1)    SHELLEY'S little-known "On the Medusa of Leonardo da Vinci in the Florentine Gallery" (1819) [see figures 86, 27] belongs to the same genre as Keats's "Ode on a Grecian Urn" and Byron's description of the Dying Gladiator in Canto IV of Childe Harold's Pilgrimage--that of ekphrasis, or the poetic description of a work of art. Formally speaking, ekphrasis is any description which brings a person, place, or thing vividly before the mind's eye.[1] But this definition is only useful if we are thinking in broad terms, or if we are invoking the rhetorical function of the word as set forth in Aphthonius's Progymnasmata, one of the earliest textbooks of prose style. For our purposes, ekphrasis can be understood as a means of citing one work of art within another, in Leo Spitzer's terms a way of reproducing "through the medium of words [...] sensuously perceptible objets d'art ".[2] In The Sister Arts, still one of the best introductions to literary pictorialism, Jean Hagstrum offers yet another, perhaps more empathetic definition: ekphrasis "refer[s] to that special quality of giving voice and language to the otherwise mute art object".[3]

2)    Although ekphrasis as a genre had been largely dormant since the early fourth and fifth centuries, when Christian writers had used it as a way of celebrating the elaborate architecture of Byzantine churches, it underwent a revival in the eighteenth century.[4] The writings of Winckelmann and Lessing, combined with the excavations (some would say pilferings) of Lord Elgin helped to stimulate a resurgent interest in sculpture as well as in the visual arts generally. Starting with Matthew Prior's "Picture of Seneca Dying in a Bath: By Jordain" (1720), the period witnessed a steady industry in ekphrasis. Among the better known pieces were John Dyer's "Epistle to a Famous Painter", Edward Young's "On Michael Angelo's Famous Piece of the Crucifixion", John Byrom's "Verses written under a Print, representing the Salutation of the Blessed Virgin", and Henry Hart Milman's "The Apollo Belvedere", which was reprinted several times, most famously in the Annals of tbe Fine Arts.[5] An indication that the genre had been wholly revived by the Romantic period was the decision in 1806 by both Cambridge and Oxford universities to sponsor a competition for the best poem on a work of art or on ancient culture. [6]

3)    Shelley's "On the Medusa" properly belongs to a tributary of this popular current, the portrait ekphrasis, inaugurated by Alexander Pope in "Epistle: To a Lady" (1735). During the eighteenth century, it became increasingly common for male poets to write poems celebrating portraits of beautiful women. Thomas Tickell's "On a Lady's Picture"[7] stands as a typical example of a form which was devoted to a laudatory inspection and cataloguing of female accoutrements. Towards the latter half of the century, poems on portraits began to appear more and more frequently in the fashionable collections of fugitive poetry and in the popular magazines of the day--particularly Gentleman's, Universal, and London.

4)    Like their predecessors in the Greek Anthology or Philostratus's Imagines, these poems tend to remark on the verisimilitude of the representation and to praise the artist's mimetic powers. In noting the beauty of the feminine image, however, the poets often use the metaphor of ravishment, and perhaps more significantly, muteness, to describe their own condition. While the portrait "breathes" life, the viewer stands breathless, paralyzed by the sublimity of the image. A kind of aesthetic osmosis occurs whereby the material qualities of the representation are transferred to the awestruck observer, who becomes powerless and inert, himself a kind of spectacle. The poet's challenge in these ekphrases is therefore to walk a fine line between making the image speak (in Hagstrum's terms), and avoiding the loss of speech himself. He must elude the fate of the University Prize poems--a passive, purely metonymic description of the object--and yet also devise a way of being sufficiently in control of his own imagination and language so as to enter fully into the image. He must translate its beauties into words without compromising the boundaries of his own fragile selfhood.

5)    In order to achieve this, the poet adopts a rhetorical protocol in which all the peripheral elements of ekphrasis--the artist's life and materials, the historical context, etc.--are employed as friendly aides de camp in occupying or decentering the image. J. Holland's translation of Pliny's "On the Picture of Medea", published in the Gentleman's Magazine for April, 1804, is a good example of this reliance on external elements:

6)    That an alliance with the figure of Medea is dangerous becomes apparent through the poet's careful use of frames. Holland's view of the picture, mediated in the first instance by Pliny, is further off-set by the illusion that the artist himself stands close by, benevolently watching over our shoulder. Though the poem is entitled, "On the Picture of Medea", the opening lines concern "the great master" and emphasize his artistic and mimetic powers. The implication is that Medea's image remains dangerous even in paint, and should be sequestered. Holland must take steps to discharge her threatening power by making the artist as well as the process of translation part of the dynamics of perception.

7)    In addition, the poet is careful to point out the artist's discipline, and his strict control over subject matter and form. His pencil may contain "magic", but it is a magic that knows its limitations. As important to the picture as the myth is the delicate choice of a moment in the plot directly preceding its climactic bloodbath; indeed, the artist's discretion is as crucial to the poem as his imaginative powers, for in the end he "draws the veil" over subsequent events. At work in this ekphrasis is a subconscious interest in decorum and restraint, and a corresponding leeriness about aesthetic identification.

