted works of material culture (nonanthropomorphic for the most part and perhaps aniconic) and the tenor of Pausanias’ enumeration of them is revealing. It shows a ritual-sensitive visuality in which the pilgrim-viewer submits to the liturgical rule book of a holy site in order to be offered its sacred experience. It demonstrates a deeply focused interest in the detailed precision of ritual, in which the topographical orchestration of a prime sacred center is rendered, and is experienced by writer and readers alike, both as a temporal process of liturgical action and as a spatial progress through a series of monuments whose order and meaning are dependent on their ritual relations with one another. It is unfortunate that none of these altars have survived, so that all we have to work with now is Pausanias’ own text. However, while many of his projected readers would have been to Olympia (which was, after all, at the center of pilgrimage, tourist, and athletic routes in Greece) and seen the altars, some may not have visited the site—and for these, at least, Pausanias’ account would have constituted a principal mode of access to perhaps the major panhellenic sanctuary in the Greek-speaking world.38

Twice Pausanias announces that he is describing the altars in the order used by the locals in sacrifice. First he tells us what he will do, saying, “Let me proceed to describe all the altars in Olympia. My narrative will follow the order in which the Eleans are accustomed to sacrifice on the altars” (5.14.4). Later on in his account of the altars, he reminds us of his chosen descriptive strategy, when he notes, “The reader must remember that the altars have not been enumerated in the order in which they stand, but the order followed by my discussion is that followed by the Eleans in their sacrifices” (5.14.10). Even as he opens this part of his account, Pausanias insists on a ritual-centered dispensation for viewing these monuments in the sanctuary. He is aware of making an odd descriptive choice, but makes it anyway, since to him (in the case of objects whose prime purpose is sacrificial) it is a more natural evocation of what he wants to describe. What matters to him, in effect, is not a topography of geographical accuracy, a map of juncture and position, but a topography of ritual correctness, in which the temporal unfolding of a series of sacred actions becomes the dominant frame for his account. It is significant that, in an author all too often unfairly impugned for lack of literary or intellectual sophistication,39 the writing (and the reading) of this innovatively organized description is itself a ritualized reiteration of the liturgical process. In reading Pausanias on the altars, we move from altar to altar in a vicarious reenactment of Pausanias’ own participation in an ancient ritual activity which was repeated, so we are told, “every month.”40

The account of the altars opens with that of Olympic Zeus, to which Pausanias devotes by far the largest space.41 Interestingly, the focus on ritual informs even the altar’s material construction: “It has been made from the ash of the thighs of the victims sacrificed to Zeus” (5.13.8). This peculiarity, for which Pausanias finds parallels at Pergamon and Samos, prepares us for a discussion of minute ritual details (5.13.10–11). Pausanias tells us precisely how the altar is designed—with its lower level, or prothysis, used for the killing of sacrificial animal victims and its upper level used as the site of the burning of the thighs. Again, what is effectively a discussion of the altar’s form (after that of its material) is realized in terms of its ritual functions. The differences between the two levels extend beyond the materials of manufacture (the steps to the prothysis are stone, those to the upper part “are, like the altar itself, composed of ashes”) to issues of access: Pausanias is careful to tell us that on days when they are not excluded altogether from Olympia, women, both virgins and married women, can only ascend to the prothysis while men can go to the higher level.

After having elaborated the structure of the monument in these entirely ritual-centered terms, Pausanias informs us that sacrifice is offered daily both by individuals and by the Eleans as a whole. Moreover, once a year on a specified day (the 19th of the month of Elaphius), ash is mixed with water from the river Alpheius, and no other river, to make a paste with which the altar is daubed. These rituals prepare the way for discussing the specific sacrificial action for which the altar is the setting. Pausanias tells us that the Eleans always use “wood of the white poplar and of no other tree” for sacrifices to Zeus, which he attributes to the fact that “when Heracles sacrificed to Zeus at Olympia, he himself burned the thigh bones of the victims upon wood of the white poplar.”42

