When Robert Fleischer first met me - on Sept. 5, 1978 - I am sure he wished he hadn't, for he had to buy me dinner. We had just attended a reception at the British Museum in London for the participants to the XI International Congress of Classical Archaeology, and I had approached him to comment on the paper he had delivered that day, about a new copy of the Polykleitan Diskophoros from Ephesos. I happened to mention that I had much enjoyed his book on the Ephesian Artemis, and he seemed so surprised that I had really read it that we decided to continue our exchange of ideas over dinner. Unfortunately, after eating, I discovered I did not have enough money with me to pay for my share, and he had to treat me! Our next encounters were more favorable, and we forged a strong friendship in 1989 when he was the Kress Visiting Lecturer for the Archaeological Institute of America, with base at Bryn Mawr College from January to May of that year. We taught a joint seminar (on Hellenistic Ruler Portraits) and have been in touch ever since, because I often need his generous advice on sculptural matters of common interest.

To honor Robert Fleischer in this auspicious occasion (but archaeologists do not get older - they simply become more interesting!), I want to discuss an East Greek goddess - not the formidable Artemis of Ephesos encased in her elaborate apparel but the equally formidable, if naked, Aphrodite of Knidos. My comments should be taken as an intensely personal reaction to some recent pronouncements on this famous statue by Praxiteles that has become the focus of attention for both men and women, although the former tend to look at it as an object of male desire while the latter see it as an example of female humiliation. The pertinent literature is vast, but I have chosen to refer primarily to two studies, by Andrew Stewart and Nanette Salomon, that may be said to epitomize these approaches. (1)

According to Stewart, Praxiteles depicted Aphrodite being surprised at her bath, either before or after having taken it, and caught in a play of glances. On the one hand, she seems to shield herself from a viewer who has just entered the temple; on the other, she turns her head to look at a second "intruder" to whom she appears revealed, drapery and gesture being inadequate to cover her femininity from an oblique perspective. Her body has an "open" and a "closed" side (the proper left and right respectively), and a slight stoop that creates an inviting concavity asking to be filled. Her mouth parted in a smile is a further invitation to the male invasion since, according to ancient lore, it connects directly with her womb. The consequences of such an alluring rendering are to be seen on one of her thighs, as explained by an ancient anecdote, regardless of its veracity; yet the sculptor had even omitted to render the goddess' vulva, "presumably in order to shield her body from the male spectator's penetrating glance - or worse." Aphrodite is caught in a "putative love triangle" ... "of voyeuristic complicity and erotic rivalry," although capable of manipulating the human members of it at her own terrifying will. (2)

For Stewart, therefore, Praxiteles' creation embodied a goddess, although one dangerously playing with men and humanly seductive in her sexuality. By contrast, Salomon can only see the woman in the composition. Stressing the differences in the rendering of the male versus the female nude (a contraposition established also by Stewart), she comments: "The hand that points also covers and that which covers also points. We are, in either case, directed to her pubis which we are not permitted to see. Woman, thus fashioned, is reduced in a humiliated way to her sexuality." The Knidia's attitude, in her interpretation, reflects the fear of a woman sensing a sudden presence and its implicit threat of violence; the statue, is the "image of a woman on display whose primary characteristic is that she does not wish to be seen." (3)

The attribution of such human feelings to Praxiteles' creation is not a recent occurrence. The interpretation of a woman embarrassed by her nudity in front of an observer had already been introduced by a most authoritative study: J. J. Bernoulli, Aphrodite. Ein Baustein zur griechischen Kunstmythologie (Leipzig 1873). Christine Havelock has now provided an illuminating commentary on that work, by setting it in the context of its own times, when explanations were needed to justify what seemed an unprecedented revealing of the female body by a Greek fourth-century master. Thus, not only was a progression established, from fully to partially draped to completely nude renderings in sculpture, as if gradually to prepare the ancient viewers, but also a motivation for the disrobing had to be supplied - hence the suggestion of Aphrodite's bath, although not mentioned by the ancient source on the Knidia. (4) This explanation has been accepted ever since, even when different rationales are proposed for Aphrodite's bathing. We shall return to this point below.