8)    At the conclusion of the poem, Holland draws another formal demarcation between the painting and his own medium. Even as the artist discreetly refuses to portray the "dire sequel of the dreadful tale", the poet finds no difficulty in doing just this. He goes on to sum up Medea's fate in a sanguine little couplet, made even more mischievously dramatic by caesura and dash: "Else had we seen a parent's hand embrued, / Suffice the horrid thought, in filial blood--" The effect here, whether the poet intends it or not, is firmly to separate the temporal elements of his narrative from the static ones in the picture. Whereas the poet completes the myth without breaching decorum, the painter will "stain" the canvas if he chooses to depict it. For the painter, the deed and its portrayal become synonymous crimes. In fact, the artist's restraint, personified in his "fault'ring touch", carries moral overtones which do not obtain for the poet. He bears a direct responsibility for the action that the poet manages to evade if only because he is working in a different medium.

9)    Implicit in this contradiction is a fear of the powerful feminine image, what Mitchell calls "iconophobia".[8] To write about the massacre of Medea's brood is to distance the event, turn it from an image into an idea. Inscribed, the myth becomes far less threatening to the poet, who can banish a "horrid thought" with greater alacrity than a fearsome image.[9] The tendency to privilege the unseen over the seen runs through the poem: the artist, we recall, is praised for painting "the tumults of Medea's mind; / Her inward struggles", rather than for any purely technical displays of color or design. In fact, the closer he comes to writing or inscription (he uses a "pencil" instead of a brush), the closer he comes to a masterpiece. This bias against painting, along with the propensity to replace visual art with writing, to subordinate the spatial to the temporal, governs Holland's ekphrasis and serves as a defense against the irrational fears which the image generates. If he successfully turns Medea into language, then the poet has exorcized a large part of her mesmerism, and has deferred the muteness that traditionally accompanies any perception of a sublime object.[10]

10)    Much of the antagonism between the sister arts implicit in Holland's poem originates in a method of approaching the aesthetic object devised by theoreticians of ekphrasis like Lessing.[11] In Laokoon, Lessing draws up a set of finely delineated moral and aesthetic boundaries between the arts. He argues that because it is temporal, poetry is inherently more flexible than painting and can therefore express a greater degree of imaginative freedom and originality. Since painting is innately static, its attempts to portray sequence are necessarily experimental and, for Lessing, dangerous. Any conjunction between the arts, any ut pictura poesis then, must be undertaken (if at all) with a great deal of caution.

11)    This debate over the proper relations between the arts finds its clearest embodiment (and perhaps its inception) in the myth of Perseus and Medusa [figure 28], whose confrontation reflects a primitive allegory of ekphrasis. The hero's pilgrimage to the Gorgon, who remains concealed in her cave/museum surrounded by other works of sculpture (though of her own making), reminds us of any traditional visit to an objet d'art by an ekphrastic hopeful. Instead of a camera, guidebook, and notepad, however, Perseus comes equipped with a more combative set of paraphernalia. For unlike the Theseus or the Apollo Belvedere [figure 84], the Medusa represents an aesthetic trap that threatens to realize what is only a metaphoric possibility in the ut pictura poesis tradition: the fate of speechlessness and paralysis. The myth's predominant themes, in fact, have to do with perception, with the power of images, and with the ways Perseus--as prototypical male observer--controls and subdues a dangerous feminine icon.[12]

12)    One of the primary sources of consolation for the hero's fear in this instance stems precisely from the poetics of ekphrasis and its handling of the trompe l'oeil. Just as Perseus decapitates Medusa through an elaborate sleight-of-hand, through the careful manipulation of a mirror, so too does he slay the dragon who threatens Andromeda by directing an illusory shadow of his own figure across the ocean. As Louis Marin has observed, the hero prevails by turning the threatening image into a representation, and therefore something necessarily distanced: "Comment conquérir le regard maintenant qu'il a l'oeil? [...] Par la ruse du miroir, par la ruse de la représentation."[13] (How to overcome the gaze now that he has the eye? [...] By the trick of the mirror, by the trick of representation.) Both ploys are not only cunning, relying for their effectiveness on a dexterous orchestration of visual images, but demonstrate the deadly possibilities of ekphrasis. If nothing else, Perseus becomes the first great practitioner of the genre, a sly conjurer much more than a brawny god.

13)    Similar to Holland's poem on Medea, the Perseus myth is arranged so that the central ekphrastic encounter is framed and distanced by other experiences which appear to establish Perseus's artfulness and prowess. His initial deception of the Graiae and later, his chivalric rescue of Andromeda, surround what is a far more ambiguous rendezvous in Medusa's cave. Within these larger structural parameters, we find a corresponding interest in protective, or what Graves calls "prophylactic" charms.[14] Before setting out on his adventures, Perseus must secure the proper tools from the Stygian Nymphs. His collection eventually comes to include a magic wallet for Medusa's head (a kibisis), a helmet which makes him invisible, a sickle or sword, winged sandals, and of course, a brass shield. These devices represent the physical equivalents of all the careful rhetoric Holland and the eighteenth-century poets of ekphrasis mobilize to ensure their own and the viewer's safety. They are necessary precautions in mitigating the threat that the image poses.