What we are offered here, and I would contend this is typical of Pausanias and can be paralleled frequently in his Description of Greece,43 is a striking instance of ritual-centered visuality. There is great empirical precision in the observations, but their descriptive force is evocative not of how the monument looks or where it is so much as of how it works. Everything about the altar in Pausanias’ vision relates to the nature of its rituals, which are themselves part of the sacred definition of Olympia as one of ancient polytheism’s most important religious centers. The altar’s meaning in this context is not merely archeological or aesthetic; it is above all a continuing site for the execution of traditional religion. In the context of Roman-dominated Greece, such an assertion of an ancient (that is, pre-Roman) piety, which was constantly affirmed by reference to mythological, myth-historical, and divine figures who functioned as founders of rites and guarantors of their validity, had political and cultural undertones. Such assertions constituted an affirmation of identity, with at least some flavor of resistance to the appropriations of the imperial center.44

This effect of reperforming traditional religion is specially marked in the enumeration of the sixty-nine altars in the order of the Elean ritual procession. In part this serial presentation is a way of celebrating the congruence of deities within the Altis—a celebration enacted as much by the monthly ritual on which Pausanias bases his narrative as by Pausanias’ description. For example: “They sacrifice to Hestia first, secondly to Olympic Zeus, going to the altar within the temple, thirdly to Zeus Laoetas and to Poseidon Laoetas. This sacrifice too it is usual to offer on one altar. Fourthly and fifthly they sacrifice to Artemis and to Athena, Goddess of Booty, sixthly to the Worker Goddess.”45 Through the sacrificial process, and equally through the logic of reciting the deities’ names in the text, the presence of the gods (of each of the sixty-nine altars) is affirmed. As ever, Pausanias is careful to note when an altar is shared by more than one deity, when the recipient of an altar’s sacrifice is in doubt or disputed, and who was the altar’s dedicator (if this can be found out).46

The liturgical order of enumeration requires Pausanias to take what has been described as a “very leisurely and erratic course” in which he more than once retraces his steps and tells us so.47 For instance, having opened with the altar of Olympian Zeus before beginning the liturgical enumeration, Pausanias returns to it first within the Eleans’ monthly procession at 5.14.8, again when he comes to the altar of Zeus Descender (“this altar is near the great altar made of the ashes,” 5.14.10), and finally at the close of his enumeration of the altars (5.15.9). Indeed, it is at the moment of return that he reminds us of the processional order along which he has structured his description (5.14.10). In effect, the text at this moment demonstrates the tension between a topographically straightforward narrative and the one upon which its author has embarked. This tension is itself evidence both of the difficulty of rendering ritual as description and of Pausanias’ deliberate choice of ritual as his preferred frame for viewing.

At the end of the description of the altars, as a kind of summing up, Pausanias recounts the nature of the sacrificial action as he did in his account of the great altar: “Each month the Eleans sacrifice once on all the altars I have enumerated. They sacrifice in an ancient manner; for they burn incense with wheat which has been kneaded with honey, placing also on the altars twigs of olive, and using wine for a libation. Only to the Nymphs and the Mistresses, it is not their custom to pour wine in libation, nor do they pour it on the altar common to all the gods” (5.15.10). The precise details are again important— that it be wheat (rather than, say, barley) kneaded with honey, that the twigs be olive, that wine be used in the correct libations and for the right deities. These careful enumerations of small details, coupled with the broader sense of a ritual process, bring the gods of the Altis alive through their living sacrifical association with their altars. From our point of view, they provide invaluable contextual and ethnographic richness to understanding the culture of ancient paganism’s sacred visuality.