In attributing human "feelings" to the composition, scholars may have been (and still could be) influenced by Athenaios' account (13.590) that Phryne, a courtesan and Praxiteles' mistress, had posed for his statue. Despite the fact that similar anecdotes about artists in love with their models are known through time, and are therefore likely fabrications, the belief persists that a real woman inspired the Knidia, and - if indeed she was Phryne - not just a woman but a prostitute, with all the subconscious reactions such belief may inspire. (5) Two points here are important. First of all, prostitution in antiquity was not illegal and was often sacred, especially in the East; therefore - had the anecdote been considered true - it would not have produced in Knidos the kind of reaction it engenders in more recent commentators. (6) Second, Greek sculptural practices, at least for the Classical period, are not known to have included the use of live models; even the Hellenistic phase, with its alleged introduction of casts, appears to have focused on details and inspiration, perhaps on sketches, from reality rather than on the direct copying of a specific human form. Indeed, there is nothing particularly feminine in the Knidia, at least to judge from the Roman copies at our disposal.

Perhaps the most damaging statement is this last. We have, in fact, no fragment surviving from Praxiteles' creation (7), all our analyses are based on considerably later copies. To be sure, the basic lines of the composition are known through depictions on Knidian Imperial coins, thus ensuring that the various elements of it - the cloth, the vessel, the gesture - are part of the original. But everything else may have been altered and modified not only to suit the requirements of a Roman clientele but also to accommodate the more limited skills or the reduced economic investment of the copyist. The many statuettes of the type, beginning with the earliest ones from Delos, are more in the line of souvenirs than of true replicas, especially since the prototype - a cult image - would not have been accessible for casts or even precise measurements. (8) The pose with thighs more tightly compressed and drapery pulled closer to the body, as if to provide some coverage, may have introduced technical modifications in order to supply a more substantial support and to eliminate the strut connecting cloth to hip in the larger replicas. Stewart seems to prefer the startled to the relaxed version, but here the coins do not offer confirmation, and the change toward a more "narrative" content may have resulted from the passing of time, toward the Roman Imperial period with its more stringent ideas of decor and propriety. As for the strut in the larger copies (differently placed in each), it too is not verified by the coins, and Praxiteles, an experienced carver, may not have needed it. (9)

The "femininity" of the figure varies from replica to replica, and the two headless torsos in the Terme Museum and the Louvre are almost masculine or adolescent in their spare forms. But even the most statuesque examples (the Colonna and the Belvedere) are certainly not typical of a Mediterranean woman with full breasts, rounded abdomen and - especially - wide hips and ample thighs. Contrast, for instance, the seated and reclining goddesses (K, L, and M) on the east pediment of the Parthenon, to see what the standards of female beauty may have been in Classical times. Seen from behind, moreover, the Colonna replica itself looks almost masculine. (10) The Roman copyists may have "slenderized" their prototype - just as the Aphrodite of Arles, in the Louvre, was trimmed down to suit the taste of Louis XIV. (11) No fourth-century original sculpture of a naked female is now extant to provide a meaningful comparison, but the very proportions of the Knidia are too elongated to be derived from a real woman. lf the scale of the Colonna replica is to be believed, at 2.04 m. in height she is well beyond the range of an ancient Mediterranean female, and so is even the Braschi Aphrodite in Munich, which stands at 1,65 m. On all these grounds, it seems to me highly unlikely that Phryne - or any other model - might have posed for Praxiteles.

Much has been made of the omission of the vulva, despite the fact that the goddess points to her genitals. On the one hand, therefore, the image "has been taken to task" by feminist studies because of this "pudica gesture" that suggests shame, as mentioned above; on the other hand, the same theoretical trends have found the sculptor deficient in his attitude toward women for not having represented the female anatomy in its full reality. Yet the naming of the gesture is a modern convention that has no reflection in ancient literature; as for the apparent absence of the vulva, other reasons may be found. (12)

The first that comes to mind is that not Praxiteles but the Roman copyists may have omitted the detail, for typical Roman modesty. (13) Another possibility is that, in the original, the position of the right hand prevented or rendered unnecessary the actual carving of the cleft. Perhaps the hand adhered to the pubis (to strengthen the arm) even if only with the fingertips, or covered it so well as to obscure the feature. Regrettably, the extant copies vary extensively in this respect and cannot be taken as secure reflections of the prototype. Finally, the vulva might have been simply painted (again, for technical or aesthetic reasons), and the surviving marbles either neglected or have lost the marking. Indeed, the Knidia was painted and not just in details: her entire flesh might have received ganosis with tinted wax, as was the practice in ancient sculpture - which may have considerably softened the "luminous radiance of the finely crystaled stone" even if it did not fully cover the impurity that gave rise to the love-making anecdote. (14)