14)    At the center of the myth lies an elemental ekphrasis which, as we have come to see in the genre, reveals more about the observer than the observed. Louis Marin has argued that when Perseus sees and then decapitates the Medusa, he is symbolically purging himself of all "evil" emotions, effectively destroying his own unruly passions:

15)    Although he is able to eradicate these "passions mauvaises", his victory is necessarily qualified because of its dependence on trompe l'oeil (and because of the intervention of Athena, who "guides his hand").[16] We might note that neither Ovid nor Apollodorus describes any facial aspects in their portraits of the Medusa, and though Perseus looks on the image in his shield, he is unable to relate more than a few incidental details of hair and teeth.[17] Freud has argued that the encounter offers a clear instance of castration anxiety, and it seems likely given Perseus's ambivalent relation to the female characters, that the incident indeed reveals a lesson in masculine insecurity.[18]

16)    The remainder of the myth, however, is designed to assuage this sense of male powerlessness and inadequacy. As soon as Medusa is beheaded, Perseus is rewarded with a winged and glorious white horse, who leaps from the Gorgon's spilled blood like a kind of jubilant phallic trophy. As an image of masculine resurrection and reassurance, Pegasus cannot be bettered in all of mythology. The horse immediately gives Perseus a freedom to move about in space and time that dramatizes the exact antithesis of the fate suffered by his predecessors. After he has escaped Medusa's oppressive den, he soon meets up with Atlas and roundly defeats him (aided, of course, by Medusa's head). As if this instance of wish fulfillment were not enough, Perseus shortly comes across Andromeda in a scene that further restores his confidence and compensates for his fear before the Medusa. Naked and chained to a rock, she presents the image of an ideal doll that Perseus may caress and fondle in safety; for she is at once immediate, beautiful, and harnessed. She offers no threat, bound as she is, but exists as an object already under control, an object straight out of Perseus's most vivid fantasies of female bondage and submission. Since masculine and feminine roles have been at last realigned--Andromeda is now the figure who is properly static, paralyzed--Perseus can claim her and reestablish a degree of comforting patriarchy.[19]

17)    Like the wallet that Perseus carries in order to protect himself from an arrant view of Medusa, the myth cleverly disguises its real concern which has to do with the fragility and tenuousness of male sexuality. Its central, though unspoken theme, involves Perseus's illusion of mastery, his fitful fantasy of control in the face of a terrifying feminine power. Such an illusion, girdered by an entire infrastructure of symbolic psychological safeguards and compensations, is completely undermined in Shelley's own poem. Here the scaffolding carefully erected between observer and observed is dismantled, not piece by piece, but all at once, and the speaker remains fixedly staring at the Medusa as if she were actually present. "On the Medusa" describes what Perseus really saw when he beheld the Gorgon, and why he could not write it down. "If ekphrastic poetry has a 'primal scene,'" Mitchell muses, "this is it."[20]

[This link will take you to the Romantic Circles Electronic Edition of Shelley's poem. Use the Back button on your browser to return.]

18)    In many ways these lines culminate the portrait-poem genre. Shelley's speaker, in his abandonment and surrender to the image, brings to the surface all the fears of an observer confronted by a dangerous female other. Where the eighteenth-century poems and the myth of Perseus are concerned with developing a poetics of cautious approach and with adopting defensive measures to stave off the harmful effects of the image, "On the Medusa" deconstructs this entire strategy by removing all barriers and prophylactic frames. The poem expresses a desire to transcend distinctions of genre and gender by inscribing the image of the Medusa on its own text, writing it into the very language of the stanzas. As such, it is the most daring of the ekphrases we have seen and the most potentially reckless.

19)    As a matter of course, the poem works to eradicate any actual or rhetorical frames. We know that the subject is a painting only through the title; after that, as Jacobs and Mitchell rightly maintain, Medusa is described as if she were actually present to the observer. Moreover, Shelley does not, as Holland, resort to the painter's authority and benevolent guardianship to rescue us from the horrors of the myth. Instead, Leonardo vanishes as soon as he's mentioned; indeed, as Shelley no doubt knew, the painting was only attributed to Leonardo. In this sense, the origination of the aesthetic object, rather than being confirmed and celebrated in the poem, is put radically in question. Medusa appears as an entirely autonomous entity, without authorship and importantly, without her conqueror Perseus. Shelley undermines the artist's patronage and his control, as well as male dominion generally, as he focuses attention squarely on the feminine image. Even the ostensibly unassailable role of the spectator as active viewer (the roving eye so common to the eighteenth-century portrait poems) comes under fire; as Mitchell argues: The speaking and seeing subject of this poem does not speak of or (in a sense) even see Medusa: the opening three words appropriate both these roles: "It lieth, gazing." Medusa, the supposed "seen object" of the poem, is presented as herself the active gazer; other possible "gazers" on this spectacle are presented as passive recipients. The voice of the poem is simply a passive recorder, an "ever-shifting mirror" that traces the "unending involutions" of its subject. The pun on "lieth" suggests that the mute ekphrastic object awaiting the ventriloquism of the poetic voice already has a voice of its own.[21]