Before we ask how the sacred images themselves speak to their worshippers, it is worth noting that Pausanias’ evidence, exceptionally rich, full, and detailed though it is, is not unique. Other ancient authors corroborate the ritual-centeredness of Pausanian visuality in interesting ways. The Sacred Tales of the distinguished second-century orator Aelius Aristides evoke a pattern of repeated pilgrimage to healing temples (especially those of Asclepius) in Greece and Asia Minor. There a variety of medicinal rituals (bathing, abstinence from bathing, fasting, vomiting, enemas) are enjoined upon the devotee by his god, who appears in dreams and who also instructs Aristides to write it all down. At one point the orator comments: “What should one say of the matter of not bathing? I have not bathed for five consecutive years and some months besides, unless, of course, in winter time, he ordered me to use the sea or rivers or wells. The purgation of my upper intestinal tract has taken place in the same way for nearly two years and two months in succession, together with enemas and phlebotomies, as many as no one has ever counted, and that with little nourishment and that forced.”48

Here we have a ritual culture mapped not around the liturgy of a sanctuary but about the workings (internal and external) of the body itself. Doubtless Aristides’ ability to see his god—not just in statues but beyond them in the dream-visions which his text repeatedly affirms and in which his spiritual path of healing was incrementally enjoined upon him—could not be separated from the ascetic effects of his ritual activities. This kind of personal preparation for a divine vision or for healing is attested in Lucian’s De Dea Syria, where pilgrims shave their heads and eyebrows, use only cold water for drinking and bathing, and always sleep on the ground until their sacred journey is completed (section 55). Pausanias himself, as well as the Delphic priest Plutarch, testifies to similar personal rituals, though with considerably less autobiographical color than Aelius Aristides.49

Beside this body-oriented focus, in which the ritual context seems intensely personal, texts like the De Dea Syria or Philostratus’ hagiography of the pagan holy man Apollonius of Tyana support what might be termed the more broadly sociological context of visuality within a regularly repeated festival full of people.50 The last part of the De Dea Syria presents a prime pilgrimage center in the east in full ritual and festival action (sections 42–60). The climax of the Life of Apollonius has Philostratus’ sage (lauded as himself an object of pilgrimage in Olympia) perform a Socratic dialogue on the nature of a festival (8.15, 18).51 The monthly Elean sacrificial liturgy and the ash-daubing on the 19th of Elaphius are examples of such public holidays. These form some of the proudest set-pieces in Pausanias’ account.52

Viewing the God

A limestone relief of the second or third century a.d. from the temple of Atargatis at Dura Europos in Syria, now in the Yale University Art Gallery, depicts the two principal gods of the city of Hierapolis, Atargatis and Hadad, enthroned in divine splendor (figure 1.5).53 In his De Dea Syria (sections 31–32), Lucian translates the identities of these deities for the convenience of his Greek readership as Hera (for Atargatis) and Zeus (for Hadad).54 Yet he describes them precisely. Of the two, Atargatis was the more important, as is shown by her larger size, her larger throne, and her larger footstool in the relief. This is corroborated by Lucian, who devotes the bulk of his description to Hera (section 32) and who regards the whole sanctuary as being under her protection (sections 1 and 16). The gods sit together in the temple chamber, Hera enthroned on lions and Zeus on bulls. Both are crowned (Lucian mentions the “tower” on Hera’s head, as well as her girdle), and, at least from the evidence of the written description, both were adorned with costly gems. Between them on the relief is a rather strange object which resembles a Roman military standard.55 Lucian mentions this too: “Between the two statues stands another golden image, not at all like the other statues.”56

If we ask how such icons were viewed, at least in the context of a celebratory pilgrimage text like the De Dea Syria,57 Lucian provides evidence which augments his empirical description. Unlike Zeus who, according to Lucian, “certainly looks like Zeus in every respect” (section 31), Hera “also has something of Athena, Aphrodite, Selene, Rhea, Artemis, Nemesis and the Fates” (section 32). In effect, she encompasses all the major female deities in the Greco-Roman pantheon. Her special nature is marked out by her precious adornments and gems brought by devotees from all over the known world (Lucian mentions pilgrims who include Greeks, Egyptians, Indians, Ethiopians, Medes, Armenians, Babylonians, Phoenicians, Cappadocians, Cilicians, Arabs, and Assyrians, sections 10 and 32). Her power is especially manifested by her gaze: “There is another wondrous feature in the statue. If you stand opposite and look directly at it, it looks back at you and as you move its glance follows. If someone else looks at it from another side, it does the same things for him” (section 32).