As a last hypothesis, we could assume that the vulva was intentionally omitted to emphasize the inviolability of the goddess, but this seems unlikely. Perennial or restorabie virginity is not one of the attributes of Aphrodite, and her very gesture - of lifting her drapery from the vessel, thus, virtually, unstoppering it - was probably meant to be read as an allusion to sexual intercourse. As ancient gynecological treatises mention, the female uterus was visualized as an inverted jar, and the nomenclature of its various parts corresponded to the terms used for vases. In addition, kredemnon, the word for a garment covering the head, also meant a lid or stopper. Since the length of the cloth held by the Knidia varies from replica to replica, we cannot tell whether the original garment was a mantle or, more probably, a shawl, a kredemnon, whose removal would have opened the way to the goddess' fertility. The hand over the pubis and the hand over the jar would therefore have carried the same signification. (15)

The cloth was certainly not a bath towel, as some have described it, and the vase may not have been for a bath. We have mentioned above Bernoulli's reasons for creating the context. Havelock tried to salvage the imagery by suggesting that water for bathing alludes to preparations for imminent love-making, and that the hydria symbolizes Aphrodite's "powers, her fertility, her unending freshness and youth." A smaller vase might have been for perfumed ointments, also part of the bath ritual. (16) Here again I am not sure we can recapture the original: if the vase contained water for the bath, why not a loutrophoros or a lebes gamikos? Why so much variation from replica to replica, unless the main signification of the object was a "vessel"? That Aphrodite stopped in Knidos to wash the salt from her body after her birth from the sea - as one version of her myth goes - is not even remotely suggested by Praxiteles' statue with its neat and elegant coiffure not at all indicative of bathing activities.

The Knidia was a cult image; as such, it would have embodied symbolism, but definitely not narrative. There is no need to have recourse to travel stopovers, excuses for nudity, reasons for bathing, explanations for momentary actions. The turn of the Aphrodite's head has been caught in the same web of conjectures. lt could have been "the consequence of the formal and abstract conception embodied in the contraposto position of the body"; in fact, it has often been remarked that the Knidia is the female counterpart of Polykleitos' Doryphoros, not simply in popularity through ancient times, but also in stance and chiastic pattern. Or the turn could have been the depiction of a natural reaction to the unexpected arrival of a viewer, according to the "narrative" version - hence Stewart's emphasis on glances. (17) Yet a consideration of the setting for the statue may undermine these observations.

Most replicas of the Knidia are of heroic size, over or around 2 meters in height, very close, in fact, to the Doryphoros. A base proportionate to the scale of the figure would also have been used. Therefore no normally tall human being could have been caught in the goddess' glance, if one follows the trajectory of her eyes looking absent-mindedly into the distance. The more moderately sized replicas (the Braschi Aphrodite, for instance) would equally stand too high on their pedestal for a crossing of glances. I would agree that a slight turn of the head is a compositional device in vogue since the Severe period, perhaps to break the strict frontality of Archaic statues but also to impart a sense of life and animation to the sculptures. Yet we traditionally expect a cult image to establish direct eye contact with the worshippers - why is the Knidia looking elsewhere? Perhaps the statue looked out toward the sea - a prominent presence at Knidos - in keeping with her local epithet, euploia, "of good navigation," as protector of sailors and seafarers. Regrettably we still ignore where Praxiteles' statue stood.

lt seems now certain that the round structure excavated in the early 1970s by an American team directed by Iris Love is not a temple to Aphrodite. First of all, it is not a monopteros but a tholos with regular walls; its outer colonnade was Corinthian, not Doric, as originally postulated on analogy with the round building at Hadrian's Villa that housed a replica of the Knidia. Technical details, moreover, show that it was contemporary with the neighboring altar inscribed to Apollo Karneios and signed by two second-century sculptors. Finally, and more damaging of all, is the discovery that dedications to the temple owner name Athena, thus making the tholos a building to that deity. (18) According to Lucian's description, the Aphrodiseion was surrounded by trees and other plants - open ground incompatible with the paved area near the tholos. Although other suggestions have been advanced as to the possible location of Aphrodite's temple, no confirmation has yet been found. (19)