20)    The answer to Carol Jacobs's query: "Who is the gazer - Perseus, his predecessors, the painter, the poet, the reader?"[22] thus becomes a glib "none of the above". It is Medusa who usurps the traditional generic role of the male observer, relinquishing her status as passive objet d'art to Shelley. Such an astonishing volte face manifests itself on all levels of the poem, including those of syntax and diction.[23] It should come as little surprise that there are three important lacunae--in lines 18, 37, and 44 (the last comes in the additional stanza)--and that they each involve lapses in the poet's descriptive powers. Medusa cripples the most essential faculty of any would-be ekphrastic writer: the capacity to generate adjectives that sensuously and imaginatively reproduce the painting in words. The resulting gaps may occur, as Neville Rogers has argued, because Shelley "lacked, or could not remember, some single word needed to complete a line (and) was unable to halt his creative process while he sought it",[24] but it is more compelling to suggest that they appear because the speaker is beginning to feel the effects of Medusa's "graven" image. His own "lineaments" are turning into hers, becoming "stone".

21)    Instead of brief slips, then, these are key moments of absence in the poetics of ekphrastic encounter. They symbolize the imminent collapse of the imagination, the spaces where the dark serpentine ink can no longer pursue its "inextricable error", and the imperious Medusa threatens to impose her pallid reign. Indeed, the terrifying whiteness of the page reflects the ghostly cheek of the Gorgon and the vapors which drift up from her mouth.[25]

22)    The pun on "lie" that Mitchell notes is not the only one Shelley intentionally exploits. Historically, the notion of visual deception is built into the very fabric of ekphrastic description. Shakespeare's ekphrasis of the painting on the siege of Troy in The Rape of Lucrece provides an excellent example:

23)    The artwork seduces the viewer by offering up its illusion or "conceit" of the world as the world; by far the most frequently used word in this ekphrasis, as we might expect, is "seem". And yet the illusion of Troy is so convincing that Lucrece briefly forgets that what she looks on is a simulacrum, and "tears the senseless Sinon with her nails" (l. 1564),[27] imagining that he is her assailant Tarquin. Even though moments later she reminds herself that "his wounds will not be sore" (l. 1568), the painting has already worked its magic by deceiving the heroine and temporarily suspending her ability to differentiate reality from representation.

24)    Ekphrasis means to "speak out", and thus as we have already seen, the poet's challenge is to give the object precisely the words that it lacks. But if the object figures a lie, then the poet is confronted with a dilemma. Does he promote the very duplicity which is the artwork's raison d'être, or does he offer a realization of this illusion as part of his own description, in effect, like Shakespeare sublimating the issue into paradox? ("In scorn of nature, art gave lifeless life", l. 1374).[28] If the latter, is not the poet betraying his own desire for mimesis, for verisimilitude? It seems whatever path the poet chooses--the famous anecdote of Zeuxis's grapes cleverly illustrates this problem--he cannot avoid conspiring with the artist by inviting the audience to participate in a false world.

25)   The irony in Shelley's depiction of the Florentine Medusa, then, stems from the fact that even though the viewer is immediately apprised that the representation "lieth", he nevertheless remains powerless in preventing his own identification with her. Like Lucrece, he cannot help being affected by the image he knows to be false. Whereas in the genre of ekphrasis an awareness of the artwork's deceptive nature typically offers the viewer his means of escape, his way of eluding the dangerous image, in Shelley's poem this awareness provides no such solace. In fact it makes no difference. Even as a representation, Medusa's gaze is paralyzing and lethal. And this is perhaps Shelley's most profound comment on the myth--that Perseus's shield, that all apotropaic devises for that matter, have little efficacy against the potency of images, particularly feminine ones.

26)    There may be still another pun on "lieth", but to understand it we need to look for a moment at the actual painting. One of the first things we notice in glancing at the portrait is the peculiar angle at which the Medusa is depicted [figure 86]. Instead of displaying the Gorgon directly facing the viewer as in Caravaggio's famous painting [figure 27; cf. figure 28], the artist has chosen to arrange the face diagonally, pointing away from us, so that only half the features are visible. By reversing and foreshortening the visage, the artist both defamiliarizes the object, and mitigates its potential threat. Medusa directs her gaze at the skies rather than towards the trembling spectator (even in Caravaggio's version the Gorgon's eyes are not permitted to meet the viewer's); furthermore, "Medusa is already dead by the time she appears in art",[29] and what we see of her is not only partial, an inert fragment, but has been removed from its original and powerful context in the dark cave.

27)    With this in mind, Shelley's pun may also involve the painter's cowardice, his resort to protective frames. The iconography "lieth" because it manipulates point of view, rotates and inverts the head so that it no longer jeopardizes the observer. This realignment of perspective tempts our vision, but never holds it long enough to mesmerize us or to induce paralysis. As prospective Perseuses, we are absolved not only of the responsibility of confronting the look of recrimination, but also of viewing the more gruesome details of our crime - on closer inspection, the severed neck is revealed to be schematized, pointing away from the viewer.