Here the divine being of the goddess is presented on multiple levels: through the value and splendor of her ornaments, through her occupation of a sociological position in the center of a whole sacred world looking both east (to the Parthian lands not controlled by the Roman empire) and west (to Lucian’s Greek-speaking readers), and above all through her power to hold the gaze of the individual worshipper. Her power encompasses the devotee’s body, looking back at the viewer and watching him or her; and it extends beyond the individual to encompass all worshippers collectively, since she can presumably hold and follow the gazes of all the pilgrims who look at her. Indeed, through the miraculous ruby on her forehead (from which “a great light shines by night and the whole temple is illumined by it as if by lamps,” section 32), she herself generates the light by which she may be seen.

This passage affords a rare glimpse of the effect of a great cult deity in antiquity. She is the center-piece and ultimate cult object both of personal rituals (head shaving, sacrifices, the dedication of locks of hair, sections 55–60) and of large public festivals. The latter include processions of all the deities to the temple’s great lake (sections 45–47), the festival observed by the sea (in which Lucian did not himself participate, section 48), the fire or lamp festival (section 49), and the festival in which the temple’s castrate priests (galli) perform various acts of self-mortification, of which the ultimate is the self-castration of those destined by the act to become future galli (sections 50–53). It is as if her gaze encompasses all these activities—from mass pilgrimage to the most intensely personal acts of physical asceticism, castration being the most extreme example.

Returning to the relief, one may argue that the sculpture—by emphasizing the paired enthronement of Atargatis-Hera and Hadad-Zeus—is less emphatic than Lucian in proclaiming the centrality of Hera’s power. To a significant extent, the supremacy of Hera was established by liturgical and ritual custom rather than just iconography. For instance, in the procession to the lake at Hierapolis, Lucian says that the image of Hera always went first (section 47). The difference here between the implications of what Lucian says and what one might infer directly from the Dura relief indicates not so much the primacy of text over image as the fact that images on their own represent only a small part of the visual culture of their use and function, which in this case a text helps to supply.

What the iconography of the Dura relief does suggest, however, is that it was frontality which Lucian had in mind in his account of the encompassing power of the goddess’s gaze. Certainly this is what distinguishes not only Atargatis but many of antiquity’s cult statues from the naturalism of Classical mimesis. Comparison with a mythological relief like that of Perseus and Andromeda (figure 1.2), where the figures are wrapped up in their own world, oblivious to the viewer who intrudes on their incipient love affair like a voyeur, may seem forced, since one can argue that it is hardly a pairing of like with like; but no naturalistic statues offer the arresting frontal gaze of such cult icons. Indeed, one definition of naturalism is precisely the avoidance of the kind of gaze that encompasses all who worship at Hierapolis.58 Where Atargatis and Hadad demand that worshippers be incorporated in the gods’ sacred world by eyeballing those that approach into submission,59 the typical images of naturalism look away, involved in their own worlds, their own narratives, and their own realities. Naturalistic images elicit a series of identifications, objectifications, and narratives from us as viewers to read our way into the picture, as it were. Philostratus and Achilles Tatius provide excellent models of how that might be done. Here, in the sacred visuality of Hierapolis, itself the product of a complex ritual culture, viewing is much more direct.