Even dedications to Athena, however, consist of terracotta busts of women holding their breasts. lt has been argued that these are not images of the deity but of female devotees making the offering; the fact, nonetheless, remains that symbols of fertility were prominent in Knidos. I am convinced by Blinkenberg's explanation that Praxiteles had been commissioned to make a cult image reflecting a local tradition, as exemplified by many figurines of earlier date. Pliny's mention (NH 36.20) of a travelling master peddling around his statues makes no sense, in view of the investment of time and money involved in the creation of the marbles, let alone the difficulty of transportation. (20)

To some extent, however, Pliny's improbable account is revealing: if his readers could accept that Praxiteles went around like a travelling salesman, his renown could not have been established and must have accrued to him solely because of the Knidia. Yet all modern commentators harp on the originality of his composition, and even those who believe in a previous iconographic tradition point out how inspired was the sculptor's solution to the problem posed for a Greek by the commission. Here again I think we are misled by our post-Renaissance notions that focus on the distinctiveness of artistic creation and the fame of individual masters. Whatever is truly known about Praxiteles' oeuvre does not make him a revolutionary or innovating genius. (21) On the other hand, we do not need to justify the making of a naked image in terms of an inevitable progression; even fifth-century statues with "transparent" drapery would have seemed less revealed when the surface of their garment was covered with solid paint. (22) Through my years of studying Greek sculpture I have come to the conclusion that a linear development of its style is an impossible theory fabricated primarily by our own desire to produce a rational sequence in the absence of objective evidence and documents.

To be sure, the validation of a work of art lies in its having meaning for each viewer and each generation; it is therefore correct for Stewart to look at the Knidia from his male-oriented point of view, for Salomon to read in it her own sense of vulnerable femininity, for other interpreters to derive from it the message that to them seems most relevant - as long as we do not try to attribute such intentions to the fourth-century Greeks, whose mentality and viewpoint we cannot hope to fathom, let alone approximate. With this caveat, I venture to offer my own conception of Praxiteles' Knidia.

lt was a religious icon, a focus of positive cult, made on specific commission for a place with a strong mixed tradition, both Greek and Near Eastern. As such, it was not narrative or anecdotal but symbolic and powerful, an object of awe in its nudity used as an attribute. Instead of "sexuality," however, we should speak of "fertility," an important concept in antiquity when children's mortality was high and population growth was desirable. Therefore the goddess - in an epiphanic rather than a momentary pose - not only pointed to her reproductive organs but also alluded to them in the unstoppering of her vessel. Her image at superhuman size, aloof and unconcerned, dangerous because naked, would have inspired respect and even fear in those, male and female alike, who went to her temple openly to worship - not pruriently to peep, since the consequences of seeing a goddess naked against her will were terrible and well known in antiquity. Such feelings may have changed with time, (23) but we need not credit the story of the stain, which probably was occasioned by the impurity in the stone rather than vice versa. In this respect, I want to close by recounting the making of a modern anecdote, as told by W. B. Dinsmoor in 1955 to the students of the American School of Classical Studies at Athens assembled on the Akropolis.

At that time, one of the columns of the east façade of the Propylaia, to the left of the central passage, still had a broken metal saw embedded between two of its drums. lt was a 19th-century tool, used (by F. C. Penrose ?) when, in oder to measure the columns accurately, an attempt was made to dislodge pebbles and debris that had infiltrated the join through times of disrepair caused by earthquake or war. When the intrusions were successfully sawn to dust, the upper block fell back in place, entrapping the blade which was left behind. During later restorations within this century, the Greek authorities wanted to remove the saw, but the local guides begged them to leave it there, and Dinsmoor had heard one of them tell some tourists the following story - convincing because of its clever mixing of facts and ancient sources with personal interpretation:

Originally, the Greeks built their columns monolithic, but then the fashion changed and drums were preferred. Perikles therefore ordered his favorite slave to cut the existing columns of this gateway into drums. While the slave was carrying out his task, however, he fell to the ground and was seriously hurt. Perikles was much distressed and prayed to the Health Goddess to restore his man. Plutarch tells us that she appeared to Perikles in a dream, the slave was healed, and the grateful statesman erected an inscribed altar and a Statue to Hygieia, which [pointing] you can see there, next to the outermost column. And that [pointing upward] is the saw that the slave was using when he fell.