28)    Our interest in Medusa's face is further discouraged by the artist's vivid rendering of the serpents. The eye is more taken with the gleaming convolutions of the snakes, the chiaroscuro of their "mailed radiance", than with the pale uniformity of Medusa's head. In point of fact, they are the first things we see in the picture, occupying nearly the whole of the foreground, and squirming, it seems, across the very borders of the composition itself. Their motion provides a striking contrast to the placid, frozen mien of the Gorgon. As a type of phallic compensation for Medusa's threat of castration, the serpents are the equivalent of Pegasus and are as lovingly dwelt upon as the mythical savior (note Shelley's own pun on "mailed radiance"). The artist certainly delights in the burgundies and blacks of their scales far more than he does in the pale grays of Medusa's cheek. Moreover, the clear diagonal that runs from the hovering bat in the top left to the squat toad at the right has the effect of firmly dividing the two planes of the canvas. The serpents occupy one sphere, the Medusa and her host of venomous rodents the other. This separation only reinforces our sense that the snakes are not physically attached to the head in the first place, that they wind and coil around the Medusa but never seem to touch her; in fact, she has hair of her own which may echo the undulations of the serpents, but never modulates into their "unending involutions".

29)  In addition to the fascinating diversion provided by the serpents, the painting offers still other means of deferring the lethal image. In every corner of the work the artist places surrogate observers who perform our gaze, as it were, by proxy. Shelley himself mentions two of these victims--the "poisonous eft" and "ghastly bat"--but there are more: a lizard perched on a rock at the far left hand side of the painting; a phantom rat which seems to hover just above and to the left of the Medusa; and a sinister toad crouching tensed, as if ready to spring right into the path of Medusa's vision. These creatures absorb the virulence of Medusa's gaze and assume the grave responsibility of a direct view.

30)  In his poem Shelley ignores each one of these protective measures in favor of an immediate confrontation with the eponymous image. "It" is what he's concerned with insofar as "its signifies Medusa's head rather than the entire composition. His poem acknowledges what the artist tries so desperately to conceal through his preoccupation with marginalia--that in the very middle of the composition, emphasized by the intersection of the painting's most prominent diagonals, we encounter Medusa's eye or 'I'. The effect of this eye is uncanny, for while the position of the head makes it appear to be staring upward, its strong iconographic placement along the left-to-right slope resists such an obvious angle of vision. This ambiguity suggests, eerily, that the eye is gravitating towards us. It is precisely this sort of uneasiness that Shelley capitalizes on in his poem, turning the disturbing relationship between Medusa's 'eye' and the viewer's 'I' into his central theme. At stake here is nothing less than the integrity of the observing self under siege from a feminine 'I' who threatens to usurp its dominion.

31)  Any traditional expectations we might have regarding firm distinctions between subject and object are almost immediately disappointed in the poem:

32)  The verbal "are seen" not only inaugurates the subsequent use of a passive recording voice in the text, but also contains a crucial ambiguity. Do we see the "far lands", does the speaker, or does Medusa? The adverb "tremblingly" offers no relief from this dilemma, since the word could just as easily describe Medusa's faltering vision as our own or the speaker's. At the outset of the poem, then, Shelley entangles and confuses the three most important perspectives in any ekphrastic exercise. Each viewer is implicated in the gazing of his/her counterparts.

33)  The second stanza realizes literally what is always only an implicit threat in the genre of ekphrasis. The "gazer's spirit", however, rather than his actual person is turned into stone. In this sense, "On the Medusa" constitutes a Romantic rewriting of the Perseus myth, an internalization of its precursor's more elemental fears. Medusa's visage paralyzes the "thought", not the body; it converts the observer's rational and critical faculty into an appreciative sensual organ capable of recognizing and affirming "the melodious hue of beauty". It is Medusa's ability to suspend the viewer's reflexive visual impulses that ultimately humanizes and harmonizes the strain. Thus, the terror expressed throughout the poem stems less from the Gothic trappings that occupy the painting's margins, than from the observer's temporary experience of muteness and dislocation; it stems from a sudden uncontrollable empathy with Medusa's voiceless predicament, and constitutes a disruption of the speaker's soul as violent in its way as the Gorgon's own beheading.

34)  In Shelley's "thought no more can trace", we hear an echo of Keats's famous complaint that the Grecian urn "tease[s] us out of thought" (l. 44). [30] Indeed, both poets are worried about the artwork's innate capacity to make us suspend critical reflection, to lure the viewer into a half-conscious liaison he can neither control nor resist. For like the urn's, the Medusa's realm is that of silence, a place where masculinized discourse breaks down. Once she has seduced the observer away from any "irritable reaching after fact or reason", then she can begin working to feminize him.

35)  That some sort of transformation has occurred following the speaker's glance at the Medusa becomes almost immediately apparent in the poem's imagery. Until now rather straightforward and conventional, the poem's language suddenly changes, loosening up and becoming marvelously irrational and inventive. In a flurry of synesthesia, Shelley blends music and vision ("melodious hue of beauty"), sense and sight ("the glare of pain"), and converts an ekphrasis into a symphony ("the strain"). Medusa's act of engraving her lineaments onto the poetic observer creates a type of aesthetic revolution whereby, as William Carlos Williams writes, "the ear and the eye lie / down together in the same bed".[31] This represents an imaginative gesture of what Mitchell calls "ekphrastic hope", a gesture that eliminates the borders between feminine object and male observer as well as between the senses.