Visuality and the Sacred

The confrontation with the direct gaze of the deity, a kind of gaze which is one of the most striking formal elements of medieval icons (see, for example, the St. Peter icon from Sinai and especially the medallions at the top, figure 3.1) as well as pre-Christian cult statues, is a key aspect of ancient ritual-centered visuality, at least as presented by Lucian. The viewer has prepared for his or her epiphany with the god through a series of ritual acts, whether they be physical mortifications and abstinences (like Aristides’ purges and enemas or the head shaving of Hierapolis), liturgical processes (like Pausanias’ Elean procession), or the ordeals of a long pilgrim’s journey.60 Viewer-pilgrims, already taken out of their normal social realities through rituals which affect on the external level the body itself and on a more interior level the individual’s sense of subjectivity, bring their identities to the house of the god.61 This house is already an especially holy place. As Lucian says, in relation to the temple at Hierapolis, “no other temple could be more sacred, nor any other region more holy. . . . The gods [here] are especially manifest to the inhabitants. For statues among them sweat and move about and give oracles, and a shouting often occurs in the temple when the sanctuary is locked, and many have heard it” (section 10). Within the temple, at the culmination of the journey (from a pilgrim’s personal point of view) and at the pivotal center of the site (from the viewpoint of both liturgical action and sacred geography), the viewer confronts the god.

It is important that the vision of the god be seen as a culmination of a ritual process. The texts we have been using are complex documents, with political, ideological, and literary purposes as well as religious ones. But it is in their most strongly defined aspects as insider texts, written by religious devotees, that their evidence for ritual-centered visuality (always an initiate’s, and never a skeptic’s, way of viewing) can be assessed. Both authors are explicit about being religious insiders: Pausanias particularly in his reticence to give away initiate secrets,62 and Lucian (very elegantly) in the last sentence of the De Dea Syria, where, having discussed a ritual of hair dedication in the temple, he comments, “When I was still a youth I, too, performed this ceremony and even now my locks and name are in the sanctuary” (section 60).

Viewing the sacred is a process of divesting the spectator of all the social and discursive elements which distinguish his or her subjectivity from that of the god into whose space the viewer will come.63 In the reciprocal gaze of divine confrontation, there is a form of visuality in which the image does not just look back at the viewer, but in which the viewer has specifically made the journey in order that the image should look back. Far from the paranoia surrounding the voyeurs of naturalism, who may discover themselves being seen watching what they should perhaps not have seen,64 this is an intentional confrontation, prepared by ritual. The word for it in Greek is theoria, which from Plato to the church fathers means contemplation, meditation, vision.65

The difference from the visuality of naturalism is fundamental. For in mimesis, the viewer stands apart from the world of the image, which operates illusionistically in its own space ( just like nature) and according to its own narrative logic. That space and logic may be realistic (like our own world, our own sense of perspective, time, form, and so forth), but looking at it is like looking through a screen into someone else’s life. The viewer is invited to enter that world vicariously and voyeuristically, to penetrate the screen through an act of imaginative fantasy. Appropriations of the world of the image may involve identifying with the narratives governing its characters (as Achilles Tatius invites us to do), or imagining what the story might be and perhaps second-guessing the psychological motivations of its characters (as Philostratus likes to do). But there is no contact. Indeed, the more the possibility for contact is offered, and the more the image’s illusionism tempts us into believing that it is real, the closer we come to the tragedy of Narcissus or the deception of Pliny’s birds, who flew up to the canvas of Zeuxis’ famous painting only to find that the grapes they desired were but pigment.66 Ultimately, because there is no contact in the regime of naturalist representation, there is only longing, nostalgia, and frustrated erotic desire.67

By contrast, in ritual-centered viewing, the grounds for a direct relationship have been prepared. The viewer enters a sacred space, a special place set apart from ordinary life, in which the god dwells. In this liminal site, the viewer enters the god’s world and likewise the deity intrudes directly into the viewer’s world in a highly ritualized context. The reciprocal gaze of this visuality is a kind of epiphanic fulfillment both of the viewer-pilgrim, who discovers his or her deepest identity in the presence of the god, and of the god himself, who receives the offerings and worship appropriate to his divinity in the process of pilgrimage rites.

I have been arguing for a visuality which is deeply different from that of naturalism. Let me follow Norman Bryson in this definition of the visuality of naturalistic art: “Between subject and object is inserted the entire sum of discourses which make up visuality, that cultural construct, and make visuality different from vision, the notion of unmediated visual experience. Between retina and world is inserted a screen of signs, a screen consisting of all the multiple discourses on vision built into the social arena.”68 Ritual-centered visuality is no less a cultural construct. But its aim, within a sacred context, is to undermine the multiple discourses of the social arena, the screen of signs produced by and carried over from “everyday life.” Instead the pilgrim is put through a process of purification of body and mind, in which the self is prepared in a liminal space for the meeting with a being from the Other World.

This ritual-centered visuality may be defined in many ways—as the putting aside of normal identity and the acquisition of a temporary cult-generated identity, or as the surrendering of individuality to a more collective form of subjectivity constructed and controlled by the sacred site, or as the provision of the deity as a vessel into which individual pilgrims can pour their devotions and their aspirations. But its positive definition (which is always open to contestation, depending on how much of an insider’s or an outsider’s view one takes) is less important than what this kind of visuality negates. In effect, ritual-centered visuality denies the appropriateness of a Philostratean strategy of interpreting images through the rules and desires of everyday life. It constructs a ritual barrier to the identifications and objectifications of the screen of discourse and posits a sacred possibility for vision, which is by definition more significant since it opens the viewer to confronting his or her god.

The formal appearance of a particular image is less important here than what one might call the naturalism or non-naturalism of viewing. Some cult images (Aphrodite of Cnidos, for instance) were very naturalistic indeed, but the correct ritual preparations and attitudes could prevent the viewer from succumbing to the dangers of voyeuristic projection.69 The naturalism of the Cnidian Aphrodite—one of antiquity’s sexiest and yet most sacredly charged cult images—shows that we are not looking at mutually exclusive visualities that were separate in antiquity, though they may seem so to modern sensibilities. Rather we have a dynamic spectrum of interchanging visualities that appear to have existed in a permanent dialectic and that could manifest together in the same viewer. Temples were not only the centers of pilgrimage but also the prime sites for ancient art galleries (in which images capable of all the naturalistic wiles extolled by Pliny or Philostratus were some of the main exhibits), while cult icons like the Cnidian Aphrodite were among the ultimate pinacles of illusionistic verisimilitude. Pausanias, the Greco-Roman viewer whose narrative gives us by far the richest evidence for responses to art, is certainly capable of both kinds of visuality in his text.70 At issue, then, was the kind of viewing which a spectator might choose at any one time; and of great importance to the choice a spectator might make was the ritual context. What is, by our standards, so strange about ancient visuality is its conflation of regimes of spectatorship, incorporating what may seem to us such antithetical archetypes as “medieval” and “Renaissance,” “abstract” and “naturalistic.”

Ultimately, in a sacred context, naturalistic images ceased to be necessary, since the kinds of viewing they enticed were at odds with the suppression of such viewing ideally encouraged and policed by ritual. In the context of Greek art and cultural attitudes toward art in the Roman empire, this kind of sacred visuality became gradually associated with archaism. The entire text of Lucian’s De Dea Syria is self-consciously archaizing, especially in its use of Herodotus’ dialect of Ionic Greek instead of the more usual Attic of the period. Pausanias specifically praises the divine nature of the archaic when he comments on the statues of the mythical sculptor Daedalus: “All the works made by Daedalus are somewhat uncouth to look at, but something divine stands out in them.”71 Yet ironicially the archaic would become the modern. It would be in the visual space of sacred ritual, a space inherited from pagan antiquity though of course transformed by the tenets of its own doctrine, that medieval and especially Byzantine Christianity would establish not only its very particular and characteristic sacred forms but also its remarkable theological theorizing of the image.

Return to Book Description

File created: 8/7/2007

Questions and comments to:
Princeton University Press

New Book E-mails
New In Print
PUP Blog
Princeton APPS
Sample Chapters
Princeton Legacy Library
Exam/Desk Copy
Recent Awards
Princeton Shorts
Freshman Reading
PUP Europe
About Us
Contact Us
PUP Home

Bookmark and Share