36)  Such a gesture enables the poet radically to alter his conception of the image. Whereas in the beginning Medusa had been a monstrous "it", a "dead face", by the end the poet can refer to his subject more nobly, as "A woman's countenance". In the course of the stanzas the observer has been humanized, taught to affirm "Beauty" rather than "Truth". Expecting to encounter the unspeakable and grotesque, the poet is surprised to find that "[i]t is less the horror than the grace" that paralyzes his rational faculty. Shelley's poem is thus a profound confirmation of Hélène Cixous's realization that if men could only look the Medusa straight in the eye they might find her laughing rather than glaring.[32]

37)  A final clue that the speaker's conversion is serious, that his perspective has been permanently altered by his experience occurs in the additional stanza. If it was not clear in the originally printed text, Shelley makes the parallel explicit here: "the ever-shifting mirror" thrown up by the varnished gleams of the serpents' folds and the Medusa's exhalation reflects both the poetic observer, and perhaps more importantly, his poem. The text represents a mirror-image of Medusa in all her painful mutilation, "The fragment of an uncreated creature". Like the breath issuing from that severed head, there is something struggling to be born in Shelley's own verse, something fragmented and "trunkless", yet lovely.[33] With the recognition that beauty may inhere even in the most unlikely (and unsightly) of subjects, the poet opens up the possibility that his own draft, rough-hewn and ragged though it may seem, nevertheless contains the seeds of "life in death". The poem's triumph lies in coming to terms with the terror of its own half-articulated composition, and in valorizing the silence, displacement, and irresolution of its stanzas. The lacunae in the text are finally upheld and celebrated; they come to represent a type of absence that affirms a poetics of aesthetic encounter which remains wholly antithetical to the predatory gazing of the eighteenth century, a benevolent poetics which embraces silence, difference, and the other.

38)  Lying prone in the frightful landscape, then, the Medusa becomes an emblem of the poem itself, as well as a perfect embodiment of an ekphrastic fear that is overcome and turned to advantage. What makes Shelley's poem so very different from its counterparts in the eighteenth century, is that the poetic observer never considers a return to the safety of Lessing's generic borders as a possible option; in other words, he refuses to reestablish or endorse the segregation of the sister arts. Instead of putting Medusa back in the picture, the speaker bears her lineaments finely etched on his soul, and carries them out of the museum and into the world.

39)  "On the Medusa" represents the poet's meeting and reconciliation with the primal forces of his own imagination. In her paradoxical loveliness and horror, her shattered and pathetic splendor, Medusa is the very personification of Shelley's own idiosyncratic poetic. As Marin states: "Deux Méduses donc en une: l'horreur, la monstruosité et l'éclatante beauté; fascination des contraires dans le mélange de l'un et de l'autre."[34] (Thus [there are] two Medusas in one: horror, monstrosity and dazzling beauty; the fascination of opposites in the mixture of the two.) One suspects that what stares back at the poet is shockingly familiar, a reflection of his most private fears about the power of poetic images and their potential for both good and evil, life and death. That the symbol of imagination is curiously androgynous tells us something of Shelley's own psyche (as well as the increasing complexity of the genre). A welter of masculine snakes atop a superbly feminine countenance, the Medusa embodies a potent image of sexual ambivalence.[35]

40)  If her physiognomy does not offer Shelley sufficient grounds for elegy, Medusa's politics certainly do. The implications of her oppression confirm the poet's own belief that the imagination is intimately bound up with political freedom and all "strugglings to the last" against authority. The last couplet, with its confident defiance and culminating march of spondees--"A woman's countenance, with serpent-locks, / Gazing in death on Heaven from those wet rocks"--shows that Shelley's perception of Medusa has come a long way in five brief stanzas. Whereas at first it "lieth", looking towards the "midnight sky" (the redundancy of "supine" in line 2 only reinforces her passivity, her deference, and her deceit), at the end Medusa's grim visage gazes in a more determined challenge towards "Heaven"; initially ambiguous and weak, her gesture at last becomes fierce and confrontational. Hers is the rage of the disinherited, and the poet has learned enough in the course of the poem to describe her predicament with compassion.

41)  Just as the first line in the stanza brings together the terror and the beauty of looking on the Medusa, so the last lines attempt to reintegrate the planes which the picture so frantically rushes to separate--the masculine snakes and the feminine face. The rhymes pair the fertility of the "wet rocks", (earlier associated with grass) with the phallic serpents, and effectively fuse the competing elements in Medusa's character. The rift in her sexuality healed, the Gorgon becomes less a dangerous and uncontrollable figure of subversion, than a radical with a legitimate cause. In the end, Shelley makes Medusa's defiance his own.

Notes

1. George Saintsbury: A History of Criticism and Literary Taste in Europe vol. I. (London: Blackwood,1902), 491. Shelley's poetry is quoted from: Neville Rogers (ed.): Selected Poetry of Percy Bysshe Shelley (London: Oxford UP, 1968). back

2. Leo Spitzer: "The 'Ode on a Grecian Urn,' or Content vs. Metagrammar." In: Essays on English and American Literature, ed. Anna Hatcher (Princeton: Princeton UP, 1962), 72. back

3. Jean H. Hagstrum: The Sister Arts (Chicago: U of Chicago P, 1958), 18n. Definitions of ekphrasis vary according to the particular argument to be deployed; hence, the universality of Saintsbury's version naturally differs from Wendy Steiner's more philosophical and abstract "the concentration of action in a single moment of energy", (The Colors of Rhetoric [Chicago: U of Chicago P, 1982], 41) or Richard Lanham's aptly generic and technical 'A self-contained description, often on a commonplace subject, which can be inserted at a fitting place in a discourse" (A Handlist of Rhetorical Terms [Berkeley and Los Angeles: U of California P, 1968], 39). Intent on demonstrating that literature uses the plastic arts as a formal model to create "stillness", or in his term, a "still movement" in its own temporal medium, Murray Krieger takes the mimetic path ("Ekphrasis and the Still Movement of Poetry; or, Laokoon Revisited." In: The Poet as Critic, ed. Frederick McDowell [Evanston: Northwestern UP, 1967]), while C. S. Baldwin. interested more in oratory and rhetoric, defines ekphrasis as "a form of Alexandrianism [...] [which] perverts description because it frustrates narrative movement" (Medieval Rhetoric and Poetic [New York: Macmillan,1928], 29). The elegant simplicity of John Bender's definition comes the closest to our purposes: "literary descriptions of real or imagined works of visual art" (Spenser and Literary Pictorialism [Princeton: Princeton UP, 1972], 51n). back

4. For a list of ekphrastic poems as well as a valuable source of works on Hellenism, see Timothy Webb (ed.): English Romantic Hellenism 1700-1824 (New York: Barnes and Noble, 1982). back

5. Henry Hart Milman: "The Belvidere Apollo." In: Annals of the Fine Arts (London: Hurst, Robinson and Co., 1820). These poems may be found in Alexander Chalmers's popular collection, The Works of the English Poets in twenty-one volumes (London, 1810). See also Christopher Smart's "On a Picture of Miss R-G-N.". back

6. See J. H. Parker, J. Vincent and H. Slatter (eds.) Oxford Prize Poems eighth edition (London: 1834); for a discussion of these poems, see Stephen A. Larrabee: English Bards and Grecian Marbles (New York: Columbia UP, 1943). back

7. Chalmers, XI, 114. back

8. W. J. T. Mitchell: "Space and Time: Lessing's Laocoon and the Politics of Genre." In: Iconology (Chicago: U of Chicago P, 1985), 113. back

9. On the curious potency of images, and the strategies we use for controlling them, see Jeane Laude: "On the Analysis of Poems and Paintings." In: New Literary History III (Spring, 1972): 471-86. Of particular relevance to Holland's poem is the following passage: "Whether figurative or 'abstract', images assert themselves by the immediacy of their perception [...] This is especially true of images used in advertising and in television; once they have been registered, we consider them forgotten. It is by the illusion of forgetfulness that images, however furtive or accidental they may be, work upon the minds of their viewers. Whatever activity takes place is achieved by the image itself and not by its viewer. One may avoid being directed by the image by reading it. If we have at our disposal some method by which to decipher, appraise, refuse, subvert, or accept pictures knowingly through a demythifying and demystifying process, the kind of activity undertaken to accomplish this is not of the same character as that activity which strives for the comprehension of a text" (477). back

10. Thomas De Quincey's description of Piranesi's Carceri  in Confessions of an English Opium Eater uses frames even more obtrusively than Holland. In an ekphrasis which itself describes a series of framed images, the author contrives no less than three levels of remove (actually, four, if we include the fact that the ekphrasis is used solely as a metaphor for De Quincey's own dream experiences); I quote only from the preface to his account: "Many years ago, when I was looking over Piranesi's Antiquities of Rome, Coleridge, then standing by, described to me a set of plates from that artist, called his Dreams, and which record the scenery of his own visions during the delirium of a fever. Some of these (I describe only from memory of Coleridge's account) represented vast Gothic halls [...]." Confessions of an English Opium-Eater (London: J. M. Dent, 1907), 237. back

11. For a popular counterpart to Lessing's Laokoon, see the anonymous essay entitled "Coalition between Poetry and Painting", which appeared in the Universal Magazine for May of 1797. back

12. For the political implications of Shelley's poem, see Neil Hertz "Medusa's Head: Male Hysteria under Political Pressure." In Representations IV (Fall, 1983); Jerome J. McGann: "The Beauty of Medusa: A Study in Romantic Literary Iconology." In: Studies in Romanticism XI (1972): 3-25; W. J. T. Mitchell: "Ekphrasis and the Other." In: South Atlantic Quarterly XCI (Summer 1992) 695-719. [NOTE: In links will take you to electronic versions of these texts linked to the Romantic Circles Electronic Edition of Shelley's "Medusa."] back

13. Louis Marin: Détruire la Peinture (Paris: Galilee, 1977), 139. back

14. Robert Graves: The Greek Myths (Harmondsworth: Penguin, 1955), 244. back

15. Marin, 134. back

16. Marin, 243. back

17. The refusal to describe Medusa's face represents the central failing of ekphrasis itself, the resigned return to "eckphrastic indifference", or a belief in the impossibility of any direct depiction of the sculpture or picture in a poem. Tobin Siebers has astutely noted "The remarkable ability of the head of Medusa to represent what cannot be represented or what should not be represented" (The Mirror of Medusa [Berkeley: U of California P, 1983], 8). back

18. See Sigmund Freud, "Medusa's Head." (1922) In: Collected Papers, V (London: Hogarth, 1953), 105f. Philip E. Slater offers a provocative reappraisal of Freud's conclusions, arguing that the process of turning to stone may not represent a symbolic erection so much as impotence and immobility: "The enstonement is not a compensatory comfort but a feared outcome. While Freud focuses on stiffness, we can just as well stress the numb or anesthetic aspect of turning to stone" (The Glory of Hera [Boston: Beacon Press, 1968], 321). back

19. Slater's analysis is again relevant here: "Despite all the help and gifts Perseus receives, he is 'barely man enough' to kill the Medusa. One is reminded of Pelops, who required not only winged horses and chariot but also a bribed charioteer to overcome a paternal opponent. But the Perseus legend is far more extreme, with two gods aiding him, a superfluity of magic implements, a sleeping enemy at whom he cannot even look, and a goddess guiding his hand to actually commit the act. Has there ever been so helpless and dependent a hero?" (326). back

20. Mitchell, 21. back

21. Mitchell, 23. back

22. Carol Jacobs: "On Looking at Shelley's Medusa." In: Yale French Studies LXIX (1985): 163-179; here: 167. back

23. As Jacobs points out, this process can also be seen in the natural landscape: in stanza three "the rock becomes watery while, shortly thereafter, the air becomes solid" (172n). back

24. Neville Rogers: "Shelley and the Visual Arts." In: Keats-Shelley Memorial Bulletin XII (1961): 9-17; here: 17. back

25. Jacobs observes that there is a concomitant lapse in grammar as well. She argues quite rightly that the word "trace" in stanza two, usually a transitive verb, is here left without an object: "Thought is no longer the trace of something else; [...] what the text insists on is a lack in thought and this lack is coincident with a grammatical failure, with the inability of language to complete its meaning" (170). back

26. William Shakespeare: The Complete Works: Compact Edition, ed. Stanley Wells et al. (Oxford: Clarendon, 1988), 251. back

27. Shakespeare, 253. back

28. Shakespeare, 251. back

29. Siebers, 4. back

30. John Keats: John Keats (The Oxford Authors), ed. Elizabeth Cook (Oxford: Oxford UP, 1990), 289. back

31. William Carlos Williams: "Song." In: Pictures from Brueghel (New York: New Directions, 1962). back

32. Hélène Cixous: "The Laugh of the Medusa." In: New French Feminism, ed. Elaine Marks and Isabelle de Courtivron (New York: Schocken, 1981), 255. back

33. I am grateful to Keith Hanley, who pointed out the interesting parallels between this poem and "Ozymandias". Both involve fragmented or "trunkless" figures and "shattered visages"; further, both seem to be critiques of masculine authority and dominion--Medusa, after all, is a victim of Neptune's ravishment. Whereas the landscape in "Ozymandias" reflects the barren and desiccated prophecy of tyranny however, that of "On the Medusa" seems to figure forth the Medusa's fertility--even after her dismemberment. The glistening snakes, slimy toads, bats, and rodents all thrive in a dark and vegetal landscape of wet rocks and thick grass. Medusa's decapitation leads to a form of rebirth, while the dislocation in Ozymandias augurs a panorama of "lone and level sands". back

34. Marin, 135. back

35. This combination is curiously reminiscent of another portrait Shelley admired, and one which he described in similar language: "There is a fixed and pale composure upon the features: she seems sad and stricken down in spirit [...] Her head is bound with folds of white drapery from which the yellow strings of her golden hair escape, and fall about her neck" (Preface to The Cenci). The Reni portrait of Beatrice Cenci is not only remarkable for echoing the shading and iconography of the Medusa, but also, as Richard Holmes astutely points out, for bearing a striking resemblance to the famous Curran portrait of Shelley. Holmes's subsequent comments on this score are germane, particularly if we keep the Medusa in mind as an additional foil: "Both are oddly androgynous creations, whose glance is a mixture of defiance and pathos; the broad, pale forehead; the delicately arched brows; the large almond-shaped eyes; and finally the long faintly aquiline nose--these upper features are so markedly similar that it seems almost certain that Aemilia Curran was influenced by the Reni [...]. The identity between male and female, or at least the transposable or interchangeable elements between the two sexes had long been an undertheme of Shelley's writing. The masculine role frequently assigned to his heroines like Cythna or Asia, and the passivity of his male lovers like the poet of Alastor, is one of the constant and original features of his poetry." (Shelley: The Pursuit [New York: Dutton, 1975], 517). back